Opportunities in the Failing Heart Johannes Michael Ing Han Gho
Opportunities in the Failing Heart
Johannes Michael Ing Han Gho
Opportunities in the Failing Heart Š 2015 JMIH Gho, Utrecht Layout:
wenz iD.nl / Wendy Schoneveld
Cover design:
Robert van Sluis / eyefordetail.nl
Printed by:
Gildeprint - Enschede
Financial support by the Dutch Heart Foundation and the Heart & Lung Foundation Utrecht for the publication of this thesis is gratefully acknowledged. This research forms part of the Project P1.04 SMARTCARE of the research program of the BioMedical Materials Institute, co-funded by the Dutch Ministry of Economic Affairs, Agriculture and Innovation. Publication of this thesis was financially supported by Stichting Genetische Hartspierziekte PLN (http://stichtingpln.nl, Middenmeer, the Netherlands). Further financial support for the publication of this thesis by Stichting Cardiovasculaire Biologie and ChipSoft B.V. is gratefully acknowledged. ISBN 978-94-6233-100-6
Opportunities in the Failing Heart Mogelijkheden in het falende hart (met een samenvatting in het Nederlands)
Proefschrift
ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magnificus, prof.dr. G.J. van der Zwaan, ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op donderdag 29 oktober 2015 des middags te 12.45 uur
door
Johannes Michael Ing Han Gho geboren op 29 september 1986 te Leusden
Promotoren:
Prof.dr. P.A.F.M. Doevendans
Prof.dr. F.W. Asselbergs
Copromotoren: Dr. S.A.J. Chamuleau
Dr. A. Vink
TABLE OF CONTENTS
Chapter 1
PART ONE
Introduction
|
Chapter 2
9
Clinical Course Portrayed Heart Failure Following Myocardial Infarction: a Cohort Study of Incidence
19
and Prognostic Factors in 24 745 Patients Using Linked Electronic Records Chapter 3
Heart Failure Following ST-elevation Myocardial Infarction: an AGNES
43
Cohort Study of Incidence and Prognostic Factors
PART TWO Chapter 4
|
Finding Fibrosis Patterns A Systematic Comparison of Cardiovascular Magnetic Resonance and
55
High Resolution Histological Fibrosis Quantification in a Chronic Porcine Infarct Model Chapter 5
Endogenous Contrast MRI of Cardiac Fibrosis: Beyond Late Gadolinium
73
Enhancement Chapter 6
High Resolution Systematic Digital Histological Quantification of Cardiac Fibrosis and Adipose Tissue in Phospholamban p.Arg14del Mutation
91
Associated Cardiomyopathy Chapter 7
The Distribution Pattern of Fibrosis in Genetic Cardiomyopathy is Related to the Type of Pathogenic Mutation
109
PART THREE Chapter 8
|
Elucidating (Epi)genetic and Translating Therapeutic Pathways Chromatin Regulation in Phospholamban R14del Mutation Associated
127
Cardiomyopathy Chapter 9
Cell Therapy, a Novel Remedy for Dilated Cardiomyopathy?
175
A Systematic Review Chapter 10
Xenotransplantation of Human Cardiomyocyte Progenitor Cells Does Not
199
Improve Cardiac Function in a Porcine Model of Chronic Ischemic Heart Failure Chapter 11
General Discussion
221
Samenvatting in het Nederlands
232
List of publications
236
Acknowledgements / Dankwoord
238
Curriculum vitae
242
Chapter
1
Introduction
CHAPTER 1
INTRODUCTION Heart failure (HF) is a syndrome with typical signs and symptoms that can result from abnormal cardiac structure or function.1 It can lead to impaired quality of life, decreased functional capacity, hospital admissions and mortality. The historical terminology used to describe HF has been based on measurement of left ventricular ejection fraction (EF). Currently the diagnosis of HF has been divided in HF with reduced (HFrEF) or preserved EF (HFpEF). Causes of HF in the western world include hypertension, valvular heart disease and (non-ischaemic) cardiomyopathies and the leading cause is ischaemic heart disease (Figure 1).2, 3 Epidemiology of heart failure Heart failure incidence increases with age4 and in a previous Dutch community based middleaged cohort study 4.4% of the individuals were diagnosed with HF during a median follow-up of 11.5 years of which 66% with HFrEF and 34% with HFpEF.5 Research studying the incidence of HF following myocardial infarction (MI) is scarce and mainly stems from the thrombolytic era. There is conflicting evidence whether the long-term incidence of HF following MI is increasing or decreasing. An analysis of the Framingham Heart Study regarding long-term trends following MI demonstrated an increase in incident HF between the 1970s and 1990s, possibly due to improved survival after MI.6 The 5-year incidence of HF following MI rose from 27.6% (1970 to 1979) to 31.9% (1990 to 1999). In a different study using the Swedish hospital discharge register with MI patients from 1993-2004, the 3-year risk of HF remained high despite a decrease observed in hospitalisations for HF.7 In MI patients 65 years or older, they found a 3-year hospital HF diagnosis of 31.5% (1993 to 1995) and 28.0% (2002 to 2004). Portraying the incidence of and prognostic factors associated with HF following MI in the current era is important for future observational studies, clinical trials, measuring health outcomes and guideline development.
Figure 1. Causes of heart failure. Figure created with data derived from 3.
10
INTRODUCTION
Pathophysiology Heart failure has been associated with myocardial fibrosis and arrhythmias and different
1
pathophysiological processes can result in focal and diffuse myocardial fibrosis.8 After myocardial injury, maladaptive changes can lead to pathologic remodelling of the left ventricle with dilatation, impaired contractility and relaxation.9 In patients with left ventricular systolic dysfunction, systemic responses such as an elevated sympathetic tone and renin-angiotensin-aldosterone system can have detrimental effects leading to the clinical syndrome of HF. In a previous study using the Framingham Offspring Study cohort, approximately 18% of the risk of HF was attributable to parental HF.10 An other important cause of HF are non-ischaemic cardiomyopathies which are mainly inherited and during the last decades developments in sequencing allowed that an increasing amount of genetic causes are being unravelled.11 Dilated cardiomyopathy (DCM) is characterised by left ventricular dilation associated with cardiac dysfunction and ventricular arrhythmias.12 Causative genes in DCM predominantly encode cytoskeletal and sarcomeric proteins, but disturbance of calcium cycling also seems to play an important role (Table 1).13, 14 For the purpose of this thesis, we will discuss cardiac calcium homeostasis in more detail. Calcium cycling in human cardiomyocytes is mainly regulated by sarcoplasmic reticulum (SR) Ca2+ ATPase (SERCA2a), a Ca2+ pump which can replenish Ca2+ stores in the SR (Figure 2).15 Lowering of the Ca2+ concentration in the cytosol by SERCA2a causes relaxation of the cardiomyocyte and influences cardiac contractility by the availability of Ca2+ stores for the next beat. The decreased SERCA2a expression in HF has been associated with impaired cardiac contractility and relaxation.16 Phospholamban (PLN) is a small transmembrane protein in the SR, a reversible inhibitor of SERCA2a and in its dephosphorylated state PLN inhibits SERCA2a activity.15 Phospholamban has been derived from phosphate and the Greek word λαμβανειν, which means to receive or to seize.17 Upon phosphorylation, PLN dissociates from SERCA2a, thereby relieving Ca2+ pump inhibition and enhancing relaxation and contractility. If PLN becomes chronically inhibitory, it can lead to diminished contractility and several mutations in PLN have been associated with DCM including the R14del (p.Arg14del) mutation.18 Within the Netherlands a relatively large population of patients exists with the PLN R14del mutation causing HF and it is likely that this mutation arose around 575–825 years ago.19 Estimated 1 in 1400 people is carrier of the R14del mutation in the North of the Netherlands and the total number of Dutch R14del carriers is likely to be more than 2000. The R14del
Table 1. Genes associated with familial dilated cardiomyopathy. Adapted from 14. Mutation group
Gene(s)
Cytoskeletal
DES, DMD, ILK, LAMA4, LDB3, PDLIM3, SGCD, VCL
Sarcomeric
ACTC1, ACTN2, ANKRD1, CSRP3, MYBPC3, MYH6, MYH7, MYPN, TCAP, TNNC1, TNNI3, TNNT2, TPM1, TTN
Calcium cycling
PLN
Nuclear envelope
LMNA, TMPO
Desmosomal
DSC2, DSG2, DSP
Other
PSEN1, PSEN2, ABCC9, SCN5A, TAZ (G4.5), BAG3, CRYAB, EYA4
11
CHAPTER 1
Figure 2. Calcium cycling in cardiomyocytes. In response to depolarization due to sodium (Na+) influx, calcium (Ca2+) enters the cytosol through the L-type calcium channels (LTCC) in the plasma membrane. This Ca2+ influx leads to calcium-induced calcium release from the sarcoplasmic reticulum mediated by the Ca2+ release channels (ryanodine receptors; RyRs). Calcium binds to troponin in the thin filaments of myofibrils to activate muscle contraction and Ca2+ is removed from the cytosol by sarcoplasmic reticulum Ca2+ ATPase (SERCA2a) and the sodium-calcium exchanger (NCX) on the plasma membrane. Phospholamban (PLN) is a reversible inhibitor of SERCA2a and in its dephosphorylated state PLN inhibits SERCA2a activity. Upon phosphorylation by protein kinase A (PKA), through the β-adrenergic receptor pathway, or Ca2+/calmodulindependent protein kinase (CaMKII), PLN dissociates from SERCA2a, thereby relieving Ca2+ pump inhibition and enhancing cardiac relaxation and contractility. Phospholamban is dephosphorylated by protein phosphatase (PP1), which ends the stimulation phase and PP1 is regulated by inhibitor-1 (I-1).
mutation has been associated with DCM and arrhythmogenic cardiomyopathy. Insights regarding the distribution pattern of fibrosis and fatty replacement in R14del cardiomyopathy patients are limited. Furthermore, several other non-ischaemic cardiomyopathies have been associated with causal mutations and fibrosis is frequently observed in this group. Revealing distribution patterns of cardiac fibrosis in cardiomyopathies is important to elucidate mechanisms of pathophysiology and for cardiac imaging purposes.
12
INTRODUCTION
Since the discovery of deoxyribonucleic acid (DNA)20 and introduction of methods for DNA sequencing21 novel methods have emerged for high-throughput next-generation sequencing
1
(NGS).22 State of the art genetic research has the potential to lead to a more fundamental understanding of the underlying pathophysiology related to HF and to identify new therapeutic targets and diagnostic biomarkers. Despite these advances limited genetic loci associated with risk of incident HF have been identified. Epigenetics is defined as the study of heritable changes in gene expression which are not due changes in DNA sequence.23 Epigenetic modifications can make regulatory elements of genes more or less permissive to transcription factors, thereby changing gene expression. Recently, NGS technologies have been developed to profile the ‘regulome’.24 The regulome is a set of DNA transcriptional regulatory elements, e.g., promotors and enhancers. Identification of differentially regulated gene regions in HF could lead to improved understanding of disease mechanisms, risk stratification and targeted therapies. Diagnosis While clinical assessment is the keystone of patient management, clinicians often utilize additional laboratory tests or imaging studies to aid differential diagnosis.25 Cardiac function can be assessed using different imaging modalities, e.g., echocardiography or cardiovascular magnetic resonance imaging (CMR). Precise assessment of cardiac fibrosis is important for diagnosis, predicting prognosis and to guide and monitor therapy.26 The non-invasive reference standard to assess cardiac anatomy and function is CMR26 and local cardiac function can be assessed for example using wall thickening27 or strain analyses28. Fibrosis detection using late gadolinium enhancement (LGE) CMR is possible due to prolonged retention of contrast agent in regions of myocardial fibrosis compared to healthy myocardium, resulting in an increased signal intensity in fibrotic regions using T1 weighted CMR.29 While LGE CMR provides an accurate qualitative measure of replacement fibrosis, it has several disadvantages. Use of gadolinium has potential adverse effects and does not provide a quantitative or direct measurement of cardiac collagen. The LGE method is restricted in assessment of diffuse fibrosis, results differ between different imaging studies and it is less suitable for longitudinal studies by factors affecting reproducibility.26 Possible adverse effects include rare but potentially life-threatening anaphylaxis, contrast induced nephropathy and nephrogenic systemic fibrosis.30 Currently, there is no consensus on the preferred method for LGE quantification although this would be useful for interpreting diagnostic studies, determining prognostic value and measuring therapeutic outcomes. Therefore it would be of interest to compare CMR parameters with histological fibrosis. Recently, several novel CMR techniques have been proposed for fibrosis detection. Contrastenhanced T1-mapping can generate a map representing the T1 relaxation times by directly quantifying T1 values per voxel.31 This method provides quantitative assessment of extra cellular volume (ECV) and could be an improvement over LGE, especially in patients with diffuse fibrosis. Direct detection of myocardial fibrosis might be possible when the acquisition is tuned to the changes in collagen content, could obviate the need for use of contrast agents, reduce patient burden and health costs. Endogenous contrast methods to assess fibrosis with CMR include non-contrast T132 and T1rho mapping33.
13
CHAPTER 1
Advanced therapy for HF: cardiac regeneration One of the most important modalities to regenerate the failing heart is heart transplantation.34 Unfortunately, the availability of donor organs is a major challenge and heart transplantation comes at a price with accompanying morbidity and mortality.35 Several other therapies have been of interest to induce cardiac regeneration, including cell therapy.36 In theory, cell therapy can be used to repair the failing heart and proposed working mechanisms include formation of new muscle cells and blood vessels.37 Given the lack of evidence regarding transdifferentiation of cells to cardiomyocytes led to the hypothesis that paracrine mechanisms contribute to improvement in cardiac function.38 While stem cell therapy has been extensively studied in preclinical39 and clinical40 studies of ischaemic HF, evidence in the setting of non-ischaemic DCM with myocardial dysfunction seems scant. Thus far, multiple cell types have been studied for cardiac regeneration, but the optimal cardiac regenerative therapy has yet to be discovered. Our research group previously reported the successful isolation of human cardiomyocyte progenitor cells.41 These cells could form functional cardiomyocytes and while this might be a promising regenerative therapy for HF patients, further research is needed before translation to the clinic. Outline of this thesis To overcome the challenges in HF the intent of this thesis is to provide opportunities from bench-to-bedside to elucidate pathways associated within the failing heart with the ultimate aim to improve clinical outcome for patients. Chapter 2 and 3 investigate the incidence of and prognostic factors associated with HF following MI in an observational cohort study, using linked electronic health record (EHR) sources from primary and secondary care in England (Chapter 2) and using patients with follow-up for incident HF after a first STEMI in the Arrhythmia Genetics in the NEtherlandS (AGNES) study (Chapter 3). In Chapter 4, we systematically compare in vivo CMR techniques with myocardial fibrosis assessment. The CMR parameters of interest are LGE CMR, myocardial strain (derived from feature tracking) and wall thickening. In Chapter 5, an overview is presented of novel methods for cardiac fibrosis detection on CMR beyond LGE, in particular methods with endogenous contrast. A novel high resolution systematic digital histological quantification method is proposed in Chapter 6 where patterns of cardiac fibrosis and adipose tissue in PLN R14del cardiomyopathy hearts are studied. In Chapter 7, we use the method previously described in Chapter 6 to determine the distribution pattern of fibrosis in multiple genetic non-ischaemic cardiomyopathies. In Chapter 8, we compare the activity of transcriptional regulatory elements between hearts of patients with the PLN R14del mutation and controls. Finally, Chapter 9 and 10 depict studies related to the translation of cell therapy towards the clinic. Chapter 9 describes a systematic literature review on cell therapy in preclinical and clinical studies for non-ischaemic DCM. In Chapter 10, we study the effects of human cardiomyocyte progenitor cells in a large animal model of chronic ischaemic HF. In summary, we tried to unravel the basic mechanisms leading to HF, but also the pathways relevant for cardiac regeneration.
14
INTRODUCTION
REFERENCES 1.
2. 3.
4.
5.
6. 7.
8. 9. 10. 11. 12. 13.
14. 15. 16.
17. 18.
19.
20. 21. 22. 23. 24.
1
McMurray JJ, Adamopoulos S, Anker SD, Auricchio A, Bohm M, Dickstein K, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J. 2012;33:1787-847. Mosterd A, Hoes AW. Clinical epidemiology of heart failure. Heart. 2007;93:1137-46. Baldasseroni S, Opasich C, Gorini M, Lucci D, Marchionni N, Marini M, et al. Left bundle-branch block is associated with increased 1-year sudden and total mortality rate in 5517 outpatients with congestive heart failure: a report from the Italian network on congestive heart failure. Am Heart J. 2002;143:398-405. Bleumink GS, Knetsch AM, Sturkenboom MC, Straus SM, Hofman A, Deckers JW, et al. Quantifying the heart failure epidemic: prevalence, incidence rate, lifetime risk and prognosis of heart failure The Rotterdam Study. Eur Heart J. 2004;25:1614-9. Brouwers FP, de Boer RA, van der Harst P, Voors AA, Gansevoort RT, Bakker SJ, et al. Incidence and epidemiology of new onset heart failure with preserved vs. reduced ejection fraction in a community-based cohort: 11-year follow-up of PREVEND. Eur Heart J. 2013;34:1424-31. Velagaleti RS, Pencina MJ, Murabito JM, Wang TJ, Parikh NI, D’Agostino RB, et al. Long-term trends in the incidence of heart failure after myocardial infarction. Circulation. 2008;118:2057-62. Shafazand M, Rosengren A, Lappas G, Swedberg K, Schaufelberger M. Decreasing trends in the incidence of heart failure after acute myocardial infarction from 1993-2004: a study of 175,216 patients with a first acute myocardial infarction in Sweden. Eur J Heart Fail. 2011;13:135-41. Weber KT, Sun Y, Bhattacharya SK, Ahokas RA, Gerling IC. Myofibroblast-mediated mechanisms of pathological remodelling of the heart. Nat Rev Cardiol. 2013;10:15-26. McMurray JJ. Clinical practice. Systolic heart failure. N Engl J Med. 2010;362:228-38. Lee DS, Pencina MJ, Benjamin EJ, Wang TJ, Levy D, O’Donnell CJ, et al. Association of parental heart failure with risk of heart failure in offspring. N Engl J Med. 2006;355:138-47. Towbin JA, Bowles NE. The failing heart. Nature. 2002;415:227-33. Jefferies JL, Towbin JA. Dilated cardiomyopathy. Lancet. 2010;375:752-62. Harakalova M, Kummeling G, Sammani A, Linschoten M, Baas AF, van der Smagt J, et al. A systematic analysis of genetic dilated cardiomyopathy reveals numerous ubiquitously expressed and muscle-specific genes. Eur J Heart Fail. 2015;17:484-93. Hershberger RE, Hedges DJ, Morales A. Dilated cardiomyopathy: the complexity of a diverse genetic architecture. Nat Rev Cardiol. 2013;10:531-47. MacLennan DH, Kranias EG. Phospholamban: a crucial regulator of cardiac contractility. Nat Rev Mol Cell Biol. 2003;4:566-77. Hasenfuss G, Reinecke H, Studer R, Meyer M, Pieske B, Holtz J, et al. Relation between myocardial function and expression of sarcoplasmic reticulum Ca(2+)-ATPase in failing and nonfailing human myocardium. Circ Res. 1994;75:434-42. Katz AM. Discovery of phospholamban. A personal history. Ann N Y Acad Sci. 1998;853:9-19. Haghighi K, Kolokathis F, Gramolini AO, Waggoner JR, Pater L, Lynch RA, et al. A mutation in the human phospholamban gene, deleting arginine 14, results in lethal, hereditary cardiomyopathy. Proc Natl Acad Sci U S A. 2006;103:1388-93. van der Zwaag PA, van Rijsingen IA, de Ruiter R, Nannenberg EA, Groeneweg JA, Post JG, et al. Recurrent and founder mutations in the Netherlands-Phospholamban p.Arg14del mutation causes arrhythmogenic cardiomyopathy. Neth Heart J. 2013;21:286-93. Watson JD, Crick FH. Molecular structure of nucleic acids; a structure for deoxyribose nucleic acid. Nature. 1953;171:737-8. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A. 1977;74:5463-7. Metzker ML. Sequencing technologies - the next generation. Nat Rev Genet. 2010;11:31-46. Eccleston A, DeWitt N, Gunter C, Marte B, Nath D. Epigenetics. Nature. 2007;447:395-. Blow MJ, McCulley DJ, Li Z, Zhang T, Akiyama JA, Holt A, et al. ChIP-Seq identification of weakly conserved heart enhancers. Nat Genet. 2010;42:806-10.
15
CHAPTER 1
25. Vasan RS. Biomarkers of cardiovascular disease: molecular basis and practical considerations. Circulation. 2006;113:2335-62. 26. Mewton N, Liu CY, Croisille P, Bluemke D, Lima JA. Assessment of myocardial fibrosis with cardiovascular magnetic resonance. J Am Coll Cardiol. 2011;57:891-903. 27. Holman ER, Vliegen HW, van der Geest RJ, Reiber JH, van Dijkman PR, van der Laarse A, et al. Quantitative analysis of regional left ventricular function after myocardial infarction in the pig assessed with cine magnetic resonance imaging. Magn Reson Med. 1995;34:161-9. 28. Hor KN, Gottliebson WM, Carson C, Wash E, Cnota J, Fleck R, et al. Comparison of magnetic resonance feature tracking for strain calculation with harmonic phase imaging analysis. JACC Cardiovasc Imaging. 2010;3:144-51. 29. Kim RJ, Fieno DS, Parrish TB, Harris K, Chen EL, Simonetti O, et al. Relationship of MRI delayed contrast enhancement to irreversible injury, infarct age, and contractile function. Circulation. 1999;100:1992-2002. 30. Bellin MF, Van Der Molen AJ. Extracellular gadolinium-based contrast media: an overview. Eur J Radiol. 2008;66:160-7. 31. Moon JC, Messroghli DR, Kellman P, Piechnik SK, Robson MD, Ugander M, et al. Myocardial T1 mapping and extracellular volume quantification: a Society for Cardiovascular Magnetic Resonance (SCMR) and CMR Working Group of the European Society of Cardiology consensus statement. J Cardiovasc Magn Reson. 2013;15:92. 32. Bull S, White SK, Piechnik SK, Flett AS, Ferreira VM, Loudon M, et al. Human non-contrast T1 values and correlation with histology in diffuse fibrosis. Heart. 2013;99:932-7. 33. Witschey WR, Zsido GA, Koomalsingh K, Kondo N, Minakawa M, Shuto T, et al. In vivo chronic myocardial infarction characterization by spin locked cardiovascular magnetic resonance. J Cardiovasc Magn Reson. 2012;14:37. 34. Chien KR. Regenerative medicine and human models of human disease. Nature. 2008;453:302-5. 35. Yacoub M. Heart transplantation: the end of the beginning. Am J Transplant. 2008;8:1767-8. 36. Segers VF, Lee RT. Stem-cell therapy for cardiac disease. Nature. 2008;451:937-42. 37. Bartunek J, Dimmeler S, Drexler H, Fernandez-Aviles F, Galinanes M, Janssens S, et al. The consensus of the task force of the European Society of Cardiology concerning the clinical investigation of the use of autologous adult stem cells for repair of the heart. Eur Heart J. 2006;27:1338-40. 38. Maxeiner H, Krehbiehl N, Muller A, Woitasky N, Akinturk H, Muller M, et al. New insights into paracrine mechanisms of human cardiac progenitor cells. Eur J Heart Fail. 2010;12:730-7. 39. van der Spoel TI, Jansen of Lorkeers SJ, Agostoni P, van Belle E, Gyongyosi M, Sluijter JP, et al. Human relevance of pre-clinical studies in stem cell therapy: systematic review and metaanalysis of large animal models of ischaemic heart disease. Cardiovasc Res. 2011;91:649-58. 40. Jeevanantham V, Butler M, Saad A, Abdel-Latif A, Zuba-Surma EK, Dawn B. Adult bone marrow cell therapy improves survival and induces long-term improvement in cardiac parameters: a systematic review and meta-analysis. Circulation. 2012;126:551-68. 41. Goumans MJ, de Boer TP, Smits AM, van Laake LW, van Vliet P, Metz CH, et al. TGF-beta1 induces efficient differentiation of human cardiomyocyte progenitor cells into functional cardiomyocytes in vitro. Stem Cell Res. 2007;1:138-49.
16
17
PART ONE
|
Clinical Course Portrayed
Chapter
2
Heart Failure Following Myocardial Infarction: a Cohort Study of Incidence and Prognostic Factors in 24 745 Patients Using Linked Electronic Records In preparation
Johannes M.I.H. Gho1,2, Amand F. Schmidt1, Laura Pasea1, Stefan Koudstaal1,2, Mar Pujades-Rodriguez1, Spiros Denaxas1, Anoop D. Shah1, Riyaz S. Patel1,3, Chris P. Gale4, Arno W. Hoes5, John G. Cleland6, Harry Hemingway1, Folkert W. Asselbergs1,2,3,7
1
Farr Institute of Health Informatics Research, UCL Institute of Health Informatics,
2
Department of Cardiology, Division Heart and Lungs, University Medical Center Utrecht,
3
Institute of Cardiovascular Science, University College London, London, UK
4
University of Leeds, Leeds, UK
5
Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht,
6
Imperial College London, London, UK
7
Durrer Center for Cardiogenetic Research, ICIN-Netherlands Heart Institute, Utrecht,
University College London, London, UK Utrecht, the Netherlands
Utrecht, the Netherlands
the Netherlands
PART ONE CHAPTER 2
ABSTRACT Aims To investigate the incidence of and prognostic factors associated with heart failure (HF) after myocardial infarction (MI). Methods We used primary care data from England, linked to national datasets providing information on hospital admissions, an MI registry and mortality as part of the CALIBER (CArdiovascular research using LInked Bespoke studies and Electronic health Records) programme. All patients aged 18 years or older who experienced a first MI between 1 January 1998 and 25 March 2010 were included. Heart failure was defined as a HF diagnosis in any of the CALIBER sources and used as outcome in Cox proportional hazard models. Results Of 24,745 patients with a first MI and without a prior history of HF (median follow-up of 3.7 years), 6005 (24.3%) patients developed HF, which represents a crude incidence rate of HF following a first MI of 66.1 cases (95% CI: 64.4 – 67.8) per 1000 person-years. Independent baseline factors associated with subsequent HF were: age [HR per 10 years increase: 1.45 (95%CI 1.41 – 1.49)], higher socioeconomic deprivation (5th [HR 1.27 (95%CI 1.13 – 1.42)] vs. 1st quintile), a history of hypertension [HR 1.16 (95%CI 1.09 – 1.23)], diabetes [HR 1.44 (95%CI 1.34 – 1.55)], atrial fibrillation [HR 1.63 (95%CI 1.51 – 1.75)], peripheral arterial disease [HR 1.38 (95%CI 1.26 – 1.51)], COPD [HR 1.28 (95%CI 1.17 – 1.40)] and a STEMI at presentation [HR 1.21 (95%CI 1.11 – 1.32)]. Conclusion One in 4 people developed HF within 4 years of experiencing a first MI. Key clinical characteristics readily identified patients at risk of HF following MI which may, therefore, be used to guide interventions to reduce cardiovascular disease burden.
20
HEART FAILURE FOLLOWING MI: A CALIBER STUDY
INTRODUCTION In the western world heart failure (HF) is a major medical problem, carrying a high morbidity and mortality risk and is in most cases secondary to ischaemic heart disease.1 Research studying the incidence of HF following myocardial infarction (MI) is limited, mainly originating from the
2
thrombolytic era. Torabi et al. described the natural history and prognosis of HF after MI in 1998 among 1000 British patients to show that 20.3% developed HF during their index MI admission and 33% after discharge preceding death.2 A Danish nationwide cohort study reported a temporal decrease in 90-day HF incidence after MI, from 23.6% in 1997-98 to 19.6% in 2009-10.3 A more recent Swedish study found a decrease in HF incidence from 46% in 1996-1997 to 28% in 2008 during hospitalization for index MI.4 The availability of prospectively collected electronic health records (EHRs) offers a unique opportunity to conduct clinical research.5 Moreover, in England linked EHRs provide a more accurate estimation of incident cardiovascular events.6 Prospective collected data of large cohorts using linked EHRs from primary and secondary care is lacking regarding the incidence and prognostic factors associated with HF after a first MI in the current era. The aim of this research was, therefore, to determine using linked EHRs of population-based data from primary and secondary care the incidence of and prognostic factors associated with HF following MI.
METHODS Study design Record-linkage cohort study of data sources within the CALIBER (CArdiovascular disease research using LInked Bespoke studies and Electronic health Records) dataset, which includes linked data from (i) primary care EHR data using the diagnoses coded with the Read system (Clinical Practice Research Datalink, CPRD) from 244 consenting general practices,7 (ii) secondary care administrative records (information coded using the International Classification of Diseases (ICD) and the Office of Population Censuses and Surveys Classification of Interventions and Procedures (OPCS-4), hospital records (Hospital Episode Statistics, HES)), the Myocardial Ischaemia National Audit Project (MINAP) and (iii) the mortality register (from UK Office for National Statistics, ONS).8 Setting and participants All patients aged ≼18 years old, registered in CPRD practices in England consenting to data linkage, with at least one year of up-to-standard pre-study follow-up, and who experienced a first MI (see Supplementary for MI definition) between Jan 1, 1998, to March 25, 2010 were potentially eligible for inclusion.6 The first record of MI during the patient’s study period and across the linked data sources was considered as the index event and subsequent MI records within 30 days in the other data sources were considered as representing the same event. Patients with a fatal index MI or a history of HF before their index MI were excluded. Patients were censored at de-registration from the general practice, the date of death or the administrative censoring of the dataset (March 25, 2010).
21
PART ONE CHAPTER 2
Variables and data sources The variables used in this study have been previously defined, were derived from Read, ICD-9 or ICD-10, MINAP, Multilex drug or procedure codes, and the EHR phenotyping algorithms can be found online (https://caliberresearch.org/portal). For continuous variables, the most recent measurement recorded in CPRD in the year before study entry was used as a baseline value. Data before study entry were used to determine prognostic factors, including age (years), sex, ethnicity, social deprivation (index of multiple deprivation, IMD score as recorded in CPRD, in quintiles), smoking, alcohol use, history of cardiovascular disease, previous revascularization, history of diabetes, history of thyroid disease, history of chronic obstructive pulmonary disease (COPD) and history of non-metastatic cancer. Data in the year before study entry was collected, including body mass index (BMI), vital signs before hospitalisation, biomarkers and medications prescribed before index MI. Data from MINAP were used to determine infarct characteristics, including: type of MI, site of infarction, type of reperfusion therapy, delay from symptom onset to delivery of reperfusion therapy and peak cardiac biomarkers. The primary study outcome was the first recorded HF event following MI as previously defined and published online (https:// www.caliberresearch.org/portal/show/phenotype_hf). Events were defined in primary care by a diagnosis of HF or notification of left ventricular dysfunction on echocardiogram, in secondary care by a diagnosis of HF or mortality due to HF or in ONS by mortality due to HF (see Supplementary for HF definition). The first recorded event date that satisfied the pre-defined definition was used as the incident date of HF from different sources. Statistical methods Incidence rates (cases per 1000 person-years) and Kaplan-Meier cumulative incidence (cumulative percentage) of HF were calculated. Kaplan-Meier curves were used to assess unadjusted survival differences between incident HF after index MI, by age (<50, 50-65, ≥65 years) and type of MI. Cumulative incidence of HF was then adjusted for competing risk of mortality as first event.9 The associations of baseline variables with the onset of HF following MI were explored using Cox proportional hazard models. The proportional hazards assumption was checked graphically and using a chi-square test, throughout slight deviations were seen at the end of follow-up which could readily be approximated by additive time. All models were stratified for general practice and calendar period (3 year periods between 1998 and 2010). Models were sequentially adjusted for: (1) age and sex, (2) cardiovascular risk factors, (3) type of MI, and (4) co-morbidities and prescribed medication. Associations are presented as hazard ratio’s (HRs) with 95% confidence intervals (95%CI) and two-sided p-values using an alpha of 0.05. Analyses were performed using R (version 3.1.2). Sensitivity analyses Missing data for baseline smoking, BMI, and systolic and diastolic blood pressure were multiply imputed using the mice package,10 and results across 20 imputed datasets were pooled using Rubin’s rules (Supplementary).11 Finally, to account for the fact that patients may have died before developing HF after a first MI (e.g., competing risk by mortality), all analyses were repeated using Fine and Gray models.
22
HEART FAILURE FOLLOWING MI: A CALIBER STUDY
Ethical approval CALIBER was approved by the Lewisham Local Research Ethics Committee (ref: 09/H0810/16 date: 08/04/2009) and the Ethics and Confidentiality Committee (ECC) (ref: ECC 2-06(b)/2009 CALIBER dataset). CALIBER has been registered with the University College London Data Protection Officer (ref: Z6364106/2009/2/26). CALIBER EHR data are anonymised; individual
2
informed consent was not sought from study participants. This study was approved by the ISAC (Independent Scientific Advisory Committee) for MHRA Database Research (protocol no. 14_198R) and the MINAP Academic Group. This study is registered with ClinicalTrials.gov, number NCT02384213.
RESULTS Subjects and baseline characteristics We identified 52,270 patients with a first MI recorded in CALIBER between 1998 and 2010 (Figure 1). After exclusion of patients with a fatal index MI (n = 15,104), a prior history of HF (n = 3392) or those with insufficient baseline data (n = 9029) a cohort of (n = 24,745) index MI patients were included in primary analyses. This cohort comprised 4547 patients with STsegment-elevation MI (STEMI), 6892 with Non-STEMI (NSTEMI) and 13,306 with unclassified MI. Baseline characteristics are described in Table 1. Of all the MI patients 16,134 (65.2%) were men, 4594 (42.2%) were current smokers, 12,409 (50.1%) had a history of hypertension and 3056 (12.3%) had a history of diabetes at baseline. In total, 6154 (24.9%) were prescribed betablockers and 5062 (20.5%) ACE inhibitors before their index MI. Incidence Patients contributed 90,817 person-years of follow-up after MI to the study. During a median time of 3.7 years (IQR: 1.4 â&#x20AC;&#x201C; 6.7), 6005 patients (24.3%) developed HF. The crude incidence rate of HF following a first MI was 66.1 cases (95% CI: 64.4 - 67.8) per 1000 person-years. During the first 30 days from index MI 2581 (10.4%) patients developed HF (Figure 2A). In patients event free during the first 30 days, from day 30 onwards 2779/21286 (13.1%) of MI patients developed HF during 5-year follow-up (Figure 2B). Death occurred at the time of first HF presentation for 137 (2.3%) patients. The 30-day HF incidence was 5.2% (126/2409) in patients younger than 50 years, 6.5% (498/7624) in 50-65 year olds and 13.3% (1957/14712) in those 65 years and older (Figure 3A). During 5-year follow-up in patients event free during the first 30 days, subsequent cumulative estimates per age group were 5.9% (131/2208), 7.6% (526/6948) and 17.5% (2122/12130), respectively (Figure 3B). The 30-day incidence of HF was similar between NSTEMI (922/6892, 13.4%) and STEMI (610/4547, 13.4%) patients, but lower in the unclassified MI subgroup (1049/13306, 7.9%) (Figure 3C). Among patients who were event free during the first 30 days, a higher incidence of HF at 5 years was observed amongst NSTEMI (692/5746, 12.0%) compared with STEMI (384/3812, 10.1%) subgroup (Figure 3D).
23
PART ONE CHAPTER 2
Figure 1. Flowchart of study population CALIBER = Cardiovascular disease research using Linked Bespoke studies and Electronic health Records; CPRD = Clinical Practice Research Datalink; HF = heart failure; MI = myocardial infarction; NSTEMI = NonST-segment-elevation myocardial infarction; STEMI = ST-segment-elevation myocardial infarction.
Figure 2. Crude incidence of HF after index MI Kaplan-Meier curves for crude incidence of HF after MI. A. 30 day follow-up after index MI. B. 5 year followup in patients who survived the first 30 days and did not develop HF during the first 30 days (30 days event free). HF = heart failure; MI = myocardial infarction.
24
HEART FAILURE FOLLOWING MI: A CALIBER STUDY
2
Figure 3. Crude incidence of HF after index MI by subgroup Kaplan-Meier curves for crude incidence of HF after MI stratified by subgroup. A. 30 day follow-up after index MI stratified by age group. B. 5 year follow-up in 30 day event free patients stratified by age group. Log-rank test of >=65 year old patients compared to patients aged 50-65 years p < 0.001. C. 30 day follow-up after index MI stratified by MI subtype. D. 5 year follow-up in patients who survived the first 30 days and did not develop HF during the first 30 days stratified by MI subtype. Log-rank test of NSTEMI compared to STEMI p < 0.001. HF = heart failure; MI = myocardial infarction; NSTEMI = Non-ST-segment-elevation myocardial infarction; STEMI = ST-segment-elevation myocardial infarction.
Cumulative incidence in competing risks data At 30 days, 10.4% (2581/24735) developed HF and only one patient died without developing HF. Of patients event free during the first 30 days, from day 30 onwards 12.7% (2823/22147) developed HF and 4.7% (1048) died without first being diagnosed with HF (Figure 4). Of the patients with an index MI between 1998-2001 17.9% (938/5229) developed HF compared with 21.6% (1496/6939) of patients with MI between 2004-2007 during a 3-year follow-up. The estimated 3-year cumulative HF incidence in MI patients age 65 years and over between 19982001 was 23.5% (2977) compared with 27.2% (4219) between 2004-2007.
25
PART ONE CHAPTER 2
Table 1. Baseline characteristics at index myocardial infarction STEMI patients NSTEMI n = 4547 patients n = 6892
MI patients not Total MI further patients specified n = 24745 n = 13306
Unknown (%)
Follow-up time (years), median (IQR)
2.8 (1.2–4.6)
2.3 (0.9–4.1)
5.9 (2.3–8.7)
3.7 (1.4–6.7)
0%
Mean age, years (SD)
65.5 (13.4)
70.5 (13.6)
67.8 (12.8)
68.1 (13.2)
0%
Male sex
3243 (71.3%)
4160 (60.4%)
8731 (65.6%)
16134 (65.2%)
0%
White
3407 (77.4%)
5281 (79.1%)
9240 (74%)
17928 (76.1%)
4.8%
Asian
85 (1.9%)
104 (1.6%)
214 (1.7%)
403 (1.7%)
Black
19 (0.4%)
29 (0.4%)
43 (0.3%)
91 (0.4%)
Other
334 (7.6%)
456 (6.8%)
718 (5.8%)
1508 (6.4%)
Underweight
25 (1.8%)
60 (2.3%)
58 (1.7%)
143 (1.9%)
Normal
360 (26.6%)
720 (27.2%)
920 (27.2%)
2000 (27.1%)
Overweight
557 (41.1%)
1043 (39.4%)
1400 (41.4%)
3000 (40.7%)
Obese
412 (30.4%)
824 (31.1%)
1000 (29.6%)
2236 (30.3%)
918 (20.3%)
1282 (18.7%)
2731 (20.6%)
4931 (20%)
0.4%
Current smoker
1360 (50.5%)
1380 (33%)
1854 (46.3%)
4594 (42.2%)
56%
Excess alcohol
63 (9.4%)
113 (8.5%)
120 (8.6%)
296 (8.7%)
86.3%
History of atrial fibrillation
353 (7.8%)
1058 (15.4%)
1208 (9.1%)
2619 (10.6%)
n/a
History of hypertension
2095 (46.1%)
4060 (58.9%)
6254 (47%)
12409 (50.1%)
n/a
History of peripheral arterial disease
233 (5.1%)
598 (8.7%)
911 (6.8%)
1742 (7%)
n/a
PCI
757 (16.6%)
715 (10.4%)
1295 (9.7%)
2767 (11.2%)
n/a
CABG
116 (2.6%)
320 (4.6%)
538 (4%)
974 (3.9%)
n/a
Previous TIA
135 (3%)
401 (5.8%)
632 (4.7%)
1168 (4.7%)
n/a
Previous stroke
78 (1.7%)
185 (2.7%)
234 (1.8%)
497 (2%)
n/a
History of diabetes
511 (11.2%)
1056 (15.3%)
1489 (11.2%)
3056 (12.3%)
n/a
History of thyroid disease
254 (5.6%)
597 (8.7%)
776 (5.8%)
1627 (6.6%)
n/a
History of COPD
302 (6.6%)
683 (9.9%)
973 (7.3%)
1958 (7.9%)
n/a
History of non-metastatic cancer
483 (10.6%)
970 (14.1%)
1335 (10%)
2788 (11.3%)
n/a
Heart rate, beats/min
73.0 (64–84)
72.0 (64–84)
75.5 (66–84)
73 (64–84)
91.1%
Systolic blood pressure, mmHg
140 (130–150)
140 (129–150)
140 (130–158)
140 (130–154)
34.6%
Diastolic blood pressure, mmHg
80 (74–87)
80 (70–85)
80 (73–90)
80 (72–88)
34.6%
Ethnicity
Body mass index (BMI) 70.2%
Index of multiple deprivation Most deprived quintile Risk factors before index MI
Previous revascularization
Vital signs before admission median (IQR)
Biomarkers median (IQR)
26
Troponin I (maximum)
19.38 (3.8–50.0) 1.73 (0.29–7.4)
Troponin T (maximum)
1.86 (0.6–4.8)
CK (maximum)
740 (226.0–1687) 189 (89.0–512)
0.30 (0.19–1.02) 3.88 (0.62–21.4) 88.1%
0.33 (0.13–0.87) 0.23 (0.12–0.24) 0.65 (0.18–2) 122 (63.2–148)
331 (123–1068)
90.6% 79.7%
HEART FAILURE FOLLOWING MI: A CALIBER STUDY
Table 1. Continued STEMI patients NSTEMI n = 4547 patients n = 6892
MI patients not Total MI further patients specified n = 24745 n = 13306
Unknown (%)
Haemoglobin, g/dL
14.1 (1.72)
13.5 (1.87)
13.7 (1.81)
13.7 (1.83)
67.3%
White blood cell count
8.06 (2.96)
7.83 (2.79)
7.93 (3.04)
7.91 (2.92)
69.3%
Neutrophil count
5.03 (2.13)
4.97 (2.32)
5.01 (2.48)
5 (2.35)
72.1%
Platelets
272 (84.6)
267 (89.5)
264 (86.6)
267 (87.5)
69.3%
Erythrocyte sedimentation rate
18.6 (18.2)
22.4 (21.7)
22.4 (21.8)
21.8 (21.2)
89.5%
Creatinine, Âľmol/L
99.3 (40.9)
102.1 (47.4)
102.6 (47.1)
102 (46.1)
59%
eGFR - CKD-EPI
70.6 (19.6)
66.1 (20.3)
66.2 (19.6)
67 (20)
60.7%
Random glucose concentration, mmol/L
6.90 (3.36)
6.95 (3.34)
7.61 (4.33)
7.23 (3.82)
76.7%
Fasting glucose concentration, mmol/L 6.07 (2.02)
6.00 (1.99)
6.50 (2.62)
6.18 (2.22)
92.8%
HbA1c, mmol/mol
60.4 (19.8)
57.6 (16.8)
59.0 (17.8)
58.6 (17.8)
90%
Total cholesterol
5.33 (1.42)
5.04 (1.25)
5.43 (1.29)
5.27 (1.31)
63.7%
LDL cholesterol
3.24 (1.11)
2.93 (1.08)
3.23 (1.11)
3.1 (1.11)
81.4%
Biomarkers before index MI mean (SD)
2
STEMI Site of infarction Anterior
518 (40.5%)
71.8%
Primary PCI
332 (19.4%)
62.4%
Prehospital fibrinolysis
432 (9.5%)
n/a
In-hospital fibrinolysis
1030 (60.3%)
62.4%
Median (IQR) delay from symptom to reperfusion (min)
150 (97â&#x20AC;&#x201C;280)
37.5%
Prescribed medication before index MI Antiplatelet
815 (17.9%)
2516 (36.5%)
4555 (34.2%)
7886 (31.9%)
n/a
Oral anticoagulant
69 (1.5%)
260 (3.8%)
363 (2.7%)
692 (2.8%)
n/a
Statin
882 (19.4%)
2269 (32.9%)
3006 (22.6%)
6157 (24.9%)
n/a
ACE inhibitor
692 (15.2%)
1643 (23.8%)
2727 (20.5%)
5062 (20.5%)
n/a
Angiotensin receptor blocker (ARB)
250 (5.5%)
611 (8.9%)
575 (4.3%)
1436 (5.8%)
n/a
Beta-blocker
729 (16%)
1782 (25.9%)
3643 (27.4%)
6154 (24.9%)
n/a
Calcium channel blocker
786 (17.3%)
1877 (27.2%)
2791 (21%)
5454 (22%)
n/a
Loop diuretic
227 (5%)
850 (12.3%)
1380 (10.4%)
2457 (9.9%)
n/a
Aldosterone antagonist
25 (0.5%)
77 (1.1%)
121 (0.9%)
223 (0.9%)
n/a
Digoxin
50 (1.1%)
210 (3%)
309 (2.3%)
569 (2.3%)
n/a
IQR = inter-quartile range; SD = standard deviation; MI = myocardial infarction; NSTEMI = non-ST-segment-elevation myocardial infarction; STEMI = ST-segment-elevation myocardial infarction; PCI = Percutaneous Coronary Intervention; CABG = Coronary Artery Bypass Grafting; TIA = Transient Ischaemic Attack; COPD = Chronic Obstructive Pulmonary Disease; CK = Creatine kinase; ACE = Angiotensin-converting-enzyme. Prehospital and in-hospital fibrinolysis are not mutually exclusive.
27
PART ONE CHAPTER 2
Table 2. Hazard ratio for heart failure in patients following a first myocardial infarction using multivariable Cox regression Model 1 n = 24458 n events = 5961
Model 2 n = 24458 n events = 5961
Model 3 n = 24458 n events = 5961
Model 4 n = 24458 n events = 5961
HR (95% CI)
HR (95% CI)
HR (95% CI)
HR (95% CI)
Age, per 10 years
1.51 (1.48 – 1.55)
1.49 (1.46 – 1.53)
1.46 (1.43 – 1.50)
1.45 (1.41 – 1.49)
Men
1.06 (1.00 – 1.12)
1.08 (1.02 – 1.14)
1.07 (1.01 – 1.13)
1.06 (1.00 – 1.12)
Index of multiple deprivation
overall p = <0.001 overall p = <0.001 overall p = <0.001
Q1 (least deprived)
Reference
Reference
Reference
Q2
1.09 (0.99 – 1.20)
1.08 (0.99 – 1.19)
1.08 (0.98 – 1.19)
Q3
1.20 (1.09 – 1.33)
1.20 (1.09 – 1.33)
1.19 (1.07 – 1.31)
Q4
1.20 (1.08 – 1.33)
1.20 (1.08 – 1.33)
1.17 (1.06 – 1.30)
Q5 (most deprived)
1.30 (1.16 – 1.45)
1.30 (1.17 – 1.46)
1.27 (1.13 – 1.42)
History of hypertension
1.19 (1.12 – 1.26)
1.17 (1.10 – 1.23)
1.16 (1.09 – 1.23)
History of diabetes
1.47 (1.37 – 1.58)
1.47 (1.37 – 1.59)
1.44 (1.34 – 1.55)
History of atrial fibrillation
1.65 (1.53 – 1.77)
1.63 (1.51 – 1.75)
Type of MI
overall p = <0.001 overall p = <0.001
NSTEMI
Reference
Reference
Unclassified MI vs. NSTEMI
0.89 (0.82 – 0.97)
0.89 (0.82 – 0.97)
STEMI vs. NSTEMI
1.19 (1.10 – 1.30)
1.21 (1.11 – 1.32)
History of peripheral arterial disease
1.38 (1.26 – 1.51)
History of COPD
1.28 (1.17 – 1.40)
Prescribed ACE inhibitor before MI
1.07 (1.00 – 1.14)
Prescribed Angiotensin receptor blocker (ARB) use before MI
1.00 (0.89 – 1.12)
Prescribed Beta-blocker before MI
0.94 (0.88 – 1.00)
HR = hazard ratio; CI = confidence interval; MI = myocardial infarction; NSTEMI = non-ST-segment-elevation myocardial infarction; STEMI = ST-segment-elevation myocardial infarction; COPD = chronic obstructive pulmonary disease; ACE = Angiotensin-converting enzyme.
Prognostic factors In multivariable Cox regression with complete-case analysis (n = 24,458), increasing age [HR per 10 years increase: 1.45 (95%CI 1.41 – 1.49)], higher socioeconomic deprivation (3rd [HR 1.19 (95%CI 1.07 – 1.31)] – 5th [HR 1.27 (95%CI 1.13 – 1.42)] vs. 1st quintile), a history of hypertension [HR 1.16 (95%CI 1.09 – 1.23)], diabetes [HR 1.44 (95%CI 1.34 – 1.55)], atrial fibrillation [HR 1.63 (95%CI 1.51 – 1.75)], peripheral arterial disease [HR 1.38 (95%CI 1.26 – 1.51)], COPD [HR 1.28 (95%CI 1.17 – 1.40)] and an index STEMI presentation [HR 1.21 (95%CI 1.11 – 1.32)] were all associated with increased ratios of HF (Table 2). These clinical factors also predicted outcomes in competing risks analysis, with the difference of an unclassified MI [HR 1.17 (95%CI 1.10 – 1.24)] and beta-blocker prescriptions in the year prior to index MI [HR 0.93 (95%CI 0.87 – 0.99)].
28
HEART FAILURE FOLLOWING MI: A CALIBER STUDY
In sensitivity analyses, the results of the adjusted complete-case analysis (n = 4911) were similar to those obtained from the multiple imputation analysis except that we found a negative association of being overweight at index MI [HR: 0.82 (95%CI 0.75 – 0.90)], a positive association of smoking [HR: 1.17 (95%CI 1.07 – 1.28)] and no association of systolic [HR: 1.00 (95%CI 1.00 – 1.00)] or diastolic blood pressure [HR: 1.00 (95%CI 1.00 – 1.00)] on admission
2
(Supplementary).
Figure 4. Cumulative incidence curves for HF adjusted for mortality as first event after index MI Cumulative incidence curves adjusted for competing risk of mortality as first event. 5-year follow-up in patients who survived the first 30 days and did not develop HF during the first 30 days (30 days event free).
DISCUSSION In this large population-based longitudinal study using linked EHRs data from primary care, hospital admissions, a MI registry and the death register, one quarter of patients who experienced a first MI developed HF during a median follow-up of 3.7 years. The incidence rate of HF following a first MI was 66 cases per 1000 person-years. Moreover, the 3-year incidence was high - over a quarter of patients aged 65 years or older developed HF. Increasing age, diabetes and atrial fibrillation, were important prognostic factors associated with the risk of incident HF following MI. HF incidence We found that HF incidence following MI is high in the first 30 days after index MI (10.4%). From day 30 onwards, there was a more gradual increase in the proportion of HF patients to 13.1% during the 5-year follow-up. The cumulative incidence of HF following MI derived from competing risk analyses was similar. These findings are comparable with the decreasing trend
29
PART ONE CHAPTER 2
seen in the incidence of HF patients after myocardial infarction from another study by comparing within 30 days of admission, 30-days to 1 year and 1-3 years follow-up.12 There is uncertainty whether the HF incidence is increasing13, 14 or decreasing3, 4, 12, 15, 16 in calendar time, the majority of these studies had short-term follow-up and did not include patients managed in primary care. Ezekowitz et al. found a relative 25% increase comparing 1994 to 2000 5-year in-hospital HF rate in an elderly (â&#x2030;Ľ65 years of age) MI patient cohort.14 Using a national sample of Medicare beneficiaries in the United States of America, the HF hospitalization following acute MI decreased from 16.1 per 100 person-years in 1998 to 14.2 per 100 person years in 2010 (p<0.001).16 We did not observe an altered HF incidence in this study that might be elucidated due to capturing milder HF cases, managed in primary care, from multiple linked sources. The low cumulative incidence of mortality as first event during the first 30 days can be attributed to exclusion of patients with a fatal index MI. We found that 28.9% (n = 15,104) of total MIâ&#x20AC;&#x2122;s were fatal which was in line with our broader definition, in comparison to death within seven days of the index MI6 we used a 30-day interval to define fatal MI. Our findings regarding 3-year cumulative HF incidence are in line with the overall HF incidence after a first MI from 2002-2004 (35-64 year: 11.52% and 65-84 year: 27.96%) in a Swedish study (1993-2004).12 They also found a higher incidence of HF in women (6%) compared to men, where we did not found a significant association of gender with HF incidence. We know that the uptake of effective treatment for STEMI, such as primary PCI was slower in the UK than in Sweden.17 This could explain the relatively low percentage of primary PCI for STEMI in our study (19.4%). The higher cumulative incidence in NSTEMI compared to STEMI patients might be explained by the on average higher age with co-morbidities in NSTEMI compared to STEMI patients. Prognostic factors In a previous UK study incidence of HF after MI during 1998 also steeply increased with age.18 In the latter study, of patients <65 years of age half developed HF during a 6-year follow-up compared to 73% of patients aged 65-75 years and 87% of patients aged >75 years. Socioeconomic deprivation has previously been shown to be an independent predictor of HF development and associated with an increased incidence of HF.19-21 Socioeconomic deprivation has also been associated with more frequent hospital admission and higher mortality in patients with HF.22, 23 In a previous CALIBER study with linked EHRs the hazards of HF increased linearly with higher socioeconomic deprivation in patients initially free of CVD.24 In the current study, we observed that higher socioeconomic deprivation in patients with MI also increased the risk for incident HF. Several other studies also previously linked co-morbidities before admission hypertension, diabetes, atrial fibrillation, peripheral arterial disease and COPD to an increased risk of HF following MI.12, 14, 25-27 In a previous multivariable analysis, users of beta-blockers after the index MI were less likely to develop HF.14 In competing risks analyses, we found that betablocker use in the year before MI is an independent predictor for a lower risk of HF after MI. In summary, these prognostic factors can be used to identify patients suitable for clinical trials of novel therapies to target HF following MI.
30
HEART FAILURE FOLLOWING MI: A CALIBER STUDY
Strengths and limitations The linkage of multiple EHR sources from primary and secondary care leads to a more accurate estimation regarding incidence and prognostic factors associated with HF following MI. This allowed for â&#x20AC;&#x2DC;real worldâ&#x20AC;&#x2122; data from a relatively large sample size with contemporary therapy and to adjust for a large range of possible confounders. The CPRD consists of a sample of English
2
general practices, but seem representative of the UK population.28 The ascertainment of cardiovascular outcomes was not based on clinical criteria (e.g., validated questionnaires and properly conducted physical examinations) and there may have been changes in medical coding over time. However, we used stringent code lists to determine variables and outcomes, which have been previously defined, validated and published (https://caliberresearch.org/portal).6-8 Using the first MI recorded in the database without a prior history of HF, might have introduced bias due to left truncation or selection bias.29 There is a possible delay between primary care and secondary care records, as we know from previous research that CPRD tends to record MI after HES or MINAP admission dates.6 The lower 30-day HF incidence in unclassified MI patients primarily derived from primary care could be partly explained because of a delay in coding. Therefore we showed cumulative incidence rates in patients (alive and HF event free within the first 30 days) from 30 days after index MI to account for a delay in recording of MI in primary care. We were unable to differentiate between HF with preserved ejection fraction (HFpEF) and HF with reduced ejection fraction (HFrEF) as we had no access to detailed (echocardiographic) parameters to assess diastolic dysfunction, but it is likely that the majority of our HF patients had developed systolic dysfunction after MI.
CONCLUSION In this large cohort study using linked EHRs in the UK from primary and secondary care we observed that about 1 in 4 people will develop HF within 4 years after the occurrence of a first MI and increasing age, higher socioeconomic deprivation, a history of hypertension, diabetes, atrial fibrillation, peripheral arterial disease or COPD and a STEMI at presentation were important risk factors. Identified amendable prognostic factors can be used to decrease the incidence of HF following MI.
31
PART ONE CHAPTER 2
REFERENCES 1. 2.
3.
4.
5. 6.
7. 8.
9. 10. 11.
12.
13.
14.
15. 16.
17.
18. 19.
32
Mosterd A, Hoes AW. Clinical epidemiology of heart failure. Heart. 2007;93:1137-46. Torabi A, Cleland JG, Khan NK, Loh PH, Clark AL, Alamgir F, et al. The timing of development and subsequent clinical course of heart failure after a myocardial infarction. Eur Heart J. 2008;29:859-70. Gjesing A, Gislason GH, Kober L, Gustav Smith J, Christensen SB, Gustafsson F, et al. Nationwide trends in development of heart failure and mortality after first-time myocardial infarction 1997-2010: A Danish cohort study. Eur J Intern Med. 2014;25:731-8. Desta L, Jernberg T, Lofman I, Hofman-Bang C, Hagerman I, Spaak J, et al. Incidence, Temporal Trends, and Prognostic Impact of Heart Failure Complicating Acute Myocardial Infarction: The SWEDEHEART Registry (Swedish Web-System for Enhancement and Development of Evidence-Based Care in Heart Disease Evaluated According to Recommended Therapies): A Study of 199,851 Patients Admitted With Index Acute Myocardial Infarctions, 1996 to 2008. JACC Heart Fail. 2015;3:234-42. Hripcsak G, Albers DJ. Next-generation phenotyping of electronic health records. J Am Med Inform Assoc. 2013;20:117-21. Herrett E, Shah AD, Boggon R, Denaxas S, Smeeth L, van Staa T, et al. Completeness and diagnostic validity of recording acute myocardial infarction events in primary care, hospital care, disease registry, and national mortality records: cohort study. BMJ. 2013;346:f2350. Herrett E, Gallagher AM, Bhaskaran K, Forbes H, Mathur R, van Staa T, et al. Data Resource Profile: Clinical Practice Research Datalink (CPRD). Int J Epidemiol. 2015. Denaxas SC, George J, Herrett E, Shah AD, Kalra D, Hingorani AD, et al. Data resource profile: cardiovascular disease research using linked bespoke studies and electronic health records (CALIBER). Int J Epidemiol. 2012;41:1625-38. Satagopan JM, Ben-Porat L, Berwick M, Robson M, Kutler D, Auerbach AD. A note on competing risks in survival data analysis. Br J Cancer. 2004;91:1229-35. van Buuren S. Multiple imputation of discrete and continuous data by fully conditional specification. Stat Methods Med Res. 2007;16:219-42. Marshall A, Altman DG, Holder RL, Royston P. Combining estimates of interest in prognostic modelling studies after multiple imputation: current practice and guidelines. BMC Med Res Methodol. 2009;9:57. Shafazand M, Rosengren A, Lappas G, Swedberg K, Schaufelberger M. Decreasing trends in the incidence of heart failure after acute myocardial infarction from 1993-2004: a study of 175,216 patients with a first acute myocardial infarction in Sweden. Eur J Heart Fail. 2011;13:135-41. Velagaleti RS, Pencina MJ, Murabito JM, Wang TJ, Parikh NI, Dâ&#x20AC;&#x2122;Agostino RB, et al. Long-term trends in the incidence of heart failure after myocardial infarction. Circulation. 2008;118:205762. Ezekowitz JA, Kaul P, Bakal JA, Armstrong PW, Welsh RC, McAlister FA. Declining in-hospital mortality and increasing heart failure incidence in elderly patients with first myocardial infarction. J Am Coll Cardiol. 2009;53:13-20. Fox KA, Steg PG, Eagle KA, Goodman SG, Anderson FA, Jr., Granger CB, et al. Decline in rates of death and heart failure in acute coronary syndromes, 1999-2006. JAMA. 2007;297:1892-900. Chen J, Hsieh AF, Dharmarajan K, Masoudi FA, Krumholz HM. National trends in heart failure hospitalization after acute myocardial infarction for Medicare beneficiaries: 1998-2010. Circulation. 2013;128:2577-84. Chung SC, Gedeborg R, Nicholas O, James S, Jeppsson A, Wolfe C, et al. Acute myocardial infarction: a comparison of short-term survival in national outcome registries in Sweden and the UK. Lancet. 2014;383:1305-12. Torabi A, Cleland JG, Rigby AS, Sherwi N. Development and course of heart failure after a myocardial infarction in younger and older people. J Geriatr Cardiol. 2014;11:1-12. McAlister FA, Murphy NF, Simpson CR, Stewart S, MacIntyre K, Kirkpatrick M, et al. Influence of socioeconomic deprivation on the primary care burden and treatment of patients with a diagnosis of heart failure in general practice in Scotland: population based study. BMJ. 2004;328:1110.
HEART FAILURE FOLLOWING MI: A CALIBER STUDY
20. Hawkins NM, Jhund PS, McMurray JJ, Capewell S. Heart failure and socioeconomic status: accumulating evidence of inequality. Eur J Heart Fail. 2012;14:138-46. 21. Ramsay SE, Whincup PH, Papacosta O, Morris RW, Lennon LT, Wannamethee SG. Inequalities in heart failure in older men: prospective associations between socioeconomic measures and heart failure incidence in a 10-year follow-up study. Eur Heart J. 2014;35:442-7. 22. Philbin EF, Dec GW, Jenkins PL, DiSalvo TG. Socioeconomic status as an independent risk factor for hospital readmission for heart failure. Am J Cardiol. 2001;87:1367-71. 23. Struthers AD, Anderson G, Donnan PT, MacDonald T. Social deprivation increases cardiac hospitalisations in chronic heart failure independent of disease severity and diuretic nonadherence. Heart. 2000;83:12-6. 24. Pujades-Rodriguez M, Timmis A, Stogiannis D, Rapsomaniki E, Denaxas S, Shah A, et al. Socioeconomic deprivation and the incidence of 12 cardiovascular diseases in 1.9 million women and men: implications for risk prediction and prevention. PLoS ONE. 2014;9:e104671. 25. Ali AS, Rybicki BA, Alam M, Wulbrecht N, Richer-Cornish K, Khaja F, et al. Clinical predictors of heart failure in patients with first acute myocardial infarction. Am Heart J. 1999;138:1133-9. 26. Lewis EF, Moye LA, Rouleau JL, Sacks FM, Arnold JM, Warnica JW, et al. Predictors of late development of heart failure in stable survivors of myocardial infarction: the CARE study. J Am Coll Cardiol. 2003;42:1446-53. 27. Lewis EF, Velazquez EJ, Solomon SD, Hellkamp AS, McMurray JJ, Mathias J, et al. Predictors of the first heart failure hospitalization in patients who are stable survivors of myocardial infarction complicated by pulmonary congestion and/or left ventricular dysfunction: a VALIANT study. Eur Heart J. 2008;29:748-56. 28. Parkinson JP, Davis S, van Staa T. The General Practice Research Database: now and the future. Mann R, Andrews EB, eds Pharmacovigilance: John Wiley; 2007. p. 341-8. 29. Hazelbag CM, Klungel OH, van Staa TP, de Boer A, Groenwold RH. Left truncation results in substantial bias of the relation between time-dependent exposures and adverse events. Ann Epidemiol. 2015;25:590-6.
2
33
PART ONE CHAPTER 2
SUPPLEMENTAL MATERIAL Table S1. Hazard ratio for heart failure in patients following a first myocardial infarction using the competing risks regression based on Fine and Gray’s proportional subhazards model to account for the competing risk of all cause mortality Model 1 n = 24458
Model 2 n = 24458
Model 3 n = 24458
Model 4 n = 24458
HR (95% CI)
HR (95% CI)
HR (95% CI)
HR (95% CI)
Age, per 10 years increase
1.41 (1.38 – 1.44)
1.39 (1.36 – 1.42)
1.37 (1.34 – 1.40)
1.36 (1.33 – 1.39)
Men
1.04 (0.99 – 1.10)
1.06 (1.01 – 1.12)
1.06 (1.00 – 1.12)
1.05 (1.00 – 1.11)
overall p = <0.001
overall p = <0.001
overall p = <0.001
Q1
Reference
Reference
Reference
Q2
1.11 (1.02 – 1.21)
1.11 (1.02 – 1.20)
1.10 (1.02 – 1.20)
Q3
1.21 (1.12 – 1.32)
1.21 (1.11 – 1.31)
1.19 (1.10 – 1.30)
Q4
1.24 (1.14 – 1.35)
1.24 (1.14 – 1.35)
1.22 (1.12 – 1.32)
Q5
1.35 (1.25 – 1.47)
1.35 (1.24 – 1.46)
1.31 (1.20 – 1.42)
History of hypertension
1.15 (1.09 – 1.21)
1.14 (1.08 – 1.20)
1.15 (1.08 – 1.22)
History of diabetes
1.38 (1.29 – 1.48)
1.39 (1.29 – 1.49)
1.37 (1.27 – 1.47)
History of atrial fibrillation
1.52 (1.41 – 1.63)
1.51 (1.40 – 1.62)
Type of MI
overall p = <0.001
overall p = <0.001
NSTEMI
Reference
Reference
Unclassified MI vs. NSTEMI
1.16 (1.09 – 1.23)
1.17 (1.10 – 1.24)
STEMI vs. NSTEMI
1.14 (1.05 – 1.24)
1.15 (1.06 – 1.24)
Index of multiple deprivation
History of peripheral arterial disease
1.29 (1.19 – 1.41)
History of COPD
1.19 (1.09 – 1.29)
ACE inhibitor use before MI
1.01 (0.95 – 1.08)
Angiotensin receptor blocker (ARB) use before MI
0.96 (0.87 – 1.07)
Beta-blocker use before MI
0.93 (0.87 – 0.99)
Overall p-values are shown for categorical covariates. ACE = Angiotensin-converting enzyme; CI, confidence interval; COPD = chronic obstructive pulmonary disease; MI = myocardial infarction; NSTEMI = non-ST-segment-elevation myocardial infarction; STEMI = ST-segment-elevation myocardial infarction; Q = quintile; HR, ratio of the subdistribution hazards.
34
HEART FAILURE FOLLOWING MI: A CALIBER STUDY
Table S2. Hazard ratio for heart failure in patients following a first myocardial infarction in sensitivity analyses Model 5 without multiple imputation
Model 5 with multiple imputation
Model 5 competing risks
HR (95% CI)
HR (95% CI)
HR (95% CI)
Age, per 10 years increase
1.49 (1.38 – 1.60)
1.48 (1.44 – 1.52)
1.45 (1.36 – 1.54)
Men
1.04 (0.90 – 1.20)
1.07 (1.01 – 1.13)
1.07 (0.90 – 1.20)
Body mass index (BMI)
p = 0.006
p = <0.001
p = < 0.001
Underweight
0.79 (0.48 – 1.32)
0.74 (0.56 – 0.99)
0.71 (0.48 – 1.32)
Normal
Reference
Reference
Reference
Overweight
0.84 (0.71 – 0.99)
0.82 (0.75 – 0.90)
0.84 (0.71 – 0.99)
Obese
1.12 (0.94 – 1.34)
0.97 (0.88 – 1.07)
1.17 (0.94 – 1.34)
p = 0.040
p = 0.001
p = < 0.001
Index of multiple deprivation Q1
Reference
Reference
Reference
Q2
1.22 (0.94 – 1.58)
1.08 (0.98 – 1.19)
1.35 (0.94 – 1.58)
Q3
1.34 (1.03 – 1.75)
1.18 (1.06 – 1.30)
1.39 (1.03 – 1.75)
Q4
1.24 (0.94 – 1.63)
1.16 (1.05 – 1.29)
1.31 (0.94 – 1.63)
Q5
1.46 (1.09 – 1.94)
1.25 (1.12 – 1.40)
1.61 (1.09 – 1.94)
Smoking status
p = 0.85
p = 0.002
p = 0.93
Non-smoker
Reference
Reference
Reference
Current smoker
0.95 (0.78 – 1.15)
1.17 (1.07 – 1.28)
1.03 (0.78 – 1.15)
Ex-smoker
0.98 (0.83 – 1.15)
1.04 (0.95 – 1.13)
0.99 (0.83 – 1.15)
History of hypertension
1.13 (0.95 – 1.35)
1.16 (1.09 – 1.24)
1.19 (0.95 – 1.35)
History of diabetes
1.26 (1.08 – 1.47)
1.47 (1.36 – 1.58)
1.24 (1.08 – 1.47)
History of atrial fibrillation
1.48 (1.25 – 1.76)
1.64 (1.52 – 1.76)
1.37 (1.25 – 1.76)
Type of MI
p = 0.024
p = <0.001
p = 0.69
NSTEMI
Reference
Reference
Reference
Unclassified MI vs. NSTEMI
0.83 (0.70 – 0.99)
0.89 (0.81 – 0.96)
1.02 (0.70 – 0.99)
STEMI vs. NSTEMI
1.06 (0.89 – 1.27)
1.20 (1.10 – 1.31)
1.07 (0.89 – 1.27)
History of peripheral arterial disease
1.55 (1.27 – 1.89)
1.36 (1.24 – 1.49)
1.35 (1.27 – 1.89)
History of COPD
1.19 (0.97 – 1.46)
1.24 (1.13 – 1.36)
1.22 (0.97 – 1.46)
Systolic blood pressure, mmHg
1.00 (0.99 – 1.00)
1.00 (1.00 – 1.00)
1.00 (0.99 – 1.00)
Diastolic blood pressure, mmHg
1.00 (0.99 – 1.01)
1.00 (1.00 – 1.00)
1.00 (0.99 – 1.01)
ACE inhibitor use before MI
1.01 (0.87 – 1.18)
1.08 (1.00 – 1.15)
1.06 (0.87 – 1.18)
Angiotensin receptor blocker (ARB) use before MI
0.95 (0.77 – 1.18)
1.01 (0.91 – 1.13)
0.93 (0.77 – 1.18)
Beta-blocker use before MI
0.87 (0.75 – 1.01)
0.95 (0.89 – 1.01)
0.90 (0.75 – 1.01)
2
Using multivariable Cox regression without and with multiple imputation (n = 4911, n events = 1160) and the competing risks regression based on Fine and Gray’s proportional subhazards model to account for the competing risk of all cause mortality Additional prognostic covariates: body mass index, smoking, systolic and diastolic blood pressure. Overall p-values are shown for categorical covariates. HR = hazard ratio; CI = confidence interval; MI = myocardial infarction; NSTEMI = non-ST-segment-elevation myocardial infarction; STEMI = ST-segment-elevation myocardial infarction; COPD = chronic obstructive pulmonary disease; ACE = Angiotensin-converting enzyme; Q = quintile.
35
PART ONE CHAPTER 2
Multiple imputation algorithm Multiple imputation was implemented using the â&#x20AC;&#x2DC;miceâ&#x20AC;&#x2122; package for R, to replace missing values in prognostic covariates (body mass index, smoking, systolic and diastolic blood pressure).1 Imputation models included general practice and 3-year periods from 1998 to 2010 of index myocardial infarction (MI) and: 1. All the baseline covariates as used in the multivariable analysis (age, sex, index of multiple deprivation, hypertension, diabetes, type of MI, body mass index, smoking, systolic blood pressure, diastolic blood pressure); 2. Baseline measurements of covariates not considered in the multivariable analysis (alcohol intake, haemoglobin, white blood cell count, creatinine, alanine aminotransferase (ALAT), total cholesterol, HDL cholesterol); 3. Baseline medications prior to index MI (antiplatelet, oral anticoagulant, statin, ACE inhibitor, angiotensin receptor blocker (ARB), beta-blocker, calcium channel blocker, loop diuretic, aldosterone antagonist, digoxin); 4. Coexisting medical conditions (history of atrial fibrillation, peripheral arterial disease, depression, cancer, chronic obstructive pulmonary disease); 5. The Nelson-Aalen hazard and the event status for the outcomes analysed.
REFERENCES 1.
36
van Buuren S. Multiple imputation of discrete and continuous data by fully conditional specification. Stat Methods Med Res. 2007;16:219-42.
HEART FAILURE FOLLOWING MI: A CALIBER STUDY
Phenotype coding lists CALIBER myocardial infarction phenotype Fatal or non-fatal myocardial infarction (MI) as recorded in any of the constituent CALIBER data sources. The type of myocardial infarction (ST elevation MI, STEMI; or non ST elevation MI, NSTEMI) is recorded only in MINAP and GPRD. Read codes used to identify myocardial infarction in primary care records (Clinical Research Practice Datalink). Category
Clinical code
Clinical term
2
Datasource lookup
STEMI (3)
G30X000
Acute ST segment elevation myocardial infarction
NSTEMI (4)
G307100
Acute non-ST segment elevation myocardial infarction
10562
Acute MI not further specified (5)
323..00
ECG: myocardial infarction
7783
12229
Acute MI not further specified (5)
3233.00
ECG: antero-septal infarct.
26975
Acute MI not further specified (5)
3234.00
ECG:posterior/inferior infarct
26972 55401
Acute MI not further specified (5)
3235.00
ECG: subendocardial infarct
Acute MI not further specified (5)
3236.00
ECG: lateral infarction
52705
Acute MI not further specified (5)
323Z.00
ECG: myocardial infarct NOS
59032
Acute MI not further specified (5)
889A.00
Diab mellit insulin-glucose infus acute myocardial infarct
61670
Acute MI not further specified (5)
G30..00
Acute myocardial infarction
241
Acute MI not further specified (5)
G30..12
Coronary thrombosis
2491
Acute MI not further specified (5)
G30..13
Cardiac rupture following myocardial infarction (MI)
30421
Acute MI not further specified (5)
G30..15
MI - acute myocardial infarction
1677
Acute MI not further specified (5)
G30..16
Thrombosis - coronary
13571 12139
Acute MI not further specified (5)
G300.00
Acute anterolateral infarction
Acute MI not further specified (5)
G301.00
Other specified anterior myocardial infarction
5387
Acute MI not further specified (5)
G301000
Acute anteroapical infarction
40429
Acute MI not further specified (5)
G301100
Acute anteroseptal infarction
17872
Acute MI not further specified (5)
G301z00
Anterior myocardial infarction NOS
14897 8935
Acute MI not further specified (5)
G302.00
Acute inferolateral infarction
Acute MI not further specified (5)
G303.00
Acute inferoposterior infarction
29643
Acute MI not further specified (5)
G304.00
Posterior myocardial infarction NOS
23892 14898
Acute MI not further specified (5)
G305.00
Lateral myocardial infarction NOS
Acute MI not further specified (5)
G306.00
True posterior myocardial infarction
63467
Acute MI not further specified (5)
G307.00
Acute subendocardial infarction
3704
Acute MI not further specified (5)
G307000
Acute non-Q wave infarction
9507
Acute MI not further specified (5)
G308.00
Inferior myocardial infarction NOS
1678
Acute MI not further specified (5)
G309.00
Acute Q-wave infarct
30330
Acute MI not further specified (5)
G30B.00
Acute posterolateral myocardial infarction
32854
Acute MI not further specified (5)
G30X.00
Acute transmural myocardial infarction of unspecif site
29758
Acute MI not further specified (5)
G30y.00
Other acute myocardial infarction
34803 28736
Acute MI not further specified (5)
G30y000
Acute atrial infarction
Acute MI not further specified (5)
G30y100
Acute papillary muscle infarction
62626
Acute MI not further specified (5)
G30y200
Acute septal infarction
41221
Acute MI not further specified (5)
G30yz00
Other acute myocardial infarction NOS
46017
Acute MI not further specified (5)
G30z.00
Acute myocardial infarction NOS
14658
Acute MI not further specified (5)
G31y100
Microinfarction of heart
68357
Acute MI not further specified (5)
G38..00
Postoperative myocardial infarction
32272
Acute MI not further specified (5)
G380.00
Postoperative transmural myocardial infarction anterior wall
46112
Acute MI not further specified (5)
G381.00
Postoperative transmural myocardial infarction inferior wall
46276
Acute MI not further specified (5)
G384.00
Postoperative subendocardial myocardial infarction
41835
Acute MI not further specified (5)
G38z.00
Postoperative myocardial infarction, unspecified
68748
Acute MI not further specified (5)
Gyu3400
[X]Acute transmural myocardial infarction of unspecif site
96838
37
PART ONE CHAPTER 2
Hospital Episode Statistics Category (code)
ICD10 code
ICD10 term
Acute MI not further specified (5)
I21
Acute myocardial infarction
Category (code)
OPCS 4
OPCS 4 term
Transluminal coronary thrombolysis (2)
K50.2
Percutaneous transluminal coronary thrombolysis using streptokinase
Transluminal coronary thrombolysis (2)
K50.3
Percutaneous transluminal injection of therapeutic substance into coronary artery NEC
OPCS
Office for National Statistics (ONS) Category (code)
ICD10 code
ICD10 term
Death from MI (3)
I21
Acute myocardial infarction
Death from MI (3)
I22
Subsequent myocardial infarction
Death from MI (3)
I23
Certain current complications following acute myocardial infarction
Office for National Statistics (ONS) prior to 2000 (ICD9) Category (code)
ICD9 code
ICD9 term
Death from myocardial infarction (3)
410
Acute myocardial infarction
Death from myocardial infarction (3)
4110
Other acute and subacute forms of ischemic heart disease ; Postmyocardial infarction syndrome
Death from myocardial infarction (3)
4297
Ill-defined descriptions and complications of heart disease ; Certain sequelae of myocardial infarction, not elsewhere classified
Myocardial Ischaemia National Audit Project (MINAP) 1 Acute STEMI 2 Acute NSTEMI
38
HEART FAILURE FOLLOWING MI: A CALIBER STUDY
CALIBER heart failure phenotype Read codes used to identify myocardial infarction in primary care records (Clinical Research Practice Datalink). Category (code)
Clinical code
Clinical term
Datasource lookup
Heart failure due to valvular disease (3) G580400
Congestive heart failure due to valvular disease
94870
Heart failure due to hypertension (4)
G210.00
Malignant hypertensive heart disease
50157
Heart failure due to hypertension (4)
G210000
Malignant hypertensive heart disease without CCF
95334
Heart failure due to hypertension (4)
G210100
Malignant hypertensive heart disease with CCF
72668
Heart failure due to hypertension (4)
G211100
Benign hypertensive heart disease with CCF
52127
Heart failure due to hypertension (4)
G21z100
Hypertensive heart disease NOS with CCF
62718
Heart failure due to hypertension (4)
G230.00
Malignant hypertensive heart and renal disease
67232
Heart failure due to hypertension (4)
G232.00
Hypertensive heart&renal dis wth (congestive) heart failure
21837
Heart failure due to hypertension (4)
G234.00
Hyperten heart&renal dis+both(congestv)heart and renal fail
57987
Heart failure due to other cause (5)
G1yz100
Rheumatic left ventricular failure
22262
Heart failure NOS (6)
1O1..00
Heart failure confirmed
9913
Heart failure NOS (6)
662W.00
Heart failure annual review
30779
Heart failure NOS (6)
662p.00
Heart failure 6 month review
83502 24503
Heart failure NOS (6)
8B29.00
Cardiac failure therapy
Heart failure NOS (6)
8H2S.00
Admit heart failure emergency
32898
Heart failure NOS (6)
9Or0.00
Heart failure review completed
19380
Heart failure NOS (6)
G400.00
Acute cor pulmonale
8464
Heart failure NOS (6)
G41z.11
Chronic cor pulmonale
5695 5141
Heart failure NOS (6)
G554000
Congestive cardiomyopathy
Heart failure NOS (6)
G554011
Congestive obstructive cardiomyopathy
68766
Heart failure NOS (6)
G58..00
Heart failure
2062
Heart failure NOS (6)
G58..11
Cardiac failure
1223
Heart failure NOS (6)
G580.00
Congestive heart failure
398
Heart failure NOS (6)
G580.11
Congestive cardiac failure
2906
Heart failure NOS (6)
G580.12
Right heart failure
10079
Heart failure NOS (6)
G580.13
Right ventricular failure
10154 9524
Heart failure NOS (6)
G580.14
Biventricular failure
Heart failure NOS (6)
G580000
Acute congestive heart failure
23707
Heart failure NOS (6)
G580100
Chronic congestive heart failure
32671 27884
Heart failure NOS (6)
G580200
Decompensated cardiac failure
Heart failure NOS (6)
G580300
Compensated cardiac failure
11424
Heart failure NOS (6)
G581.00
Left ventricular failure
884
Heart failure NOS (6)
G581.11
Asthma - cardiac
23481
Heart failure NOS (6)
G581.13
Impaired left ventricular function
5942 5255
Heart failure NOS (6)
G581000
Acute left ventricular failure
Heart failure NOS (6)
G582.00
Acute heart failure
27964
Heart failure NOS (6)
G58z.00
Heart failure NOS
4024 17278
Heart failure NOS (6)
G58z.12
Cardiac failure NOS
Heart failure NOS (6)
G5yy900
Left ventricular systolic dysfunction
8966
Heart failure NOS (6)
G5yyA00
Left ventricular diastolic dysfunction
12550
Heart failure NOS (6)
R2y1000
[D]Cardiorespiratory failure
20324
Left ventricular dysfunction (3)
585f.00
Echocardiogram shows left ventricular systolic dysfunction
11284
Left ventricular dysfunction (3)
585g.00
Echocardiogram shows left ventricular diastolic dysfunction
11351
2
39
PART ONE CHAPTER 2
Hospital Episode Statistics Category (code)
ICD10 code
ICD10 term
HF - hypertension cause (4)
I110
Hypertensive heart disease with (congestive) heart failure
HF - hypertension cause (4)
I130
Hypertensive heart and renal disease with (congestive) heart failure
HF - hypertension cause (4)
I132
Hypertensive heart and renal disease with both (congestive) heart failure and renal failure
HF - other unspecified cause (5)
I260
Pulmonary embolism with mention of acute cor pulmonale
HF - not otherwise specified (6)
I50
Heart failure
Office for National Statistics (ONS) Category (code)
ICD10 code
ICD10 term
Hypertensive heart failure (4)
I110
Hypertensive heart disease with (congestive) heart failure
Hypertensive heart failure (4)
I130
Hypertensive heart and renal disease with (congestive) heart failure
Hypertensive heart failure (4)
I132
Hypertensive heart and renal disease with both (congestive) heart failure and renal failure
Heart failure with other specified cause (5)
I260
Pulmonary embolism with mention of acute cor pulmonale
Heart failure with unspecified cause (6)
I50
Heart failure
Office for National Statistics (ONS) prior to 2000 (ICD9) Category (code)
ICD9 code
ICD9 term
Heart failure with other specified cause (5)
4151
Acute pulmonary heart disease ; Pulmonary embolism and infarction
Heart failure with unspecified cause (6)
428
Heart failure
Myocardial Ischaemia National Audit Project (MINAP) CCF 1
40
41
PART ONE
|
Clinical Course Portrayed
Chapter
3
Heart Failure Following ST-elevation Myocardial Infarction: an AGNES Cohort Study of Incidence and Prognostic Factors In preparation
Johannes M.I.H. Gho1, Pieter G. Postema2, Maartje Conijn1, Nienke Bruinsma2, Jonas S.S.G. de Jong2, Connie R. Bezzina3, Arthur A.M. Wilde2, Folkert W. Asselbergs1,4,5
1
Department of Cardiology, Division Heart and Lungs, University Medical Center Utrecht, Utrecht, the Netherlands
2
Department of Cardiology, Heart Center, Academic Medical Center, Amsterdam, The Netherlands
3
Department of Clinical and Experimental Cardiology, Academic Medical Center, Amsterdam, The Netherlands.
4
Durrer Center for Cardiogenetic Research, ICIN-Netherlands Heart Institute, Utrecht, The Netherlands
5
Institute of Cardiovascular Science, Faculty of Population Health Sciences, University College London, London, United Kingdom
PART ONE CHAPTER 3
ABSTRACT Background Aim of the current study is to determine the contemporary incidence, risk factors and prognosis of heart failure (HF) after ST-elevation myocardial infarction (STEMI). Methods We used the Arrhythmia Genetics in the NEtherlandS (AGNES) observational cohort study to identify patients with a first STEMI from 2001 onwards (n=1459). Heart failure during follow-up was defined as a hospitalisation for HF or an outpatient clinic visit for HF. Cox regression was performed to estimate the relationship between baseline covariates and the onset of HF. Results Follow-up was completed for 1360 patients with an overall median follow-up time of 6.7 years. A total of 85 patients (6.3%) developed HF during follow-up. In multivariable analysis, peak CKMB levels (HR 1.11 per 100U/L, 95%CI 1.11 â&#x20AC;&#x201C; 1.22) and an LAD culprit lesion (HR 2.82, 95%CI 1.50 â&#x20AC;&#x201C; 5.30) were risk factors associated with HF. Conclusion We found a relatively low long-term contemporary incidence of HF after a first STEMI in comparison to other reports.
44
HEART FAILURE FOLLOWING STEMI: AN AGNES STUDY
INTRODUCTION Heart failure (HF) is a major medical problem in the western world.1 HF is associated with substantial morbidity and mortality and contributes in a significant extent to national healthcare costs.2 The incidence of HF could be influenced by an increasing survival following an ischaemic event due to primary percutaneous coronary intervention (PCI) and because of demographic changes (ageing population).1 Data on the long-term incidence of HF after ST-elevation myocardial infarction (STEMI) is lacking and varies among different studies. In a recent nationwide Swedish
3
registry study (SWEDEHEART), HF incidence during index hospitalisation declined from 50% to 28% in the subgroup including STEMI patients between 1996 and 2008.3 In a Canadian cohort study 13.6% of STEMI patients were diagnosed with HF during index MI hospitalisation with a 1-year cumulative HF incidence of 23.4% between 2002 and 2008.4 In the same study, development of HF was associated with a 1-year mortality of 9.9% in STEMI patients.4 Here, we aim to determine the contemporary incidence and prognostic factors associated with HF after STEMI in the Netherlands percutaneous coronary intervention (PCI) era.
METHODS Study design The Arrhythmia Genetics in the NEtherlandS (AGNES) study is a Dutch multicenter observational study of which details have been published before.5-7 The AGNES cohort consisted of individuals with a first STEMI. Setting and participants This study complies with the Declaration of Helsinki. All patients or their legal representative gave written informed consent for inclusion and collection of information during follow-up was approved by the institutional ethics committee. Patients with first STEMI were recruited at seven heart centers in The Netherlands from 2001 onwards. Excluded were individuals with an actual nonâ&#x20AC;&#x201C;ST-elevation myocardial infarction, prior myocardial infarction or revascularisation prior to inclusion, congenital heart defects, known structural heart disease, severe comorbidity at discretion of the operator which would result in severely decreased prognosis, severe electrolyte disturbances, trauma at presentation, recent surgery, previous coronary artery bypass graft or use of class I and III antiarrhythmic drugs. The study cohort consisted of patients with a first STEMI. Patients who died within 30 days of the index infarct and patients with an incomplete follow-up were excluded from this analysis. Data sources Data on age, gender, cardiovascular risk factors, medication use before STEMI, infarct characteristics (e.g., type of reperfusion therapy, biomarkers and culprit lesion) were recorded at admission for index MI. Data on risk factors and family history in a subcohort were previously defined.7 Family history included sudden death and cardiovascular disease. A positive family history of sudden death was defined as a self-reported parent or sibling who died suddenly and
45
PART ONE CHAPTER 3
unexpectedly before the age of 80 years. We collected follow-up data from the hospital of index MI admission and the referral center. Collected endpoints included subsequent MI, stroke, HF, revascularization (CABG and PCI), ICD implantation, aborted cardiac arrest and all-cause mortality. In case of insufficient data we contacted general practitioners for additional information. The end points were verified by an adjudication committee (P.G.P. and F.W.A.). In this study we defined HF cases as patients with a hospital admission for HF or patients visiting the outpatient clinic for HF during follow-up. Patients without HF during follow-up served as control cases. Mortality data were verified using the Dutch Municipal Personal Records Database. Statistical methods Baseline patient characteristics, index MI characteristics and outcomes were described using descriptive statistics. Differences in continuous variables between cases with HF and controls without HF were tested using an independent t-test when data were normally distributed or otherwise using a Mann-Whitney test. Differences in categorical variables between cases and controls were determined using a chi-square test. We calculated cumulative incidence (cumulative percentage) of HF. Kaplan-Meier curves were generated to describe unadjusted cumulative HF incidence and all-cause mortality in STEMI patients using 5-year follow-up. Cox regression The associations of exposures of interest with the onset of HF were explored using Cox proportional hazard models. The primary outcome measure was patients admitted to the hospital with HF or visiting the outpatient clinic for HF. Cox regression with complete case analysis was performed to estimate the relationship between baseline prognostic factors and the development of heart failure after myocardial infarction and to determine hazard ratios (HR) with 95% confidence intervals (95%CI). The proportional hazards assumption was investigated by partial residual plots. In all analyses a p-value of <0.05 was considered as significant. Statistics were performed using IBM SPSS Statistics (Version 22, IBM Corporation, Armonk, New York, United States).
RESULTS In the AGNES cohort (n=1459), follow-up was incomplete for 74 patients due to insufficient available information (inability to retrieve follow-up information) and 25 patients died within 30 days of index MI (Figure 1). Follow-up was completed for 1360 patients. In total, 85 HF cases and 1275 controls without HF were included. Baseline and infarct characteristics Heart failure cases were significantly older at their index MI (59.9 vs. 57.2 years, p=0.001), more commonly had a positive family history (FH) of cardiovascular disease (CVD) (77.5% vs. 65.8%, p=0.032), a positive FH of sudden death (SD) (41% vs. 26.1%, p=0.006), a history of atrial fibrillation (6.1% vs. 1.4%, p=0.001), and were less likely to use statins prior to MI compared to controls (3.6% vs. 10.6%, p=0.041) (Table 1). In patients with HF median peak CK-MB levels at index STEMI were significantly higher (422.5 U/L, IQR 333.0) than in controls (181.0 U/L, IQR
46
HEART FAILURE FOLLOWING STEMI: AN AGNES STUDY
3
Figure 1. Flow chart
259.0) (p<0.001) (Table 2). Patients with HF were more likely to have suffered an anterior MI (79.8% vs. 48.9%, p<0.001) or LAD as culprit lesion (82.3% vs. 49.8%, p<0.001), whereas controls were more likely to have an RCX (3.8% vs. 15.9%, p=0.003) or RCA as culprit lesion (12.7% vs. 33.3%, p=0.001) at index STEMI. Other baseline characteristics were comparable between the two groups. There was no significant difference in ventricular fibrillation at index MI between cases (48.2%) and controls (41.4%). Heart failure incidence after MI and survival A total of 85 (6.3%) patients developed HF during an overall median follow-up time of 6.7 years (Table 3). Median follow-up was longer in HF cases than in controls (9.4 vs. 6.4 years, p<0.001). The median time until the onset of HF was 2.1 years. During follow-up, HF cases were more likely to have been subjected to revascularisation by additional PCI of the initial culprit artery (14.1% versus 7.4%, p = 0.025) or CABG (10.6% versus 5.3%, p = 0.042) than controls. We also found a higher rate of ICD implantation (44.7% versus 3.5%, p<0.001) and stroke (7.1% versus 2.6%, p = 0.017) in cases compared to controls. Overall 4.3% of the patients developed HF during 5-year follow-up (Figure 2). During follow-up 23.5% of HF cases versus 11.1% controls died (p=0.001). Overall 5-year mortality after a first STEMI was 6.7% (Figure 3). Risk factors associated with the onset of heart failure following index STEMI The proportional hazards assumption was violated and therefore the dataset was administratively censored at 10 years follow-up to ensure goodness-of-fit. A multivariable analysis (n=833) for
47
PART ONE CHAPTER 3
the outcome of HF was performed using age at index STEMI, gender, maximal measured Creatine Kinase (CK-MB) levels, the LAD culprit lesion and the calendar year of the index STEMI as predictors. Predictors associated with HF were peak CK-MB levels (HR 1.11 per 100U/L, 95%CI 1.11 – 1.22), a LAD culprit lesion (HR 2.82, 95% CI 1.50 – 5.30) and the year of index STEMI (HR 0.89 per year, 95%CI 0.83 – 0.96). Furthermore we found a trend towards an association with higher age at STEMI (HR 1.02, 95%CI 1.00 – 1.05). We did not find an association of gender with the outcome HF (male HR 0.77, 95%CI 0.40 – 1.48).
Table 1. Baseline characteristics With HF (n=85)
Without HF (n=1275)
Number of missings
p-value
Gender (female)
20 (23.5%)
261 (20.5%)
0
0.509
Mean age in years at index infarction (SD)
59.9 (10.3)
57.2 (10.7)
0
0.001
VF at index MI
41 (48.2%)
527 (41.4%)
0
0.212
Cardiovascular risk profile Mean BMI (SD)
26.6 (3.9)
26.6 (3.9)
61
0.702
FH of CVD
62 (77.5%)
817 (65.8%)
40
0.032
Current smoker
50 (50.5%)
767 (61.5%)
29
0.718
Diabetes
8 (9.6%)
94 (7.6%)
42
0.502
Hypertension
32 (39%)
378 (31.3%)
73
0.148
Atrial fibrillation
5 (6.1%)
17 (1.4%)
59
0.001
17 (22.7%)
378 (32.1)
110
0.087
FH of sudden death
High cholesterol
34 (41%)
330 (26.1%)
14
0.006
Angina 48h before STEMI
37 (45.1%)
455 (36.6%)
36
0.122
Cardiac medication before STEMI β-Blocker
13 (15.7%)
120 (9.6%)
25
0.073
Statins
3 (3.6%)
133 (10.6%)
25
0.041
Diuretics
5 (6.1%)
79 (6.3%)
24
0.943
ACE inhibitors/ARB
11 (13.3%)
98 (7.8%)
26
0.080
Aspirin/oral anticoagulation
9 (10.8%)
103 (8.2%)
23
0.401
Numbers in the columns with and without HF denote n= (%), unless specified otherwise. Abbreviations: ACE = angiotensin converting enzyme; ARB = Angiotensin receptor blocker; BMI = body mass index; FH = family history; HF = heart failure; MI = myocardial infarction; SD = standard deviation; STEMI = STelevation myocardial infarction; VF = ventricular fibrillation.
48
HEART FAILURE FOLLOWING STEMI: AN AGNES STUDY
Table 2. Characteristics of index STEMI With HF (n=85)
Without HF (n=1275)
Number of missings
p-value
73 (85.9%)
1159 (90.9%)
10
0.125
Reperfusion therapy PCI CABG
3 (3.5%)
19 (1.5%)
0.149
Thrombolysis
4 (4.7%)
34 (2.7%)
0.269
Non (spontaneous reperfusion on angiography or medication)
5 (5.9%)
53 (4.2%)
0.446
Time between symptoms and PCI (min) 180 (IQR 112)
180 (IQR 119)
0
0.670
Maximal Troponin T value (median) Îźg/L 6.26 (IQR 8.74)
2.10 (IQR 4.03)
1050
0.280
Maximal CK-MB value (median) U/L
422.5 (IQR 333.0)
181.0 (IQR 259.0)
243
<0.001
Location MI (anterior)
67 (79.8%)
617 (48.9%)
12
<0.001
65 (82.3%)
610 (49.8%)
Culprit lesion LAD
55 <0.001
RCX
3 (3.8%)
195 (15.9%)
0.003
RCA
10 (12.7%)
409 (33.3%)
0.001
1 (1.3%)
12 (1.0%)
30 (38.5%)
450 (36.9%)
LM Multivessel disease
3
0.828 62
0.780
Numbers in the columns with and without HF denote n= (%), unless specified otherwise. Abbreviations: CABG = coronary artery bypass grafting; HF = heart failure; LAD = left anterior descending artery; LM = left main artery; MI = myocardial infarction. PCI = percutaneous coronary intervention; RCA = right coronary artery; RCX = ramus circumflexus.
Table 3. Outcomes after STEMI Total (n=1360)
With HF (n=85)
Without HF (n=1275) p-value
6.7 (2.5-10.9)
9.4 (6.5-12.6)
6.4 (0.2-10.8)
<0.001
PCI culprit
106 (7.8%)
12 (14.1%)
94 (7.4%)
0.025
PCI non culprit
117 (8.6%)
10 (11.8%)
107 (8.4%)
0.283
CABG
Median follow-up in years (IQR) Revascularisations
77 (5.7%)
9 (10.6%)
68 (5.3%)
0.042
ICD
83 (6.1%)
38 (44.7%)
45 (3.5%)
<0.001
Second MI
130 (9.6%)
10 (11.8%)
120 (9.4%)
0.475
Stroke
39 (2.9%)
6 (7.1%)
33 (2.6%)
0.017
Deaths
162 (11.9%)
20 (23.5%)
143 (11.2%)
0.001
Numbers in the columns with and without HF denote n= (%), unless specified otherwise. Abbreviations: The p-value is derived from the comparison between HF cases and controls. CABG = coronary artery bypass grafting; HF = heart failure; ICD = implantable cardioverter-defibrillator; MI = myocardial infarction. PCI = percutaneous coronary intervention.
49
PART ONE CHAPTER 3
Figure 2. Kaplan-Meier curve showing the time until onset of heart failure in 5-year follow-up. The inset shows the portion of the graph with 100-90% free of heart failure. STEMI = ST-elevation myocardial infarction.
Figure 3. Kaplan-Meier curve showing the overall unadjusted 5-year survival after a first STEMI. The inset shows the portion of the graph with 100-90% cumulative survival. STEMI = ST-elevation myocardial infarction.
50
HEART FAILURE FOLLOWING STEMI: AN AGNES STUDY
DISCUSSION Using a patient cohort recruited in the Netherlands with a first STEMI we observed that 6.3% of patients developed HF during a median follow-up of 6.7 years. The median time to onset of HF was 2.1 years. The overall mortality was twice as high in the HF group versus the control group after index MI. Higher CK-MB levels and an LAD culprit lesion at index STEMI were important risk factors for the outcome of HF.
3
Incidence Studies researching the contemporary long-term HF incidence in STEMI subgroups report an incidence between 23 and 28%.3, 4 When comparing different studies, variability in reperfusion therapy changing over the years and HF definition should be taken into account. In the current study overall 90.6% of patients underwent primary PCI and 2.8% thrombolysis, which resembles a unique contemporary STEMI cohort and comparable population studies undergoing similar reperfusion therapy are lacking. Of the SWEDEHEART study patients with STEMI or left bundle branch block admitted in 2002-2003, 72.2% underwent thrombolysis and 2.78% primary PCI compared to 8.5% thrombolysis and 91.5% primary PCI in 2008.3 In the Canadian study subgroup of STEMI patients, PCI was performed in 46.8% of patients with HF during index hospitalisation, 52.5% of patients who developed HF after discharge and 60.2% in those who did not develop HF.4 Other studies were not limited to STEMI, but also included patients with non-ST-elevation infarction (NSTEMI) and were predominantly from the thrombolytic era. In the Framingham Heart Study, Velagaleti et al. found that 14.8% (21/142) of patients surviving 30 days after index MI developed congestive HF during 5-year follow-up from 1990 to 1999.8 Data of Torabi et al. showed a HF incidence of 33% after discharge from hospital admission for index MI.9 In a study by Najafi et al. an 22.4% developed HF within 28 days of index admission (after exclusion of patients who died within 28 days) and from these patients 12.4% had at least one subsequent admission with HF after 10 years follow-up.10 Epidemiologic research in Olmsted County (Minnesota, USA) found a HF incidence of 41% during a median follow-up of 6.6 years after MI.11 A more recent Danish study found a 90-day HF incidence of 19.6% in 2009-2010.12 Prognostic factors Infarct size and anterior MI are well known risk factors for adverse prognosis in STEMI patients.13 In this study, maximal CK-MB value as a proxy for infarct size was an adverse prognostic factor for the development of HF. Cardiac biomarkers, including peak CK-MB significantly correlated in previous research with infarct size and LVEF.14 We also found a LAD culprit lesion to be a risk factor for development of HF after STEMI (HR 2.82). Our study demonstrated a negative association between the year in which the index event occurred and the development of HF. This is in accordance to other studies where decreasing trends in the development of HF after MI are reported.11, 12, 15, 16 We also found a trend for higher age at the time of the index MI associated with the development of HF after a first STEMI. This was expected as age is a well-known risk factor for HF development.3, 10, 11, 15 For example, Desta et al. found an odds ratio of 1.05 (95%CI 1.05 â&#x20AC;&#x201C; 1.06) for every additional year increase in age.3 A previous study found a significant association of gender with HF incidence after MI, but we could not replicate this finding in our STEMI cohort.15
51
PART ONE CHAPTER 3
Survival We found that about 1 in 4 HF patients died during a median follow-up of 6.7 years. This is consistent with results reported in other studies in which the poor prognosis after onset of HF is described.10, 11, 16, 17 In the SWEDEHEART study, Desta et al. found a decrease in 1-year mortality for patients with clinical HF following MI from 36% to 31% between 1996 and 2008.3 Kaul et al. found a 1-year mortality rate of 10.6% in patients with HF and 2.4% mortality in patients without HF after STEMI.4 Limitations Cardiovascular endpoints might have been missed as follow-up was performed in the hospital of the index admission, for example (predominantly milder) HF cases managed by the general practitioner or HF cases managed in other hospitals. In this study we defined HF as a hospitalisation or outpatient clinic visit for HF, which is of relevance to external validity. Other studies also used administration of diuretics or symptoms of HF to define HF.3,
9, 12
The study did not systematically include consecutive STEMI patients, patients with primary VF were selectively included and patients without primary VF were random included but the cohort represents a contemporary STEMI population.
CONCLUSION This research demonstrates a low contemporary incidence of HF after a first PCI-treated STEMI in the Netherlands in comparison to other reports. Higher CK-MB levels and a LAD culprit lesion at index STEMI were important risk factors for the development of HF after STEMI.
52
HEART FAILURE FOLLOWING STEMI: AN AGNES STUDY
REFERENCES 1. 2. 3.
4. 5.
6.
7.
8.
9.
10. 11. 12.
13. 14.
15.
16.
17.
Mosterd A, Hoes AW. Clinical epidemiology of heart failure. Heart. 2007;93:1137-46. McMurray JJ, Stewart S. Epidemiology, aetiology, and prognosis of heart failure. Heart. 2000;83:596-602. Desta L, Jernberg T, Lofman I, Hofman-Bang C, Hagerman I, Spaak J, et al. Incidence, Temporal Trends, and Prognostic Impact of Heart Failure Complicating Acute Myocardial Infarction: The SWEDEHEART Registry (Swedish Web-System for Enhancement and Development of Evidence-Based Care in Heart Disease Evaluated According to Recommended Therapies): A Study of 199,851 Patients Admitted With Index Acute Myocardial Infarctions, 1996 to 2008. JACC Heart Fail. 2015;3:234-42. Kaul P, Ezekowitz JA, Armstrong PW, Leung BK, Savu A, Welsh RC, et al. Incidence of heart failure and mortality after acute coronary syndromes. Am Heart J. 2013;165:379-85 e2. Bezzina CR, Pazoki R, Bardai A, Marsman RF, de Jong JS, Blom MT, et al. Genome-wide association study identifies a susceptibility locus at 21q21 for ventricular fibrillation in acute myocardial infarction. Nat Genet. 2010;42:688-91. de Jong JS, Marsman RF, Henriques JP, Koch KT, de Winter RJ, Tanck MW, et al. Prognosis among survivors of primary ventricular fibrillation in the percutaneous coronary intervention era. Am Heart J. 2009;158:467-72. Dekker LR, Bezzina CR, Henriques JP, Tanck MW, Koch KT, Alings MW, et al. Familial sudden death is an important risk factor for primary ventricular fibrillation: a case-control study in acute myocardial infarction patients. Circulation. 2006;114:1140-5. Velagaleti RS, Pencina MJ, Murabito JM, Wang TJ, Parikh NI, Dâ&#x20AC;&#x2122;Agostino RB, et al. Long-term trends in the incidence of heart failure after myocardial infarction. Circulation. 2008;118:205762. Torabi A, Cleland JG, Khan NK, Loh PH, Clark AL, Alamgir F, et al. The timing of development and subsequent clinical course of heart failure after a myocardial infarction. Eur Heart J. 2008;29:859-70. Najafi F, Dobson AJ, Hobbs M, Jamrozik K. Late-onset heart failure after myocardial infarction: trends in incidence and survival. Eur J Heart Fail. 2008;10:765-71. Hellermann JP, Jacobsen SJ, Redfield MM, Reeder GS, Weston SA, Roger VL. Heart failure after myocardial infarction: clinical presentation and survival. Eur J Heart Fail. 2005;7:119-25. Gjesing A, Gislason GH, Kober L, Gustav Smith J, Christensen SB, Gustafsson F, et al. Nationwide trends in development of heart failure and mortality after first-time myocardial infarction 1997-2010: A Danish cohort study. Eur J Intern Med. 2014;25:731-8. Sutton MG, Sharpe N. Left ventricular remodeling after myocardial infarction: pathophysiology and therapy. Circulation. 2000;101:2981-8. Chia S, Senatore F, Raffel OC, Lee H, Wackers FJ, Jang IK. Utility of cardiac biomarkers in predicting infarct size, left ventricular function, and clinical outcome after primary percutaneous coronary intervention for ST-segment elevation myocardial infarction. JACC Cardiovasc Interv. 2008;1:415-23. Shafazand M, Rosengren A, Lappas G, Swedberg K, Schaufelberger M. Decreasing trends in the incidence of heart failure after acute myocardial infarction from 1993-2004: a study of 175,216 patients with a first acute myocardial infarction in Sweden. Eur J Heart Fail. 2011;13:135-41. Chen J, Hsieh AF, Dharmarajan K, Masoudi FA, Krumholz HM. National trends in heart failure hospitalization after acute myocardial infarction for Medicare beneficiaries: 1998-2010. Circulation. 2013;128:2577-84. Spencer FA, Meyer TE, Goldberg RJ, Yarzebski J, Hatton M, Lessard D, et al. Twenty year trends (1975-1995) in the incidence, in-hospital and long-term death rates associated with heart failure complicating acute myocardial infarction: a community-wide perspective. J Am Coll Cardiol. 1999;34:1378-87.
3
53
PART TWO
|
Finding Fibrosis Patterns
Chapter
4
A Systematic Comparison of Cardiovascular Magnetic Resonance and High Resolution Histological Fibrosis Quantification in a Chronic Porcine Infarct Model Submitted
Johannes M.I.H. Gho1*, RenĂŠ van Es1*, Frebus J. van Slochteren2, Sanne J. Jansen of Lorkeers1, Allard J. Hauer1, Joep W.M. van Oorschot3, Pieter A. Doevendans1, Tim Leiner3, Aryan Vink4, Folkert W. Asselbergs1,5,6, Steven A.J. Chamuleau1 *Contributed equally to this article and are shared first authors
1
Department of Cardiology, Division Heart and Lungs, University Medical Center Utrecht, Utrecht, the Netherlands
2
Interuniversity Cardiology Institute of the Netherlands, Utrecht, the Netherlands
3
Department of Radiology, University Medical Center Utrecht, Utrecht, the Netherlands
4
Department of Pathology, University Medical Center Utrecht, Utrecht, the Netherlands
5
Durrer Center for Cardiogenetic Research, ICIN-Netherlands Heart Institute, Utrecht, the Netherlands
6
Institute of Cardiovascular Science, Faculty of Population Health Sciences, University College London, London, United Kingdom
PART TWO CHAPTER 4
ABSTRACT Background The noninvasive reference standard for myocardial fibrosis detection on cardiovascular magnetic resonance imaging (CMR) is late gadolinium enhancement (LGE). Currently there is no consensus on the preferred method for LGE quantification. Moreover myocardial wall thickening (WT) and strain are measures of regional deformation and can provide insight into local mechanical effects of myocardial fibrosis. In this article we present a high resolution systematic digital histological quantification of cardiac fibrosis and provide a detailed comparison to various in vivo CMR imaging techniques. Aim of this research was to systematically compare in vivo CMR parameters, such as LGE, WT and strain, with histological fibrosis quantification. Methods Eight weeks after 90 minutes ischemia/reperfusion of the LAD artery, 16 pigs (79.8Âą5.8 kg) underwent in vivo CMR on a clinical 3T scanner. Cine imaging and LGE sequences were performed. Histological sections from transverse heart slices were digitally analysed for fibrosis quantification. Mean percentage of fibrosis of each digitally analysed section was related to the different CMR techniques (using Segment or feature tracking software) for that slice using a linear mixed model analysis. For each CMR derived variable, R2 values were calculated as a measure of the percentage of variability in myocardial fibrosis that can be explained by a linear association with the CMR parameters. Results The full width at half maximum (FWHM) technique for quantification of LGE yielded the highest R2 of 60%. Cine derived myocardial WT explained 16 to 36% of the histological myocardial fibrosis. The peak circumferential and radial strain measured by feature tracking could explain 15% and 10% of the variance of myocardial fibrosis, respectively. Conclusions This novel method to systematically compare CMR image data with digital histological images is feasible. Myocardial WT and strain were only modestly related with the amount of fibrosis. The fully automatic FWHM analysis technique is the preferred method to detect myocardial fibrosis.
56
COMPARISON OF CMR AND HISTOLOGICAL FIBROSIS
BACKGROUND Myocardial fibrosis has been associated with heart failure and can act as a substrate for cardiac arrhythmias.1 Following myocardial infarction (MI), loss of cardiomyocytes leads to reparative fibrosis with replacement by connective tissue. Noninvasive assessment of cardiac fibrosis is important for diagnosis, predicting prognosis and treatment planning.2 The noninvasive reference standard for fibrosis detection is late gadolinium enhancement (LGE) on cardiovascular magnetic resonance imaging (CMR). Since there is no consensus on the LGE quantification techniques,3 a detailed comparison with the reference standard of histological analysis is important. For example, accurate fibrosis quantification can predict reversible myocardial dysfunction after revascularization.4, 5 Thus far, mainly correlation studies have been performed with small endomyocardial biopsies, triphenyl tetrazolium chloride stained or ex vivo hearts.6-8 Studies using
4
whole heart slices are scarce,9 especially for focal fibrosis. While LGE provides an accurate qualitative measure of fibrosis, it requires contrast administration with potential adverse effects and does not provide a quantitative or direct measurement of cardiac collagen.2, 10 The result of LGE differs between different imaging studies and by variable intensity threshold settings and thus relies on an adequate imaging protocol. Functional assessment of the local myocardium is typically performed visually on cine CMR images. Quantitative assessment of local myocardial function (e.g., wall thickening) can also be performed with (semi-)automatic segmentation software packages.11, 12 More recently, feature tracking (FT) has been introduced as a method to assess local myocardial deformation (strain) using cine images without the need for tagged CMR scans.13 Feature tracking is relatively quick in post processing, has shown reasonable agreement with tagging CMR when looking at global strain from complete slices and might be usable between different field strengths.14-16 However, there is debate about the agreement between strain derived from FT and tagging CMR at a segmental level, as multiple studies found poor intra- and interobserver variability for segmental strain.14, 17-19 We recently developed a method for high resolution systematic digital histological quantification of (diffuse and focal) cardiac fibrosis in a whole heart slice,20 which can provide a detailed reference for comparing different CMR imaging techniques. The aim of this study was to systematically analyse in vivo CMR derived parameters and high spatial resolution digital fibrosis quantification in a chronic porcine infarct model to compare different CMR techniques with myocardial fibrosis assessment. Parameters of interest are: LGE CMR, myocardial strain and wall thickening (WT). Myocardial strain and myocardial WT are respectively assessed by FT and manual segmentation on cine MRI.
METHODS Animal model All in vivo experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals prepared by the Institute of Laboratory Animal Resources. Experiments were approved (protocol no.: 2012.II.09.145) by the local Animal Experiments Committee (DEC) (Utrecht, the Netherlands).
57
PART TWO CHAPTER 4
Our protocol regarding a porcine chronic MI model has been described in detail before.21 Eight weeks after 90 minutes ischemia/reperfusion of the proximal left anterior descending artery (LAD), 16 Dalland Landrace pigs (79.8¹5.8 kg) under continuous anesthesia underwent in vivo CMR on a clinical 3T scanner (Achieva TX, Software Release 3.2.1, Philips Healthcare, Best, the Netherlands). CMR Pigs were positioned supine with a dedicated 32-channel phased-array receiver coil over the chest and scanned using a standardized protocol. In short, for image planning scout images were obtained in short-axis and two-chamber long-axis views. ECG-gated steady-state free precession (SSFP) cine of short-axis (from apex to base of LV) and two chamber long-axis views were acquired. Thirty frames were acquired per RR cycle. Cine parameters: echo time (TE)/ repetition time (TR) 1.6/3.2ms, 13 slices, slice thickness 8 mm, resolution = 2x2mm, field of view (FOV) = 320x320mm2, bandwidth = 1200Hz and flip angle = 45°.
Figure 1. Cardiovascular Magnetic Resonance Analysis Methods A-C. The endo- and epicardial segmentations of the wall thickening (Segment), feature tracking (Image-Arena) and LGE (Segment) analyses respectively. The asterisk (*) indicates the point in the mid-lateral wall that was used for registration between CMR and histology. The automatic FWHM infarct delineation is shown in C (yellow). D. Resulting WT patterns of the four segments shown in A. The actual WT analysis used 60 sections. E. Radial strain analysis of 48 segments (Image-Arena); the result of the segment indicated with the asterisk in B is shown in red. F. The %TM data was exported in 360 segments. The digitized annotation map of the LV, as shown in Figure 2B, is then projected onto the LGE exported data for analysis. FWHM = full width at half maximum; LGE = late gadolinium enhancement; LV = left ventricle; %TM = fraction of transmurality; WT = wall thickening.
58
COMPARISON OF CMR AND HISTOLOGICAL FIBROSIS
LGE Late gadolinium enhancement CMR was performed using an inversion recovery 3D-turbogradient-echo-technique 15 minutes after double-dose intravenous bolus injection of a gadolinium based contrast agent (Gadovist, Bayer Healthcare, Berlin, Germany). First, a look-locker scout was performed for the optimal inversion time. Acquisition parameters for the LGE scan: inversion time (TI) = 200-270ms, TE/TR = 1.5/4.7ms, slice thickness = 6mm, spatial resolution = 1.5x1.5mm2, FOV = 300x300mm2, flip angle = 25°, 63 TFE shots, bandwidth = 300Hz, number of signals averaged = 2, SENSE acceleration = 2. CMR imaging analysis Segment Offline image analysis to derive WT and LGE was performed using Segment version v1.9 R3590
4
(http://segment.heiberg.se, Medviso AB, Lund, Sweden).12 In all datasets, the short-axis image corresponding to the histological slice was selected based on its location, and used for further analysis (Figure 1). In the short-axis cine images, LV endo- and epicardial borders were manually segmented in all frames. The segmentations were copied to the corresponding LGE slice (Figure 1C). Manual adjustment was performed if necessary. From the short-axis cine dataset the absolute WT (mm) per image frame of 60 LV segments was exported using the reportSlice function of Segment. The end systolic absolute WT of each segment was used for further analysis. Feature Tracking Cardiac strain analysis by FT was performed using the Image-Arena 2D Cardiac Performance Analysis software toolbox version 1.2 (TomTec Imaging Systems, Unterschleissheim, Germany). The end-diastolic endo- and epicardial contours of the slice were manually traced (Figure 1B). The mid-lateral point along the endocardial contour was marked as an anatomical reference and data was exported for registration purposes. For all 48 segments, raw data containing circumferential strain (εcc), radial strain (εrr) and WT (endo to epi distance), was exported for further analysis (Figure 1E). Viability analysis Segment For viability analysis of short-axis LGE datasets, scar was delineated using automatic full width at half maximum (FWHM) (Figure 1C), automatic uncorrected and manually corrected standard deviation (SD) from remote (2, 3 and 5SD) algorithms. For the SD methods, the remote healthy myocardium of the lateral wall was selected as remote. Manual corrections of the infarct area were performed in the subgroups. The fraction of area based transmurality (%TM), the mean infarct size fraction of the wall thickness, was analysed in 360 equal segments over the LV wall. For myocardial signal intensity (MSI) and each scar delineation method the %TM was exported using Segment’s reportBullsEye function. Histology Following CMR, the animals were sacrificed by exsanguination under general anaesthesia and the hearts were excised and cut into transverse (short-axis) 1cm thick slices beginning from
59
PART TWO CHAPTER 4
apex to base. Each third transverse slice was fixed in formalin, cut into smaller sections and a map of the heart slice was drawn to annotate the origin of each tissue specimen (Figure 2A). These sections were embedded in paraffin and stained with Massonâ&#x20AC;&#x2122;s trichrome. The slides were scanned at 20X magnification as described before.22 Images were extracted using Aperio ImageScope v.12.0.0.5039 (Aperio, Vista, CA, USA) and resized to 10% for digital analysis. Histological analysis Digital histological analysis was performed systematically as previously described, using the in house developed software package Fibroquant (http://sourceforge.net/projects/fibroquant).20 The epicardium, defined as the outer region of fatty tissue bordered by the first row of cardiomyocytes, was excluded from further analysis. The remaining myocardium, including the compact and trabeculated region was analysed as a whole. The percentage of connective tissue (blue), cardiomyocytes (red) and adipose tissue (cells with non-stained cytoplasm) was digitally quantified using Fibroquant. The results were annotated to their corresponding heart region using an automated algorithm and transformed to a standardized schematic overview (Figure 2B). Matching MRI with histology After the construction of the standardized schematic overview, the exported high detail MRI data was averaged over the regions delineated by the histological sections (e.g., Figure 1F). A landmark in the mid-lateral wall of all datasets was used as a reference point to assure optimal registration. This reference was defined as the point opposite both hinge points of the right ventricle. Statistical analysis Statistics were performed using IBM SPSS Statistics (Version 20.0, IBM Corporation, Armonk, New York, United States). We compared different CMR techniques with percentages of fibrosis per section using a linear mixed model analysis. For fibrosis, the amount of residual variance (Ď&#x192;2) within the animals and the variance (intercept, Ď&#x201E;) between animals were calculated
Figure 2. Digital Histological Analysis Methods A. Map containing the origin of the histological sections. B. The digitized left ventricular annotation map of image A. C. The manual tracings of the four regions across the histological myocardial section, corresponding to location 5 in A and B.
60
COMPARISON OF CMR AND HISTOLOGICAL FIBROSIS
(null model). Subsequently, CMR parameters were added to the model (full model). R2 (Snijders & Bosker, 2012), was calculated as 100% minus the ratio of the full and null models (equation 1),23, 24 representing the explained variance. (eq. 1)
4
Figure 3. Results of Histological Analysis A-B. Results of the histological analyses of sections 5 and 7 respectively (Figure 2A&B). C. Analysis of the whole heart of this animal. D. Mean fibrosis content in the left ventricle of all animals (n=15). LV = left ventricle; RV = right ventricle.
RESULTS Histology A total of 116 histological sections (16 animals) were successfully stained, segmented (Figure 2C) and analysed (Figure 3A&B). Digital maps with the annotated origins of each histological section were used to construct schematic histological overviews for each animal (e.g., Figure 3C). One animal (8 sections) was excluded from further analysis because of lack of histological
61
PART TWO CHAPTER 4
fibrosis due to a sampling error (Supplementary Figure 1), 108 sections remained. Mean fibrosis percentages (15 animals) were log transformed to reduce right-skewness and heterogeneity of variance. Myocardial fibrosis was mainly observed in the anteroseptal wall (Figure 3D), corresponding to the LAD territory. The 8-week old LAD infarct model used in this study yielded fractional fibrosis values of maximally 57%. Cine CMR analysis Segment Myocardial WT curves were successfully extracted for 15 animals in 60 segments. After matching the CMR segments according to the histological sections, mean end systolic absolute WT was 2.4±2.4mm (range -2.1 to 7.2mm) (Figure 4A, 5A). The explained variance of WT for histological myocardial fibrosis was 36% (Table 1). Feature tracking Data of 15 animals was successfully extracted and analysed in 48 segments. After matching the data according to histological sections the mean εcc, εrr and WT were -12.1±12.1% (range -38.5 to 24.3%), 26.9±34.7% (range -26.0 to 171.9%) and -1.6±2.5mm (range -4.6 to 7.7mm) respectively (Figure 4B, 5B-D). Feature tracking derived εcc, εrr and WT explained 15%, 10% and 16% of myocardial fibrosis respectively (Table 1).
Table 1. Comparison of Wall Thickening, Strain and LGE with Histological Fibrosis n = 15
σ2full
τ00full
R2
Cine WT (Segment)
1.07
0.02
36%
Cine WT (Image-Arena)
1.44
0*
16%
Radial Strain
1.54
0*
10%
Circumferential Strain
1.45
0*
15%
2SD
0.94
0.07
41%
3SD
0.82
0.11
46%
5SD
0.65
0.12
55%
2SD Manual Subgroup
0.66
0.03
60%
3SD Manual Subgroup
0.63
0.05
60%
5SD Manual Subgroup
0.65
0.07
58%
FWHM
0.62
0.07
60%
Myocardial Signal Intensity
0.68
0.14
52%
The null model used in the statistical analysis containing the variance in myocardial fibrosis was expressed as σ2null+ τ00null=1.71*. The full model containing the unexplained variance in myocardial fibrosis after adding explanatory CMR parameters is expressed as σ2full+ τfull. The resulting R2 value was calculated as 100% - the fraction of unexplained variance (eq. 1). *= Although all convergence criteria were satisfied, the hessian matrix was not positive definite, as a consequence the residual and random intercept variance were pooled. FWHM = full width at half maximum; LGE = late gadolinium enhancement; R2 = explained variance of myocardial fibrosis; SD = standard deviation; σ2 = intra animal variance; τ = inter animal variance; WT = wall thickening.
62
COMPARISON OF CMR AND HISTOLOGICAL FIBROSIS
4
Figure 4. Comparisons of Cardiovascular Magnetic Resonance Parameters and Fibrosis (n = 15) The dots represent 108 separate sections from 15 animals. R2 values (explained variance) are derived from a separate linear mixed model analysis. Fibrosis was compared with: A. Wall thickening results of the cine analysis (Segment); B. Circumferential strain analysis (Image-Arena); C. Myocardial signal intensity analysis (Segment); D. LGE FWHM analysis (Segment). FWHM = full width at half maximum; LGE = late gadolinium enhancement.
Viability CMR analysis Segment The mean LGE scar transmurality and MSI (15 animals) is shown in Figure 6. The variance between animals was small (Ď&#x201E; range 0 to 0.14) (Table 1). The automatic FWHM algorithm resulted in 60% explained variance for histological myocardial fibrosis. Manually corrected SD from remote methods yielded an explained variance between 58 - 60%. The variance within animals (Ď&#x192;2) decreased (0.94 to 0.65) with more stringent automatic segmentation methods (2-5 SD), while it remained similar in the manually edited SD subgroup. The MSI of the LGE images, in which no additional interaction is required, explained the myocardial fibrosis for 52%. Scar transmurality compared to fibrosis and functional parameters In the boxplots (Figure 7), the median fibrosis ranged from 1.4% in the segments without LGE to 22.7% in the segments with 75-100% LGE transmurality using FWHM. Median WT ranged
63
PART TWO CHAPTER 4
from 4.3 to 0.0mm and εcc and εrr, ranged from -15.0% to -7.7% and 40.0% to -1.0% respectively between the different increasing amounts of LGE transmurality.
Figure 5. Mean Wall Thickening and Strain Analysis (n = 15) A. The averaged results of the cine WT analysis (Segment). B-D. The averaged results of the feature tracking analysis, the WT, radial (εrr) and circumferential (εcc) strain respectively (Image-Arena). εcc = circumferential strain; εrr = radial strain; WT = wall thickening.
Figure 6. Mean Late Gadolinium Enhancement Analysis (n = 15) Averaged results of the fraction of transmurality in 360 segments for the 2, 3 and 5SD automatic methods, the manually corrected subgroup and the fully automatic FWHM analysis. In the bottom right corner, the averaged myocardial signal intensity is shown. FWHM = full width at half maximum; SD = standard deviation.
64
COMPARISON OF CMR AND HISTOLOGICAL FIBROSIS
4
Figure 7. Boxplots of Late Gadolinium Enhancement Scar Transmurality and Fibrosis or Functional Parameters (n = 15) A-D. The dots represent outliers. Late gadolinium enhancement fraction of transmurality based on FWHM divided in different subgroups (0, 0-25, 26-50, 51-75 and 76-100%) compared to fibrosis percentage, wall thickening from feature tracking, and radial or circumferential strain from FT. FT = feature tracking; FWHM = full width at half maximum; WT = wall thickening.
DISCUSSION This study demonstrates a novel systematic method to compare different in vivo CMR techniques with high spatial resolution histological analyses in a chronic porcine infarct model. To the best of our knowledge, this is the first time such a systematic comparison has been performed and this method can be readily adapted for use with other imaging modalities such as SPECT, CT, PET and echocardiography. For LGE imaging, we found that the FWHM and the 2, 3 or 5SD methods with manual correction were the best methods to quantify the amount of myocardial fibrosis. Although it is expected that measures of myocardial deformation would be affected in infarcted myocardium, myocardial WT and strain were only modestly related with the amount of myocardial fibrosis.
65
PART TWO CHAPTER 4
Myocardial WT The reference standard for the noninvasive assessment of cardiac anatomy and function is CMR.2 While cine imaging is not intended for fibrosis quantification, MI is associated with a lower amount of WT in the infarct area.6 We measured myocardial WT using two different software analysis tools, Segment and Image-Arena. Our analysis shows that WT derived from Segment performs better in explaining myocardial fibrosis compared to Image-Arena (R2 = 36% vs. 16%). In Segment, both the endo- and epicardial contours of the selected slice were manually traced in every frame. In Image-Arena the segmentation was performed in only one frame and the tracing in subsequent frames were calculated automatically by the software algorithm. The FT workflow could limit accuracy and cause variability, due to the lack to perform specific adjustments in multiple frames which is inherent to the algorithm. Furthermore, through plane motion and filtering algorithms resulting in loss of detail are likely to introduce erroneously tracked features, thereby affecting the FT results. Cardiac deformation In our analysis of the complete LV slices, both radial and circumferential strain based on FT were poorly associated with myocardial fibrosis (R2 = 10% and 15% respectively). As previously described by Cowan et al., differences in strain could occur between two regions indicated as healthy by LGE.19 From our study it can be concluded that FT based strain imaging is less applicable to identify regional myocardial fibrosis in this experimental infarct model. This might be caused by the fact that strain is not a strictly local phenomenon, but is the resulting deformation caused by the contraction of cardiomyocytes in the vicinity of the infarct. Although local myocardial deformation assessment is not a direct measurement of fibrosis, strain analysis might supply new insights in local cardiac biomechanics after ischemic injury, but rigorous clinical validation is required. Viability Compared to the reference standard of fibrosis on histology, FWHM and the 2, 3 or 5SD methods with manual correction were the best explanatory variables for variance in fibrosis. The FWHM technique uses half of the maximal signal intensity within the scar region as a threshold to determine the infarct area.25, 26 The used SD from remote methods delineate scar by pixels with an image intensity higher than the mean plus 2, 3 or 5SD from the mean in a non-infarcted remote region.6 Late gadolinium enhancement has been previously validated ex vivo using a MSI analysis compared to quantitative histopathological sections with fibrosis to discriminate between healthy myocardium, dense scar and the border zone.8 A previous clinical study showed limited reproducibility of the SD from remote techniques with manual correction and found the FWHM technique the most reproducible method.3 In our analysis, the MSI measured from LGE images resulted in worse explanatory values for myocardial fibrosis compared to FWHM (52% vs. 60%). The between animal variance of LGE MSI compared to FWHM was higher (Ď&#x201E; 0.14 vs. 0.07) which could be explained by MSI variations due to small variations in gadolinium injection time to acquisition. Because of the improved explanation for myocardial fibrosis and reduced manual interaction needed, FWHM would be the preferred method of choice from this study for infarct quantification on LGE CMR.
66
COMPARISON OF CMR AND HISTOLOGICAL FIBROSIS
In diffuse fibrosis, with less distinct LGE or remote areas, the FWHM method could be technically more difficult and T1-mapping might be more promising.9 Limitations The final resolution of the comparison between histology and MRI is determined by the method of histological sections cut from the LV, typically 7 or 8 sections per heart slice in this study. Only one short-axis slice of the CMR images was used and the high detail MRI data was averaged to match exactly with the histology data, leading to a loss of detail, especially along the infarct border zone. For future studies dissection of the entire heart using a microtome could be considered to preserve the gross cardiac anatomy. Resulting high resolution images could later be analyzed and subdivided using automated algorithms to allow for a more detailed comparison with high detail imaging modalities.
4
In the assessment of local cardiac function (e.g., WT, Îľcc and Îľrr) it is likely that a higher fibrosis percentage would have led to a stronger relation with worsening of functional parameters. The right ventricle was excluded from all comparisons because the conventional analyses applied on the cine and LGE images in this study did not allow an accurate analysis of the right ventricle. Future implications The comparison method used in this study could be applied on data from other imaging modalities such as SPECT, PET, CT, echocardiography and other CMR sequences (e.g., T1mapping) and can also be translated for use with a 3D model. While the proximal LAD ischemia/ reperfusion model used in this study produced consequent isolated anteroseptal infarctions, it would be in the interest of external validity to study other infarct sizes and locations using this method. Improved fibrosis detection with CMR will be applicable to a broad clinical spectrum ranging from diagnostic to therapeutic outcomes, including cell therapy, ablation therapy, valvular diseases, cardiomyopathies and ischemic heart disease. For example, precise identification of the fibrotic region allows for accurate therapy guidance to the target area with the aim to ultimately improve clinical outcome for patients.
CONCLUSIONS In conclusion, the novel systematic method to compare high resolution in vivo CMR imaging with detailed histological fibrosis data was feasible and can readily be applied to other imaging data. Locally measured functional parameters such as WT, and measures of myocardial deformation derived from FT: radial and circumferential strain related modestly with local myocardial fibrosis, yet can be used to gain insight into local cardiac mechanics. The fully automatic FWHM algorithm applied on the reference standard LGE CMR showed to be preferred to detect myocardial fibrosis in a chronic in vivo infarct model.
67
PART TWO CHAPTER 4
Acknowledgments We kindly acknowledge Grace Croft, Marlijn Jansen, Evelyn Velema, Joyce Visser, Merel Schurink, Martijn van Nieuwburg and Gerard Marchal for their assistance with the animal experiments. We would like to thank Jaco Zwanenburg and Martijn Froeling for performing the CMR experiments. We are also grateful to Rebecca Stellato for statistical advice.
68
COMPARISON OF CMR AND HISTOLOGICAL FIBROSIS
REFERENCES 1. 2. 3.
4.
5.
6.
7.
8.
9.
10. 11. 12.
13.
14.
15.
16.
17.
18.
Weber KT, Sun Y, Bhattacharya SK, Ahokas RA, Gerling IC. Myofibroblast-mediated mechanisms of pathological remodelling of the heart. Nat Rev Cardiol. 2013;10:15-26. Mewton N, Liu CY, Croisille P, Bluemke D, Lima JA. Assessment of myocardial fibrosis with cardiovascular magnetic resonance. J Am Coll Cardiol. 2011;57:891-903. Flett AS, Hasleton J, Cook C, Hausenloy D, Quarta G, Ariti C, et al. Evaluation of techniques for the quantification of myocardial scar of differing etiology using cardiac magnetic resonance. JACC Cardiovasc Imaging. 2011;4:150-6. Kim RJ, Wu E, Rafael A, Chen EL, Parker MA, Simonetti O, et al. The use of contrast-enhanced magnetic resonance imaging to identify reversible myocardial dysfunction. N Engl J Med. 2000;343:1445-53. Bondarenko O, Beek AM, McCann GP, van Rossum AC. Revascularization in patients with chronic ischaemic myocardial dysfunction: insights from cardiovascular magnetic resonance imaging. Eur Heart J Cardiovasc Imaging. 2012;13:985-90. Kim RJ, Fieno DS, Parrish TB, Harris K, Chen EL, Simonetti O, et al. Relationship of MRI delayed contrast enhancement to irreversible injury, infarct age, and contractile function. Circulation. 1999;100:1992-2002. Malliaras K, Smith RR, Kanazawa H, Yee K, Seinfeld J, Tseliou E, et al. Validation of contrastenhanced magnetic resonance imaging to monitor regenerative efficacy after cell therapy in a porcine model of convalescent myocardial infarction. Circulation. 2013;128:2764-75. Pop M, Ghugre NR, Ramanan V, Morikawa L, Stanisz G, Dick AJ, et al. Quantification of fibrosis in infarcted swine hearts by ex vivo late gadolinium-enhancement and diffusion-weighted MRI methods. Phys Med Biol. 2013;58:5009-28. Iles LM, Ellims AH, Llewellyn H, Hare JL, Kaye DM, McLean CA, et al. Histological validation of cardiac magnetic resonance analysis of regional and diffuse interstitial myocardial fibrosis. Eur Heart J Cardiovasc Imaging. 2015;16:14-22. Bellin MF, Van Der Molen AJ. Extracellular gadolinium-based contrast media: an overview. Eur J Radiol. 2008;66:160-7. Attili AK, Schuster A, Nagel E, Reiber JH, van der Geest RJ. Quantification in cardiac MRI: advances in image acquisition and processing. Int J Cardiovasc Imaging. 2010;26 Suppl 1:27-40. Heiberg E, Sjogren J, Ugander M, Carlsson M, Engblom H, Arheden H. Design and validation of Segment--freely available software for cardiovascular image analysis. BMC Med Imaging. 2010;10:1. Hor KN, Gottliebson WM, Carson C, Wash E, Cnota J, Fleck R, et al. Comparison of magnetic resonance feature tracking for strain calculation with harmonic phase imaging analysis. JACC Cardiovasc Imaging. 2010;3:144-51. Augustine D, Lewandowski AJ, Lazdam M, Rai A, Francis J, Myerson S, et al. Global and regional left ventricular myocardial deformation measures by magnetic resonance feature tracking in healthy volunteers: comparison with tagging and relevance of gender. J Cardiovasc Magn Reson. 2013;15:8. Schuster A, Morton G, Hussain ST, Jogiya R, Kutty S, Asrress KN, et al. The intra-observer reproducibility of cardiovascular magnetic resonance myocardial feature tracking strain assessment is independent of field strength. Eur J Radiol. 2013;82:296-301. Khan JN, Singh A, Nazir SA, Kanagala P, Gershlick AH, McCann GP. Comparison of cardiovascular magnetic resonance feature tracking and tagging for the assessment of left ventricular systolic strain in acute myocardial infarction. Eur J Radiol. 2015. Morton G, Schuster A, Jogiya R, Kutty S, Beerbaum P, Nagel E. Inter-study reproducibility of cardiovascular magnetic resonance myocardial feature tracking. J Cardiovasc Magn Reson. 2012;14:43. Wu L, Germans T, Guclu A, Heymans MW, Allaart CP, van Rossum AC. Feature tracking compared with tissue tagging measurements of segmental strain by cardiovascular magnetic resonance. J Cardiovasc Magn Reson. 2014;16:10.
4
69
PART TWO CHAPTER 4
19. Cowan BR, Peereboom SM, Greiser A, Guehring J, Young AA. Image Feature Determinants of Global and Segmental Circumferential Ventricular Strain From Cine CMR. JACC Cardiovasc Imaging. 2014. 20. Gho JM, van Es R, Stathonikos N, Harakalova M, te Rijdt WP, Suurmeijer AJ, et al. High resolution systematic digital histological quantification of cardiac fibrosis and adipose tissue in phospholamban p.Arg14del mutation associated cardiomyopathy. PLoS ONE. 2014;9:e94820. 21. Koudstaal S, Jansen of Lorkeers S, Gho JM, van Hout GP, Jansen MS, Grundeman PF, et al. Myocardial infarction and functional outcome assessment in pigs. J Vis Exp. 2014. 22. Huisman A, Looijen A, van den Brink SM, van Diest PJ. Creation of a fully digital pathology slide archive by high-volume tissue slide scanning. Hum Pathol. 2010;41:751-7. 23. Snijders TAB, Bosker RJ. Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Modeling. 2nd ed. London: Sage Publications; 2012. 24. LaHuis DM, Hartman MJ, Hakoyama S, Clark PC. Explained Variance Measures for Multilevel Models. Organizational Research Methods. 2014;17:433-51. 25. Amado LC, Gerber BL, Gupta SN, Rettmann DW, Szarf G, Schock R, et al. Accurate and objective infarct sizing by contrast-enhanced magnetic resonance imaging in a canine myocardial infarction model. J Am Coll Cardiol. 2004;44:2383-9. 26. Hsu LY, Natanzon A, Kellman P, Hirsch GA, Aletras AH, Arai AE. Quantitative myocardial infarction on delayed enhancement MRI. Part I: Animal validation of an automated feature analysis and combined thresholding infarct sizing algorithm. J Magn Reson Imaging. 2006;23:298-308.
70
COMPARISON OF CMR AND HISTOLOGICAL FIBROSIS
SUPPLEMENTAL MATERIAL
4
Supplementary Figure 1. Whole heart slice fibrosis analysis, one animal (excluded from analysis due to absence of left ventricular fibrosis). The percentage of fibrosis is shown using a color scale. Ant. = anterior; LV = left ventricle; Post. = posterior; RV = right ventricle.
71
PART TWO
|
Finding Fibrosis Patterns
Chapter
5
Endogenous Contrast MRI of Cardiac Fibrosis: Beyond Late Gadolinium Enhancement Published as J Magn Reson Imaging. 2015;41:1181-9
Johannes M.I.H. Gho1*, Joep W.M. van Oorschot2*, Gerardus P.J. van Hout1, Martijn Froeling2, Sanne J. Jansen of Lorkeers1, Imo E. Hoefer1, Pieter A. Doevendans1, Peter R. Luijten2, Steven A.J. Chamuleau1, Jaco J.M. Zwanenburg2 *Both authors contributed equally to this manuscript 1
Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands
2
Department of Radiology, University Medical Center Utrecht, Utrecht, The Netherlands
PART TWO CHAPTER 5
ABSTRACT The aim of this review is to provide an overview of detection of cardiac fibrosis with MRI using current standards and novel endogenous MRI techniques. Assessment of cardiac fibrosis is important for diagnosis, prediction of prognosis and follow up after therapy. During the last years, progress has been made in fibrosis detection using MRI. Cardiac infarct size can be assessed non-invasively with late gadolinium enhancement (LGE). Several methods for fibrosis detection using endogenous contrast have been developed, such as native T1-mapping, T1Ď -mapping, Magnetization Transfer Imaging and T2*-mapping. Each of these methods will be described, providing the basic methodology, showing potential applications from applied studies, and discussing the potential and challenges or pitfalls. We will also identify future steps and developments that are needed for bringing these methods to the clinical practice.
74
ENDOGENOUS CONTRAST MRI OF CARDIAC FIBROSIS
INTRODUCTION In myocardial tissue, the interstitial collagen matrix surrounds and supports the cardiac myocytes. An excessive increase in myocardial collagen is defined by thickening and scarring of connective tissue, usually as a result of injury. Scar tissue can lead to heart failure by decreasing ventricular systolic function, induction of diastolic dysfunction and adverse cardiac remodeling. There are different types of myocardial fibrosis that result from different etiologies. The first type is replacement fibrosis, that results from the replacement of dead cardiomyocytes. Replacement fibrosis occurs after injury such as myocardial infarction, and leads to a localized distribution of collagen, which mainly consists of type 1 collagen. The second type is interstitial reactive fibrosis, which is a progressive form of fibrosis resulting from an increase in collagen synthesis by myofibroblasts. Interstitial fibrosis normally shows a diffuse distribution. It can be seen in almost any cardiac disease, and is common in idiopathic dilated cardiomyopathy and hypertrophic cardiomyopathy,1â&#x20AC;&#x201C;3 but other pathologies, such as amyloidosis and hypertension have also been associated with infiltrative interstitial fibrosis in
5
the heart and expansion of the extracellular matrix.4,5 Regardless of etiology, maladaptive cardiac fibrosis increases the risk of potentially lethal complications such as arrhythmias and heart failure, underlining the importance of effective treatment. In addition to existing treatments, there is an emerging focus on regenerative medicine. One example is stem cell therapy, which aims to restore damaged myocardium by stem cell injection and as a result myocardial function. 6 Adequate diagnosis and precise identification of the fibrotic region will improve the effectiveness of the treatment, but also allows for better prediction of the prognosis and outcome after therapy.7 Hence, precise detection of myocardial fibrosis becomes more important.8 Current techniques such as endomyocardial biopsies can provide a qualitative and quantitative assessment of collagen. However they do require an invasive procedure and provide only regional information.2 Cardiac infarct size can be assessed non-invasively with Cardiac Magnetic Resonance Imaging (MR). The most common technique is late gadolinium enhancement (LGE), which allows for indirect detection of fibrosis using an exogenous contrast agent.9 Currently, there is an increased interest for direct detection of myocardial fibrosis without the use of an exogenous contrast agent, to overcome the limitations associated with contrast agents and to obtain a quantitative standardized measurement to assess myocardial fibrosis. This is possible by exploiting the various endogenous contrast mechanisms available using MRI. Several methods for direct detection of myocardial fibrosis have been proposed. In this review the current standard MRI method, LGE, will be described, followed by an overview of novel contrast mechanisms, such as T1, T2, T2*, T1Ď and magnetization transfer imaging (Table 1). These different endogenous MRI techniques will be illustrated with data from previous published literature and a pilot study in a porcine model of chronic ischemic heart failure (see Materials and MRI methods).
75
76 - Assessment of myocardial edema in acute phase - Direct method without need for exogenous contrast - Detection of intramyocardial hemorrhage
- Direct method without need for exogenous contrast
- T2 depends on tissue composition and is mainly influenced by edema
- T2* reflects both local tissue content and structure, which are both changed when collagen content changes
- T1Ď depends mainly on macromolecular interactions in the tissue
- Based on the transfer of magnetization from the hydrogen nuclei with restricted motion to hydrogen nuclei of free water, associated with macromolecular sensitivity
T2 mapping
T2*
T1Ď
Magnetization transfer imaging
- Direct method without need for exogenous contrast
- Direct method without need for exogenous contrast
- T1 depends on tissue composition and could change with varying collagen content
Native T1 mapping
- Assessment of diffuse fibrosis
- Not yet (pre-)clinically validated
- Homogeneity of the spin lock pulse is important - High Specific absorption rate (SAR) of spin lock pulses - Long acquisition time to obtain a T1Ď map - Specificity for fibrosis not yet known - Not yet clinically validated
- Influence of macroscopic field inhomogeneities - Shimming is very important - Unclear whether hemorrhage can be distinguished from fibrosis - Not yet clinically validated
- Only proven to work in acute phase - Minor tissue contrast in chronic myocardial fibrosis
- Only minor contrast differences (low sensitivity) - Not yet clinically validated - Specificity for collagen not yet known
- Contrast administration, causing variability and possible toxic effects - Require long acquisition time to acquire pre- and post contrast T1 map - Blood sample analysis necessary for hematocrit - Not yet clinically validated
- Quantitative assessment Extra Cellular Volume (ECV)
- Contrast enhanced T1 mapping relies on the same principle as LGE, but provides absolute T1 relaxation times after contrast injection.
Contrast enhanced T1 mapping (ECV mapping)
Methods using endogenous contrast (without contrast agents)
- Limited quantitative assessment of fibrosis - Restricted assessment of diffuse fibrosis - Contrast administration, causing possible toxic effects / variability due to lack of standardized protocol
- Clinically validated - Qualitative assessment of fibrosis - Prognostic value
Disadvantages
- Relies on difference in contrast agent washout between normal and diseased myocardium
Advantages
Late gadolinium enhancement (LGE)
Contrast-enhanced methods
Methodology
Table 1. Overview of Cardiac MR fibrosis detection methods
PART TWO CHAPTER 5
ENDOGENOUS CONTRAST MRI OF CARDIAC FIBROSIS
Contrast enhanced MRI of myocardial fibrosis Late gadolinium enhancement Late gadolinium enhancement is the non-invasive reference standard for detection of myocardial infarct size and location. The technique relies on the difference in contrast agent washout between normal and diseased myocardium.10 Gadolinium contrast agents reduce the T1 relaxation time of tissue. In regions of myocardial fibrosis or necrosis, the retention of gadolinium contrast agent is prolonged compared to healthy myocardium. This results in an increased signal intensity of this region on a T1 weighted MRI image, by adding an inversion pulse with a well-chosen inversion time to null the signal of healthy myocardium. Replacement fibrosis, e.g., due to ischemic heart disease can be clearly distinguished as it is well-delineated on LGE images. LGE cardiac MR provides prognostic value as has been demonstrated in several ischemic and non-ischemic cardiomyopathies.11–14 LGE also provides guidance to therapeutic strategies, usually in combination with other modalities (e.g., echocardiography and electromechanical mapping). This has been demonstrated for different therapeutic interventions. LGE in
5
combination with electromechanical mapping allows to guide cell injections to the target area in stem cell therapy experimentally. The use of LGE prior to cardiac resynchronization therapy 15
(CRT) can improve appropriateness of CRT, and implantable cardioverter-defibrillator (ICD) implantation. This reduces the rate of ‘non-responders’ and improves arrhythmia risk stratification by guidance of left ventricular lead deployment away from scarred myocardium for an improved clinical outcome.16–18 Despite the clinical value of LGE in determining infarct size, there are some limitations and drawbacks to this method. Adverse reactions after intravenous administration of gadoliniumbased contrast agents are rare but potentially life threatening, including anaphylactic reactions and especially in patients with renal failure contrast induced nephropathy and nephrogenic systemic fibrosis.19,20 Using data from the Food and Drug Administrations Adverse Event Reporting system between 1988 and 2012, 614 anaphylaxis cases have been reported, leading to mortality in 7.2% of the reports related to gadolinium based contrast agents (GBCA). 21 GBCAs have also been associated with nephrotoxicity, especially when given at high doses (>0.3 mmol/kg).22 The incidence nephrogenic systemic fibrosis (NSF) prior to 2008 varied from 0.26% in patients on dialysis up to 8.8% in patients with a eGFR smaller than 15 ml/min/1.73m2 without hemodialysis.23 Since the connection between NSF and GBCAs has become known, MRI protocols have changed and incorporated a contra-indication for using GBCAs in patients with severe renal failure. MRI methods based on endogenous contrast may be particularly important for these patients. Next to adverse reactions, the results from LGE vary between different imaging studies. The infarct size can be significantly overestimated, depending on the applied intensity threshold settings, or on the human observer if manual outlining is performed.9,24 The sensitivity to the intensity threshold is even higher in patients with in various cardiomyopathies that result in less focal areas of fibrosis.2 Furthermore, precise timing of the MRI exam after contrast agent administration is important to prevent over- or underestimation of scarred tissue and thus an adequate, standardized protocol is needed. Because of this variability, the LGE method is less suitable for longitudinal studies, such as follow up after treatment. Finally, the most important
77
PART TWO CHAPTER 5
limitation is the difficulty to apply LGE for the detection of diffuse interstitial myocardial fibrosis. LGE relies on the difference in signal intensity between fibrotic and normal areas to generate contrast. Lack of an area with clearly unaffected myocardium will impair detection of diffuse fibrosis by LGE. Contrast enhanced T1 mapping (ECV mapping) Recent developments have enabled the measurement of a high resolution T1 map of the human myocardium in a single breath hold, with the so called Modified Look-Locker Inversion Recovery (MOLLI) sequence.25 Similar to the LGE method, a T1 map acquired after gadolinium injection can show differences in local gadolinium concentration. T1 mapping however provides a quantitative T1 relaxation time per voxel, which alleviates the need for a remote or healthy area in the analysis. By measuring the quantitative T1 relaxation time constants instead of using qualitative T1 contrast, it is possible to overcome the limitations of LGE in patients with diffuse myocardial fibrosis.26 With the pre- and post-contrast T1 of myocardium and blood, adjusted for hematocrit, the extracellular volume (ECV) can be calculated,27 using the following formula: ECV = (1 - hematocrit) (Î&#x201D;R1myocardium/Î&#x201D;R1blood). It was shown that contrast-enhanced T1 mapping allows non-invasive quantification of diffuse myocardial fibrosis by the calculation of an ECV map.28,29 Using equilibrium contrast assessment, increased diffuse myocardial fibrosis was found in case of LV impairment and diastolic dysfunction in case of patients with severe aortic stenosis.30 In a large group of 793 patients, the ECV (in areas not containing an infarct) was found to be associated with mortality and/or the need for a heart transplant or cardiac assist device.31 The equilibrium contrast method requires 30 to 45 minutes to reach contrast equilibrium, making this a complex and time-consuming technique.28 Recent studies showed that a single bolus approach is a suitable alternative in clinical practice sufficient for most myocardial ECV applications,32 as equally accurate estimates of the ECV were found.33 Although the application of contrast enhanced myocardial T1 mapping and ECV mapping provides a valuable tool to study cardiac fibrosis, it still needs to be validated in further clinical studies.34 Nonetheless, there are still limitations that stimulate the search for other imaging techniques. Calculation of an ECV map requires the acquisition of 2 different T1 maps, one before and one 15-20 minutes after contrast injection. Furthermore, the hematocrit value should be determined, which requires the withdrawal of blood before the MRI exam. This can be done by the intravenous line that is already in place for the gadolinium injection, but creates a time-consuming procedure in clinical workflow, as the blood sample needs to be analyzed, and the resulting ECV map can only be calculated afterwards with the hematocrit value, pre- and post T1 map combined. To overcome these problems and further simplify the procedure, a single post-contrast T 1 measurement was proposed. It was shown that an isolated post-contrast T1 measurement is insufficient to provide reliable results.35 Endogenous MRI contrast of myocardial fibrosis Native T1 mapping Where the native T1 map is needed for the calculation of an ECV map, the information that can be obtained from an isolated native T1 map can also be used. It has been shown that T1 mapping
78
ENDOGENOUS CONTRAST MRI OF CARDIAC FIBROSIS
without contrast agent can provide information on diffuse fibrosis.36,37 Native T1 values provide diagnostic accuracy to discriminate between normal and diffuse fibrosis in patients with nonischemic dilated cardiomyopathy and hypertrophic cardiomyopathy.38 An example of T1 mapping to detect myocardial fibrosis in patients with aortic stenosis is shown in Figure 1.37 Benefits of native T1 mapping without the use of a contrast agent would be the independence of renal function and timing of the measurement of the outcome of the quantitative T1 map. Main drawback of the method is that reported changes in T1 after fibrosis are very small, which leads to weak sensitivity of the method.37 Moreover, since the resulting T1 values in the T1 map are sensitive to the pulse sequence and scanner used in the experiment, use of native T1 mapping for diagnosis of diffuse fibrosis might be challenging. A standardized imaging protocols should be used to overcome these problems.32 A recent study in a canine animal model has shown the first evidence that native T1 mapping at 3T can be used to determine the location, size and transmurality of chronic myocardial infarction, with high sensitivity (95 %) and high specificity (97%).39 Future work should focus on the translation and validation of this technique in patient studies.
5
Figure 1. Color maps of T1 values using shortened modified Lookâ&#x20AC;&#x201C;Locker inversion in a mid-ventricular short-axis slice with the corresponding slice with late gadolinium enhancement (LGE) imaging. The left-hand panel shows a normal volunteer (T1=944 ms). The middle panels show moderate aortic stenosis (AS) with moderate left ventricular hypertrophy (T1=951 ms). The right-hand panel shows severe AS with severe left ventricular hypertrophy (T1=1020 ms).37 (Adapted from Ref. 37, with permission)
79
PART TWO CHAPTER 5
Figure 2. T2 maps with corresponding LGE images of three animals with acute myocardial infarction. The area of the myocardium with a higher T2 correlates with the enhanced area on the LGE images. (Adapted from Ref. 41, with permission)
T2 mapping The T2 relaxation parameter is prolonged in regions of edema. Edema measured by T2 mapping in animal studies has been shown to correlate with the region of acute ischemic injury, which led to interest in using T2-weighted imaging to assess myocardial damage in the acute stage of ischemia.40 An example of detection of acute myocardial edema in an animal model with T2 mapping is shown in Figure 2.41 Also in patients the T2 relaxation time provides an effective parameter to detect myocardial edema in the case of acute myocardial infarction.42 For chronic myocardial fibrosis, T足2 is less effective to generate contrast with healthy myocardium.42 In a study in diabetic mice at ultrahigh field (11.75 T) however, a significant relation was found between collagen fractional area and T2 relaxation times.43 Further research is needed to investigate if these results can be translated to application in a clinical setting at lower field strengths. T2* mapping The T2* relaxation time constant characterizes the relaxation of the transverse magnetization, as a result of static magnetic field inhomogeneities.44 Therefore, T2* mapping enables the measurement of macroscopic and mesoscopic magnetic field inhomogeneities. Macroscopic inhomogeneities are mainly caused by inhomogeneity of the main magnetic field such as imperfect shimming or to air-tissue transitions. Mesoscopic field inhomogeneities, however, are induced by the biological structure of the tissue. For example, it has been shown that T2* mapping enables the imaging of myocardial fiber structure.45 Currently the main use of T2* mapping cardiac MR is for assessment of myocardial iron. It has been proven that T2* mapping can provide insight on cardiac iron overload in thalassemia major patients and other iron overload conditions.46,47
80
ENDOGENOUS CONTRAST MRI OF CARDIAC FIBROSIS
From cartilage, it is known that the T2* relaxation time in collagen rich tissue becomes very short.48 A short T2* means that the signal decays very quickly after excitation. To enable detection of such ‘short living’ signal, Ultra Short Echo Time (UTE) MRI can be used.49 Ex vivo results in a rat model of myocardial infarction show that the area with a short T2* as detected with UTE MRI, correlates with the area of myocardial fibrosis as determined by histology using a picrosirius red staining.50 Iron staining was negative in this study, so the T2* shortening could be attributed to the presence of collagen only. In addition, it has been shown in an in vivo mouse model that the T2* value decreases during the initial weeks, when the infarction progresses into the chronic phase.51 An example of T2* mapping in a porcine model of chronic myocardial infarction is shown in Figure 3 (see Materials and MRI methods section). Extravasation of erythrocytes, during reperfusion after acute coronary occlusion, may lead to chronic iron deposition and chronic inflammation. Recently, Kali et al. have shown a relationship between postreperfusion intramyocardial hemorrhage iron deposits and changes in signal intensity on T2* weighted images.52 Further research is necessary to investigate to what extent
5
shortening of T2* can be attributed to collagen, and what the influence is of hemorrhage on T2* shortening in myocardial fibrosis detection after infarction. A limitation for the use of T2* mapping MRI for detection of myocardial fibrosis is the influence of macroscopic field inhomogeneities. These field differences, mainly induced at the lung-heart transition, influence the T2* map and indicate false positive areas.53 Therefore, good shimming of the heart region is very important, or sophisticated image analysis software is needed that aims to correct for the effects of macroscopic inhomogeneity with the use of the phase image.54
Figure 3. Detection of chronic myocardial infarction (4 weeks post infarction) with T2* mapping in a porcine animal model, see Materials and MRI methods section. A: post mortem TTC staining of the infarction area. B: late gadolinium enhancement (LGE) image. C: native R2* (1/T2*) map. The arrow indicates the infarcted area. Artifacts arising from macroscopic field inhomogeneities can be observed by the lungtissue interface.
T1 ρ mapping T1ρ is a less known and less used MRI relaxation mechanism, which is commonly applied to study articular cartilage.55,56 The T1ρ relaxation time describes relaxation while the magnetization is in the transverse plane, in the presence of a so-called spin-lock pulse. A spin-lock pulse is a low amplitude radiofrequency (RF) pulse on-resonance with the precessing transverse
81
PART TWO CHAPTER 5
magnetization, and the relaxation rate constant of the transverse magnetization under influence of this spin-lock pulse is the T1ρ relaxation. In conventional T1 relaxation, energy is exchanged from spins to the surrounding lattice in processes occurring near the Larmor frequency. However in the presence of a spin-lock pulse, the interactions between water and other molecules (exchange and rotation correlation times) have to be close to the frequency of the spin-lock pulse in order to contribute to the relaxation. As the rotation times of macromolecules are in the order of the frequency with which the magnetizations precesses around the spin-lock pulse, T1ρ relaxation is sensitive to watermacromolecule interactions. This culminated in the hypothesis that T1ρ is sensitive to the increase in collagen macromolecules content in case of myocardial fibrosis. An increase of T1ρ relaxation time was found in a swine model of chronic myocardial infarction. The area with a significantly higher T1ρ relaxation time constant (compared to healthy myocardium) correlated with the infarcted area confirmed by histology.57,58 It was also shown that the area of the myocardium with a significant higher T1ρ relaxation time constant correlates with the infarct area indicated by the gold standard LGE, both in a swine model of chronic myocardial infarction, and a mouse model of chronic myocardial infarction.58,59 Figure 4 shows an example of the difference in T1ρ between an infarcted area and the remote myocardium in a porcine infarct model (see Materials and MRI methods section). It has been shown that with increasing spin-lock pulse amplitude, the contrast between infarct, border zone and remote tissue is increased.57,60 However, for clinical application of T1ρ MRI, the specific absorption rate (SAR) delivered to the subjects limits the maximum achievable spin-lock amplitudes. As T1ρ mapping is a relative new technique in the field of cardiac MR, future research should aim at further elucidating the potential of this technique regarding the assessment of cardiac fibrosis. Additionally, it is important that the underlying mechanism of the effect of myocardial fibrosis on T1ρ is studied further. Where in myocardial infarction tissue consistently a higher T1ρ time is found compared to healthy myocardium,61–63 studies in cartilage and protein solutions show an opposite effect with a decrease of T1ρ time corresponding with a higher macromolecule content.64
Figure 4. Detection of chronic myocardial infarction (4 weeks post infarction) with T1ρ mapping in a porcine animal model, see Materials and MRI methods section. A: post mortem TTC staining of the infarction area. B: late gadolinium enhancement (LGE) image. C: native T1ρ map. The arrow indicates the infarcted area.
82
ENDOGENOUS CONTRAST MRI OF CARDIAC FIBROSIS
Magnetization transfer imaging Magnetization transfer imaging (MTI) is a MRI technique which is based on the transfer of magnetization from the hydrogen nuclei with restricted motion to hydrogen nuclei of free water. The hydrogen nuclei with restricted motion are mainly associated to be bound to macromolecules through hydrogen bonds. The amount of magnetization transfer can be assessed, amongst other methods, by measuring the difference in signal of free water obtained with and without the application of an off-resonance RF pulse prior to imaging. Hydrogen in bound water has a broad frequency distribution, and will be partly saturated by the application of an off-resonance RF pulse. Due to the continuous exchange between bound and free water, the saturated hydrogen will exchange with the free water, thereby reducing the signal intensity of the image obtained with the application of an off-resonance pulse. Thus, MTI provides an alternative contrast compared to the commonly used mechanisms based on T1 and T2 relaxation times, but is assumed to have similarity with T1Ď relaxation.65 It has been shown ex vivo in a rat model that the magnetization transfer rate is decreased in acute and chronically infarcted myocardium compared to healthy tissue.66 It was suggested that
5
this decrease in magnetization transfer is induced by a change in macromolecular interactions with the surrounding water as a result of myocardial infarction. The use of magnetization transfer imaging has been tested in vivo in patients with acute myocardial infarction.67 By acquiring two balanced steady-state free precession images with different levels of RF power deposition, a magnetization transfer ratio (MTR) image could be calculated (Figure 5). A significantly reduced magnetization transfer ratio was found in the infarct region in patients with subacute (2-5 days) myocardial infarction, corresponding well with the LGE images.
Figure 5. Short-axis view of a 59-year old male patient with an anterior-septal infarct. Left is the late enhancement (LGE) image. Right is an magnetization transfer (MTR) map calculated from two SSFP images with different levels of RF power deposition. Reproduced from Ref 67, with permission.
83
PART TWO CHAPTER 5
Magnetization transfer imaging for detection of myocardial infarction is a promising technique. However, it needs to be investigated if this technique can also provide information on chronic myocardial infarction and the formation of fibrosis. Since T1ρ relaxation and magnetization transfer imaging are both associated with macromolecular sensitivity, it is important to investigate the underlying principle on how the concentration of myocardial fibrosis influences T1ρ and MTI.68 Materials and MRI methods All in vivo experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals prepared by the Institute of Laboratory Animal Resources. Experiments were approved by the Animal Experimentation Committee of the University Medical Center Utrecht. Two Dalland landrace pigs (69 ± 5 kg) underwent a 90-minute percutaneous balloon occlusion of the left anterior descending (LAD) coronary artery distal to the second diagonal branch, followed by reperfusion. Four weeks after myocardial infarction, an in vivo MRI was performed under anesthesia on a clinical 3T scanner (Achieva TX, Philips Healthcare, Best, the Netherlands) with a 32-channel receive coil. The pigs were sacrificed for triphenyl tetrazolium chloride (TTC) staining of serial sectioned heart slices and histology. For the T1ρ map a single slice T1ρ-prepared, multishot gradient echo sequence with Cartesian readout was used (TE/TR = 2.1 / 4.2, slice thickness = 10 mm, resolution = 1.5 x 1.5 mm2, FOV = 350 x 350 mm2, flip angle = 15 degrees, echo train length = 15. Five different pulse lengths were set for the T1ρ preparation pulse (3, 10, 20, 30, 40 ms), with a spin lock amplitude of 500 Hz (total scan time = 7 minutes). The T2* map was acquired with a multi-echo gradient echo sequence with Cartesian readout (10 echoes, first TE = 1.46 ms, ΔTE = 0.9 ms, TR = 1875 ms, 6 slices, slice thickness = 6 mm, resolution = 2x2 mm2, FOV = 300 x 300 mm2, flip angle = 90 degrees, total scan time = 5 minutes). The T1ρ and T2* maps were calculated by pixelwise fitting of an exponential decay function. LGE MRI was performed using a 3D gradient echo sequence, fifteen minutes after injection of a gadolinium based contrast agent (TE/TR = 1.38/4.3, 20 slices, slice thickness = 6 mm, resolution = 1.75 x 1.75 mm2, FOV = 320 x 320 mm2, flip angle = 25 degrees, inversion delay = 330 ms, total scan time = 3 minutes). Future implications The need for detecting diffuse fibrosis in a non-invasive quantitative way has driven the search for new methods to detect fibrosis with or without using contrast agents, e.g., T1 mapping, T2 mapping, T2* mapping, UTE imaging, T1ρ mapping, and MTI. These MRI techniques show promising initial results, but further research is needed to better understand how the contrast is formed. ECV mapping shows promising initial results in clinical studies, but has the downside that it requires contrast administration and blood sampling. Currently, in the field of non-invasive endogenous MRI detection of myocardial fibrosis, native T1, T1ρ mapping and MTI look most promising. Further development for clinical application of these techniques is required, for instance development of a clinically applicable protocol, further validation in clinical trials and correlation with histology. Furthermore, standardized measurement protocols to assess these contrasts will also be necessary. Once established, these techniques might be applicable to a
84
ENDOGENOUS CONTRAST MRI OF CARDIAC FIBROSIS
broad clinical spectrum, including ablation therapy, valvular diseases, cardiomyopathies, ischemic heart disease and stem cell therapy. Acknowledgments We kindly acknowledge Grace Croft, Marlijn Jansen, Merel Schurink, Evelyn Velema, Cees Verlaan and Joyce Visser for their assistance with the animal experiments.
5
85
PART TWO CHAPTER 5
REFERENCES 1.
2. 3. 4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14. 15.
16.
17. 18. 19.
86
Marijianowski MM, Teeling P, Mann J, Becker AE. Dilated cardiomyopathy is associated with an increase in the type I/type III collagen ratio: a quantitative assessment. J Am Coll Cardiol. 1995;25:1263–72. Mewton N, Liu CY, Croisille P, Bluemke D, Lima JA. Assessment of myocardial fibrosis with cardiovascular magnetic resonance. J Am Coll Cardiol. 2011;57:891–903. Maron BJ. Hypertrophic Cardiomyopathy. JAMA. 2002;287:1308–20. Robbers LF, Baars EN, Brouwer WP, Beek AM, Hofman MB, Niessen HW, et al. T1 mapping shows increased extracellular matrix size in the myocardium due to amyloid depositions. Circ Cardiovasc Imaging. 2012;5:423–6. Díez J, González A, López B, Querejeta R. Mechanisms of disease: pathologic structural remodeling is more than adaptive hypertrophy in hypertensive heart disease. Nat Clin Pract Cardiovasc Med. 2005;2:209–16. Koudstaal S, Jansen Of Lorkeers SJ, Gaetani R, Gho JM, van Slochteren FJ, Sluijter JP, et al. Concise review: heart regeneration and the role of cardiac stem cells. Stem Cells Transl Med. 2013;2:434-43. Gulati A, Jabbour A, Ismail TF, Guha K, Khwaja J, Raza S, et al. Association of fibrosis with mortality and sudden cardiac death in patients with nonischemic dilated cardiomyopathy. JAMA. 2013;309:896–908. Van Slochteren FJ, Teske AJ, van der Spoel TI, Koudstaal S, Doevendans PA, Sluijter JP, et al. Advanced measurement techniques of regional myocardial function to assess the effects of cardiac regenerative therapy in different models of ischaemic cardiomyopathy. Eur Heart J Cardiovasc Imaging. 2012;13:808–18. Bondarenko O, Beek A, Hofman M, Kühl H, Twisk J, van Dockum W, et al. Standardizing the Definition of Hyperenhancement in the Quantitative Assessment of Infarct Size and Myocardial Viability Using Delayed Contrast-Enhanced CMR. J Cardiovasc Magn Reson. 2005;7:481–5. Kim RJ, Fieno DS, Parrish TB, Harris K, Chen EL, Simonetti O, et al. Relationship of MRI delayed contrast enhancement to irreversible injury, infarct age, and contractile function. Circulation. 1999;100:1992–2002. Kwong RY, Chan AK, Brown KA, Chan CW, Reynolds HG, Tsang S, et al. Impact of unrecognized myocardial scar detected by cardiac magnetic resonance imaging on event-free survival in patients presenting with signs or symptoms of coronary artery disease. Circulation. 2006;113:2733–43. Assomull RG, Prasad SK, Lyne J, Smith G, Burman ED, Khan M, et al. Cardiovascular magnetic resonance, fibrosis, and prognosis in dilated cardiomyopathy. J Am Coll Cardiol. 2006;48:1977– 85. Müller KA, Müller I, Kramer U, Kandolf R, Gawaz M, Bauer A, et al. Prognostic Value of Contrastenhanced Cardiac Magnetic Resonance Imaging in Patients with Newly Diagnosed NonIschemic Cardiomyopathy: Cohort Study. PLoS One. 2013;8:e57077. Karamitsos TD, Francis JM, Myerson S, Selvanayagam JB, Neubauer S. The role of cardiovascular magnetic resonance imaging in heart failure. J Am Coll Cardiol. 2009;54:1407–24. van Slochteren FJ, van Es R, Koudstaal S, van der Spoel TI, Sluijter JP, Verbree J, et al. Multimodality infarct identification for optimal image-guided intramyocardial cell injections. Neth Heart J. 2014;22:493-500. Leyva F, Foley PW, Chalil S, Ratib K, Smith RE, Prinzen F, et al. Cardiac resynchronization therapy guided by late gadolinium-enhancement cardiovascular magnetic resonance. J Cardiovasc Magn Reson. 2011;13:29. Aggarwal NR, Martinez MW, Gersh BJ, Chareonthaitawee P. Role of cardiac MRI and nuclear imaging in cardiac resynchronization therapy. Nat Rev Cardiol. 2009;6:759–70. Vernooy K, van Deursen CJ, Strik M, Prinzen FW. Strategies to improve cardiac resynchronization therapy. Nat Rev Cardiol. 2014;11:481–93. Bellin MF, Van Der Molen AJ. Extracellular gadolinium-based contrast media: an overview. Eur J Radiol. 2008;66:160–7.
ENDOGENOUS CONTRAST MRI OF CARDIAC FIBROSIS
20. Murphy KP, Szopinski KT, Cohan RH, Mermillod B, Ellis JH. Occurrence of adverse reactions to gadolinium-based contrast material and management of patients at increased risk: a survey of the American Society of Neuroradiology Fellowship Directors. Acad Radiol. 1999;6:656–64. 21. Raisch DW, Garg V, Arabyat R, Shen X, Edwards BJ, Miller FH, et al. Anaphylaxis associated with gadolinium-based contrast agents: data from the Food and Drug Administration’s adverse event reporting system and review of case reports in the literature. Expert Opin Drug Saf. 2014;13:15–23. 22. Perazella MA. Gadolinium-contrast toxicity in patients with kidney disease: nephrotoxicity and nephrogenic systemic fibrosis. Curr Drug Saf. 2008;3:67–75. 23. Reiter T, Ritter O, Prince MR, Nordbeck P, Wanner C, Nagel E, et al. Minimizing risk of nephrogenic systemic fibrosis in cardiovascular magnetic resonance. J Cardiovasc Magn Reson. 2012;14:31. 24. Beek AM, Bondarenko O, Afsharzada F, van Rossum AC. Quantification of late gadolinium enhanced CMR in viability assessment in chronic ischemic heart disease: a comparison to functional outcome. J Cardiovasc Magn Reson. 2009;11:6. 25. Messroghli DR, Greiser A, Fröhlich M, Dietz R, Schulz-Menger J. Optimization and validation of a fully-integrated pulse sequence for modified look-locker inversion-recovery (MOLLI) T1 mapping of the heart. J Magn Reson Imaging. 2007;26:1081–6. 26. Sibley CT, Noureldin RA, Gai N, Nacif MS, Liu S, Turkbey EB, et al. T1 Mapping in cardiomyopathy at cardiac MR: comparison with endomyocardial biopsy. Radiology. 2012;265:724–32. 27. Messroghli DR, Nordmeyer S, Dietrich T, Dirsch O, Kaschina E, Savvatis K, et al. Assessment of diffuse myocardial fibrosis in rats using small-animal Look-Locker inversion recovery T1 mapping. Circ Cardiovasc Imaging. 2011;4:636–40. 28. Flett AS, Hayward MP, Ashworth MT, Hansen MS, Taylor AM, Elliott PM, et al. Equilibrium contrast cardiovascular magnetic resonance for the measurement of diffuse myocardial fibrosis: preliminary validation in humans. Circulation. 2010;122:138–44. 29. Kellman P, Wilson JR, Xue H, Ugander M, Arai AE. Extracellular volume fraction mapping in the myocardium, part 1: evaluation of an automated method. J Cardiovasc Magn Reson. 2012;14:63. 30. Flett AS, Sado DM, Quarta G, Mirabel M, Pellerin D, Herrey AS, et al. Diffuse myocardial fibrosis in severe aortic stenosis: an equilibrium contrast cardiovascular magnetic resonance study. Eur Heart J Cardiovasc Imaging. 2012;13:819–26. 31. Wong TC, Piehler K, Meier CG, Testa SM, Klock AM, Aneizi AA, et al. Association between extracellular matrix expansion quantified by cardiovascular magnetic resonance and short-term mortality. Circulation. 2012;126:1206–16. 32. Moon JC, Messroghli DR, Kellman P, Piechnik SK, Robson MD, Ugander M, et al. Myocardial T1 mapping and extracellular volume quantification: a Society for Cardiovascular Magnetic Resonance (SCMR) and CMR Working Group of the European Society of Cardiology consensus statement. J Cardiovasc Magn Reson. 2013;15:92. 33. White SK, Sado DM, Fontana M, Banypersad SM, Maestrini V, Flett AS, et al. T1 mapping for myocardial extracellular volume measurement by CMR: bolus only versus primed infusion technique. JACC Cardiovasc Imaging. 2013;6:955–62. 34. Kellman P, Wilson JR, Xue H, Bandettini WP, Shanbhag SM, Druey KM, et al. Extracellular volume fraction mapping in the myocardium, part 2: initial clinical experience. J Cardiovasc Magn Reson. 2012;14:64. 35. Miller CA, Naish JH, Bishop P, Coutts G, Clark D, Zhao S, et al. Comprehensive validation of cardiovascular magnetic resonance techniques for the assessment of myocardial extracellular volume. Circ Cardiovasc Imaging. 2013;6:373–83. 36. Dass S, Suttie JJ, Piechnik SK, Ferreira VM, Holloway CJ, Banerjee R, et al. Myocardial tissue characterization using magnetic resonance noncontrast t1 mapping in hypertrophic and dilated cardiomyopathy. Circ Cardiovasc Imaging. 2012;5:726–33. 37. Bull S, White SK, Piechnik SK, Flett AS, Ferreira VM, Loudon M, et al. Human non-contrast T1 values and correlation with histology in diffuse fibrosis. Heart. 2013;99:932-7. 38. Puntmann VO, Voigt T, Chen Z, Mayr M, Karim R, Rhode K, et al. Native t1 mapping in differentiation of normal myocardium from diffuse disease in hypertrophic and dilated cardiomyopathy. JACC Cardiovasc Imaging. 2013;6:475–84.
5
87
PART TWO CHAPTER 5
39. Kali A, Cokic I, Tang RL, Yang HJ, Sharif B, Marbán E, et al. Determination of location, size, and transmurality of chronic myocardial infarction without exogenous contrast media by using cardiac magnetic resonance imaging at 3 T. Circ Cardiovasc Imaging. 2014;7:471–81. 40. Salerno M, Kramer CM. Advances in parametric mapping with CMR imaging. JACC Cardiovasc Imaging. 2013;6:806–22. 41. Giri S, Chung YC, Merchant A, Mihai G, Rajagopalan S, Raman S V, et al. T2 quantification for improved detection of myocardial edema. J Cardiovasc Magn Reson. 2009;11:56. 42. Abdel-Aty H, Zagrosek A, Schulz-Menger J, Taylor AJ, Messroghli D, Kumar A, et al. Delayed enhancement and T2-weighted cardiovascular magnetic resonance imaging differentiate acute from chronic myocardial infarction. Circulation. 2004;109:2411–6. 43. Bun SS, Kober F, Jacquier A, Espinosa L, Kalifa J, Bonzi MF, et al. Value of in vivo T2 measurement for myocardial fibrosis assessment in diabetic mice at 11.75 T. Invest Radiol. 2012;47:319–23. 44. Reeder SB, Faranesh AZ, Boxerman JL, McVeigh ER. In vivo measurement of T*2 and field inhomogeneity maps in the human heart at 1.5 T. Magn Reson Med. 1998;39:988–98. 45. Köhler S, Hiller KH, Waller C, Jakob PM, Bauer WR, Haase A. Visualization of myocardial microstructure using high-resolution T*2 imaging at high magnetic field. Magn Reson Med. 2003;49:371–5. 46. Carpenter JP, Roughton M, Pennell DJ. International survey of T2* cardiovascular magnetic resonance in β-thalassemia major. Haematologica. 2013;98:1368–74. 47. Baksi AJ, Pennell DJ. T2* imaging of the heart: methods, applications, and outcomes. Top Magn Reson Imaging. 2014;23:13–20. 48. Reiter DA, Lin PC, Fishbein KW, Spencer RG. Multicomponent T2 relaxation analysis in cartilage. Magn Reson Med. 2009;61:803–9. 49. Holmes JE, Bydder GM. MR imaging with ultrashort TE (UTE) pulse sequences: Basic principles. Radiography. 2005;11:163–74. 50. de Jong S, Zwanenburg JJ, Visser F, van der Nagel R, van Rijen HV, Vos MA, et al. Direct detection of myocardial fibrosis by MRI. J Mol Cell Cardiol. 2011;51:974–9. 51. Aguor EN, Arslan F, van de Kolk CW, Nederhoff MG, Doevendans PA, van Echteld CJ, et al. Quantitative T 2* assessment of acute and chronic myocardial ischemia/reperfusion injury in mice. MAGMA. 2012;25:369–79. 52. Kali A, Kumar A, Cokic I, Tang RL, Tsaftaris SA, Friedrich MG, et al. Chronic manifestation of postreperfusion intramyocardial hemorrhage as regional iron deposition: a cardiovascular magnetic resonance study with ex vivo validation. Circ Cardiovasc Imaging. 2013;6:218–28. 53. Atalay MK, Poncelet BP, Kantor HL, Brady TJ, Weisskoff RM. Cardiac susceptibility artifacts arising from the heart-lung interface. Magn Reson Med. 2001;45:341–5. 54. de Leeuw H, Bakker CJ. Correction of gradient echo images for first and second order macroscopic signal dephasing using phase derivative mapping. Neuroimage. 2012;60:818–29. 55. Menezes NM, Gray ML, Hartke JR, Burstein D. T2 and T1rho MRI in articular cartilage systems. Magn Reson Med. 2004;51:503–9. 56. Li X, Cheng J, Lin K, Saadat E, Bolbos RI, Jobke B, et al. Quantitative MRI using T1ρ and T2 in human osteoarthritic cartilage specimens: Correlation with biochemical measurements and histology. Magn Reson Imaging. 2011;29:324–34. 57. Witschey WR, Pilla JJ, Ferrari G, Koomalsingh K, Haris M, Hinmon R, et al. Rotating frame spin lattice relaxation in a swine model of chronic, left ventricular myocardial infarction. Magn Reson Med. 2010;64:1453–60. 58. Musthafa HS, Dragneva G, Lottonen L, Merentie M, Petrov L, Heikura T, et al. Longitudinal rotating frame relaxation time measurements in infarcted mouse myocardium in vivo. Magn Reson Med. 2013;69:1389–95. 59. Witschey WR, Zsido GA, Koomalsingh K, Kondo N, Minakawa M, Shuto T, et al. In vivo chronic myocardial infarction characterization by spin locked cardiovascular magnetic resonance. J Cardiovasc Magn Reson. 2012;14:37. 60. Oorschot J, Kaplan M, Baldus M, Luijten P, Zwanenburg J. Analysis of T1 , T2 and T1rho spin lock field dependency in myocardial infarction tissue using HRMAS spectroscopy at 11.7T. Proc Int Soc Magn Reson Med. 2012;3069.
88
ENDOGENOUS CONTRAST MRI OF CARDIAC FIBROSIS
61. Han Y, Liimatainen T, Gorman RC, Witschey WR. Assessing Myocardial Disease Using T1ρ MRI. Curr Cardiovasc Imaging Rep. 2014;7:9248. 62. Muthupillai R, Flamm SD, Wilson JM, Pettigrew RI. Radiology Acute Myocardial Infarction : Tissue Characterization with T1rho -weighted MR Imaging — Initial Experience. Radiology. 2004;606–10. 63. Musthafa HS, Dragneva G, Lottonen L, Merentie M, Petrov L, Heikura T, et al. Longitudinal rotating frame relaxation time measurements in infarcted mouse myocardium in vivo. Magn Reson Med. 2013;69:1389–95. 64. Andrasko J. Water in agarose gels studied by nuclear magnetic resonance relaxation in the rotating frame. Biophys J. 1975;15:1235–43. 65. Ulmer JL, Mathews VP, Hamilton CA, Elster AD, Moran PR. Magnetization transfer or spinlock? An investigation of off-resonance saturation pulse imaging with varying frequency offsets. AJNR Am J Neuroradiol. 1996;17:805–19. 66. Scholz TD, Hoyt RF, DeLeonardis JR, Ceckler TL, Balaban RS. Water-macromolecular proton magnetization transfer in infarcted myocardium: a method to enhance magnetic resonance image contrast. Magn Reson Med. 1995;33:178–84. 67. Weber OM, Speier P, Scheffler K, Bieri O. Assessment of magnetization transfer effects in myocardial tissue using balanced steady-state free precession (bSSFP) cine MRI. Magn Reson Med. 2009;62:699–705. 68. Scholz TD, Ceckler TL, Balaban RS. Magnetization transfer characterization of hypertensive cardiomyopathy: significance of tissue water content. Magn Reson Med. 1993;29:352–7.
5
89
PART TWO
|
Finding Fibrosis Patterns
Chapter
6
High Resolution Systematic Digital Histological Quantification of Cardiac Fibrosis and Adipose Tissue in Phospholamban p.Arg14del Mutation Associated Cardiomyopathy Published as PloS ONE 2014;9:e94820
Johannes M.I.H. Gho1, RenĂŠ van Es1, Nikolas Stathonikos2, Magdalena Harakalova1,2, Wouter P. te Rijdt3, Albert J.H. Suurmeijer4, Jeroen F. van der Heijden1, Nicolaas de Jonge1, Steven A.J. Chamuleau1, Roel A. de Weger2, Folkert W. Asselbergs1,5, Aryan Vink2
1
Department of Cardiology, Division Heart and Lungs, University Medical Center Utrecht, Utrecht, the Netherlands
2
Department of Pathology, University Medical Center Utrecht, Utrecht, the Netherlands
3
Department of Cardiology, University Medical Center Groningen, University of Groningen, Groningen, the Netherlands
4
Department of Pathology, University Medical Center Groningen, University of Groningen, Groningen, the Netherlands
5
Interuniversity Cardiology Institute of the Netherlands (ICIN), Utrecht, the Netherlands
PART TWO CHAPTER 6
ABSTRACT Myocardial fibrosis can lead to heart failure and act as a substrate for cardiac arrhythmias. In dilated cardiomyopathy diffuse interstitial reactive fibrosis can be observed, whereas arrhythmogenic cardiomyopathy is characterized by fibrofatty replacement in predominantly the right ventricle. The p.Arg14del mutation in the phospholamban (PLN) gene has been associated with dilated cardiomyopathy and recently also with arrhythmogenic cardiomyopathy. Aim of the present study is to determine the exact pattern of fibrosis and fatty replacement in PLN p.Arg14del mutation positive patients, with a novel method for high resolution systematic digital histological quantification of fibrosis and fatty tissue in cardiac tissue. Transversal mid-ventricular slices (n = 8) from whole hearts were collected from patients with the PLN p.Arg14del mutation (age 48Âą16 years; 4 (50%) male). An in-house developed open source MATLAB script was used for digital analysis of Massonâ&#x20AC;&#x2122;s trichrome stained slides (http://sourceforge.net/projects/ fibroquant/). Slides were divided into trabecular, inner and outer compact myocardium. Per region the percentage of connective tissue, cardiomyocytes and fatty tissue was quantified. In PLN p.Arg14del mutation associated cardiomyopathy, myocardial fibrosis is predominantly present in the left posterolateral wall and to a lesser extent in the right ventricular wall, whereas fatty changes are more pronounced in the right ventricular wall. No difference in distribution pattern of fibrosis and adipocytes was observed between patients with a clinical predominantly dilated and arrhythmogenic cardiomyopathy phenotype. In the future, this novel method for quantifying fibrosis and fatty tissue can be used to assess cardiac fibrosis and fatty tissue in animal models and a broad range of human cardiomyopathies.
92
CARDIAC DIGITAL HISTOLOGICAL QUANTIFICATION
INTRODUCTION A network of extracellular matrix maintains the structural integrity of the myocardium. Due to several etiologies increased deposition of collagen and other extracellular matrix proteins can occur leading to cardiac fibrosis.1 After myocardial infarction, cardiomyocytes are replaced by connective tissue leading to reparative fibrosis. In contrast, in non-ischemic cardiomyopathies, an increase in collagen synthesis by myofibroblasts results in diffuse interstitial reactive fibrosis. In arrhythmogenic cardiomyopathy (AC), fibrosis is accompanied by an increase of adipocytes leading to so-called fibrofatty replacement.2 Myocardial fibrosis is an important part of the histological characteristics in heart failure (HF) with preserved and reduced ejection fraction and may act as a substrate for cardiac arrhythmias. Adequate detection of the amount and distribution of fibrosis in the heart is important for diagnosis, predicting prognosis, treatment planning and follow-up after therapy.3, 4 The reference noninvasive standard for indirect detection of myocardial fibrosis is late gadolinium enhancement on cardiac magnetic resonance imaging (MRI).3 Thus far, detailed histological correlation studies to validate this MRI technique are scarce. Histological assessment of cardiac fibrosis is mostly limited by the small amount of tissue available in diagnostic endomyocardial biopsies that only provides regional information.2 In addition, quantification of histological fibrosis is usually
6
performed semi-quantitatively, classifying the fibrosis in limited categories. Phospholamban is a protein in the sarcoplasmic reticulum and acts as a (reversible) inhibitor of the Ca2+ pump: sarcoplasmic reticulum Ca2+-ATPase 2a (SERCA2a). On phosphorylation it dissociates from SERCA2a and thereby activates the Ca2+ pump. This cascade regulates cardiac relaxation and contractility. Several causal phospholamban (PLN) mutations have been described in humans.5-8 The p.Arg14del (c.40_42delAGA) founder mutation in the PLN gene has been associated with dilated cardiomyopathy (DCM) and recently also with AC.9 Detailed histologic analysis of the pattern of fibrosis and fatty changes in PLN mutation associated cardiomyopathies has not been extensively studied and to the best of our knowledge has not been performed on transverse heart slices. The aim of this study was to determine the exact patterns of fibrosis and fatty changes in the myocardium of patients with the PLN p.Arg14del mutation associated cardiomyopathy in relation to their clinical phenotype. This study population was used as proof-of-principle for a novel method of high resolution systematic digital quantification of fibrosis and fatty tissue in transversal cardiac slides. In the future this method may be used for detailed histological quantification and determination of the distribution pattern of cardiac fibrosis in different types of heart disease, in addition it provides a detailed high resolution reference for imaging techniques of cardiac fibrosis.
93
PART TWO CHAPTER 6
METHODS Ethics statement The study met the criteria of the code of proper use of human tissue that is used in the Netherlands. The study was approved by the scientific advisory board of the biobank of the University Medical Center Utrecht, Utrecht, the Netherlands (protocol no. 12/387). Written informed consent was obtained or in certain cases waived by the ethics committee when obtaining informed consent was not possible due to death of the patient. Hearts obtained at autopsy (n = 2) or explantation (n = 6) were collected from patients with the PLN p.Arg14del mutation. Based on their initial clinical presentation, patients were divided in two categories: predominantly DCM or AC. Three control hearts, two donor hearts not-used for transplantation and one heart obtained at autopsy of a road accident victim, were used as reference. We used a systematic methodology for high resolution digital cardiac fibrosis quantification (Figure 1). Hearts were cut in transverse (short-axis) slices of 1 cm thick starting at the apex including both ventricles. Each fourth transverse slice was fixed in formalin and divided into smaller pieces. A map of the heart slice was drawn to annotate the origin of each tissue specimen. Subsequently the samples were embedded in paraffin and Masson’s trichrome staining was performed. The slides were scanned at 20X magnification as described previously.10 Images were extracted using Aperio ImageScope v12.0.0.5039 (Aperio, Vista, CA, USA) as a TIFF file with lossless compression. The images were resized to 10% of their original size for digital analysis. To analyze the cardiac tissue, slides were divided into several layers: the epicardial area, compact myocardium and non-compact (trabeculated) myocardium (http://sourceforge.net/projects/ fibroquant/). The epicardial area was defined as the outer region of fatty tissue bordered by the first row of cardiomyocytes. The non-compact myocardial area was defined as the endocardial trabeculated region. The compact myocardium was defined by the area between the trabeculated area and the epicardial area and was artificially divided with an equidistant line in two halves. An equidistant line is one for which every point on the line is equidistant from the nearest points on both the epicardial and trabecular segmentation. In case a midline division was not feasible, e.g., due to a thin wall or originating from the interventricular septum (without epicardium), mean values of the total myocardium were used. Thus, the myocardium was divided in four layers: trabecular myocardium, inner or outer compact myocardium and the epicardium. In all four layers the percentage of connective tissue (blue), cardiomyocytes (red) and adipose tissue (cells with non stained cytoplasm) was digitally quantified using MATLAB (Release R2012a, The MathWorks, Inc., Natick, Massachusetts, United States). The percentage per region was calculated by separating the Masson’s trichrome stained slides into its constituent stains of methyl blue and ponceau-fuchsin by performing colour deconvolution.11 This produces two grayscale images depicting the concentration of the two stainings. The resulting images are filtered using a 2D median filter and further processed using morphological dilation, thresholding and closing operations. The adipose tissue is quantified using a separate function in order to highlight the “chain-link” structure of adipocytes as seen on glass slides. The original image is
94
CARDIAC DIGITAL HISTOLOGICAL QUANTIFICATION
6
Figure 1. Overview of methodology. LV, left ventricle; RV, right ventricle; Ant., anterior; Post., posterior. A. gross showing a transverse heart slice of arrhythmogenic cardiomyopathy. B. transverse slice dissection scheme. C-F. examples of digital slide processing. Regions of interest are shown in orange lines (defining the epicardium, compact myocardium divided by an equidistant midline and trabeculated part). C. slide from the left ventricle posterior wall. D. slide C after digital processing. Red: cardiomyocytes. Blue: connective tissue. Pseudo green: adipose tissue. E. slide from the right ventricle lateral wall. F. slide E after digital processing. Red: cardiomyocytes. Blue: connective tissue. Pseudo green: adipose tissue.
converted to grayscale and filtered using a 2D median filter. After that a threshold is applied and morphological opening is performed. The median filtered image is then subtracted from the morphologically opened image and the resulting image is further morphologically processed by performing consecutive closing and opening operations. This converts the â&#x20AC;&#x153;chain-linkâ&#x20AC;? structure into a black and white image where only the adipose tissue is left. The area of each constituent is determined and a percentage is calculated based on the total area that was processed.12 The epicardial region was excluded from analysis. The resulting values (percentage fibrosis and fatty tissue) were annotated to the corresponding region in the heart using an automated algorithm (Supplementary Figure S1).
95
PART TWO CHAPTER 6
Subsequently the annotated map of the transverse slice is automatically transformed to a standardized schematic overview. The annotated map was translated to a schematic overview by determining the angular properties of each separate section from the middle of the ventricles with a precision of one degree. The results of the quantification (percentage fibrosis or fatty tissue) are displayed using an easily interpretable color scale. Statistics were performed using IBM SPSS Statistics version 20. We compared mean percentages of fibrosis and adipose tissue in different areas between the two clinical phenotypes of AC and DCM and control hearts. The left ventricle was divided into a septal, posterior, posterolateral, lateral, anterolateral and anterior part. The right ventricle was divided into a posterior and anterior part. To compare mean percentages of fibrosis and adipose tissue corrected for surface area per region and condition a repeated measures analysis was performed. After Greenhouse-Geisser correction, interactions between region and condition were explored. Post hoc tests (Tukey HSD) were performed in the absence of a significant interaction.
RESULTS Clinical characteristics of the patients are summarized in Table 1. Mean age was 48Âą16 years; 4 (50%) patients were male. Five patients were known with a clinical phenotype of DCM and 3 patients were known with a clinical phenotype of AC. The schematic overviews depict distribution and percentage of fibrosis and adipose tissue for DCM (Figure 2) and arrhythmogenic patients (Figure 3). Mean values of fibrosis and adipose tissue with standard deviations per region and condition (control, AC or DCM) are presented in Supplementary Table S1. In the 8 heart slices of PLN mutation carriers, myocardial fibrosis was mainly observed in the trabecular part of the posterolateral wall of the right ventricle and in the posterolateral (mean >38%) and in lesser extent anterolateral (mean >26%) wall of the left ventricle. In the left ventricle, fibrosis
Table 1. Patient characteristics of the PLN p.Arg14del mutation carriers (n = 8) Age; mean (SD)
48 (16)
Male sex
4 (50%)
Device ICD
4 (50%)
CRT-D
3 (38%)
Left ventricular assist device
4 (50%)
First presenting symptom Arrhythmia
4 (50%)
Heart failure
4 (50%)
Heart transplantation
6 (75%)
SD, standard deviation; ICD, implantable cardioverter-defibrillator; CRT-D, cardiac resynchronization therapy defibrillator.
96
CARDIAC DIGITAL HISTOLOGICAL QUANTIFICATION
was more pronounced in the outer layer of compact myocardium than in the myocardial layers more closely to the lumen. Mean percentage of fibrosis and adipose tissue is shown in the combined schematic overviews (Figure 4).
6 Figure 2. Schematic heart slice overview of patients with a clinical phenotype of dilated cardiomyopathy. The results of the digital quantification in heart slices in percentage of fibrosis or adipose tissue are shown using a color scale. The epicardial fat has been excluded from this overview. LV, left ventricle; RV, right ventricle; Ant., anterior; Post., posterior.
Figure 3. Schematic heart slice overview of patients with a clinical phenotype of arrhythmogenic cardiomyopathy. The results of the digital quantification in heart slices in percentage of fibrosis or adipose tissue are shown using a color scale. The epicardial fat has been excluded from this overview. LV, left ventricle; RV, right ventricle; Ant., anterior; Post., posterior.
97
PART TWO CHAPTER 6
The septum and ventral wall of the right ventricle revealed the least amount of interstitial fibrosis. Fatty changes of myocardium were predominantly observed in the entire right ventricle wall (mean 37.2±14% and 28.9±4% in AC and 26.1±20% and 24.3±12% in DCM respectively in regions 7 and 8) and in the epicardial side of the compact myocardium of the posterolateral wall of the left ventricle (mean 6.9±5% in AC and 5.8±6% in DCM in region 1). Overall fatty infiltration of myocardium was more pronounced in the right (mean adipose tissue >24%) than in the left (mean adipose tissue <11%) ventricle. Mean fibrosis in control hearts was less than 6%, mean adipose tissue was 2% or less in the left ventricle myocardium and less than 15% in the right ventricular wall (Figure 5). Mean connective and adipose tissue per condition and region (n = 8, corrected for surface area) is shown in boxplots (Figure 6). Mean values of fibrosis and adipose tissue were log transformed to reduce right-skewness and heterogeneity of variance and a repeated measures analysis was performed. Comparing fibrosis we found a significant interaction between condition and region (p<0.001), therefore post-hoc testing was not performed. For adipose tissue there was no significant interaction between condition and region (p=0.470). We found no significant difference between AC and DCM in pattern of adipose tissue (p=0.382). Compared to controls, we found a higher percentage of adipose tissue in AC (p=0.028) and a trend for a higher percentage of adipose tissue in DCM (p=0.126).
Figure 4. Schematic heart slice overview of patients with the PLN p.Arg14del mutation (dilated and arrhythmogenic cardiomyopathy). The results of the mean percentage of fibrosis or adipose tissue are shown using a color scale. In total 102 heart slides of 8 heart slices were used. LV, left ventricle; RV, right ventricle; Ant., anterior; Post., posterior.
98
CARDIAC DIGITAL HISTOLOGICAL QUANTIFICATION
Figure 5. Schematic overview of fibrosis and adipose tissue in heart slices of control hearts. The results of the digital quantification in percentage of fibrosis or adipose tissue are shown using a color scale. The epicardial fat has been excluded from the analysis. LV, left ventricle; RV, right ventricle; Ant., anterior; Post., posterior.
6
DISCUSSION To the best of our knowledge this is the first study that provides the exact and detailed pattern of fibrosis and fatty changes in PLN p.Arg14del mutation associated cardiomyopathy hearts. For this study we developed a novel method for systematic high resolution digital quantification of different tissue types in the heart. This quantification has been applied to Massonâ&#x20AC;&#x2122;s trichrome stained slides of transverse cardiac slices in PLN p.Arg14del mutation associated cardiomyopathies. Interestingly we found an overlap in fibrosis and fatty changes between DCM and AC in PLN p.Arg14del mutation carriers. Myocardial fibrosis was mainly observed in the posterolateral wall of the left ventricle and in less extent in the posterolateral wall of the right ventricle. Fatty tissue was more pronounced in the myocardium bordering the epicardium of the right ventricle. We found a significant higher percentage of adipose tissue in AC compared to control hearts (p=0.028). Phospholamban is a regulator of the SERCA2a pump, important for maintaining Ca2+ homeostasis and crucial for cardiac contractility.6,13 Phosphorylation of PLN increases SERCA2a activity, leading to increased cardiac relaxation and contractility for the next beat. The precise pathophysiological mechanism in PLN p.Arg14del mutation carriers leading to cardiac fibrosis and heart failure remains unknown. Transgenic mice overexpressing the mutant PLN-R14del showed extensive myocardial fibrosis, myocyte disarray, ventricular dilation and premature death, recapitulating human cardiomyopathy.8 Co-expression of the normal and mutant protein in HEK-293 cells
99
PART TWO CHAPTER 6
resulted in SERCA2a super inhibition. From these results it was inferred that the PLN p.Arg14del mutation causes inhibition of the SERCA2a pump and thereby leads to disturbed calcium metabolism and subsequently cardiac dysfunction. These data show that this process of reactive fibrosis develops according to a specific pattern, irrespective of the phenotype of the patients. Our small sample size should thereby be taken into account and because of a significant interaction post-hoc testing was not feasible for comparing the mean fibrosis values. This varying phenotype might by influenced by effect modifiers, e.g., epigenetics or intense endurance exercise.14 In addition, it has been postulated that the pattern of reactive fibrosis is determined by myocardial stress, microvascular dysfunction and sustained activation of neurohormonal and cytokine systems.15 Future research is needed to elucidate underlying pathophysiology in PLN mutation carriers related to the phenotype. Previous (not PLN mutation specific) histopathological studies have shown similar areas of predilection in AC.16 At first, affected areas in the right ventricle (RV) were described as the classic triangle of dysplasia, including the RV inflow tract, the apex and the RV outflow tract.17
Figure 6. Boxplots of mean fibrosis and mean adipose tissue in the different regions. Boxplots of mean percentage of fibrosis (A) and adipose tissue (B) per condition in 8 regions (C) corrected for surface area. Outliers are represented by small circles or stars. AC, arrhythmogenic cardiomyopathy; DCM, dilated cardiomyopathy; LV, left ventricle; RV, right ventricle; Ant., anterior; Post., posterior; 1 = LV posterolateral wall; 2 = LV posterior wall; 3 = interventricular septum; 4 = LV anterior wall; 5 = LV anterolateral wall; 6 = LV lateral wall; 7 = RV dorsal wall; 8 = RV ventral wall.
100
CARDIAC DIGITAL HISTOLOGICAL QUANTIFICATION
Further research revealed that AC is not isolated to the RV. In a clinicopathologic study by Corrado et al. left ventricular involvement was found in 76% of cases with AC affecting both the septum and LV free wall, with a predilection for the posteroseptal and posterolateral areas.18 A recent MRI study in desmosomal mutation positive patients with AC showed involvement of the basal inferior and anterior right ventricle and the posterolateral left ventricle in AC, supporting the presence of a new biventricular triangle in early AC.19 Our results support the evidence from experimental animal models that the disease process in AC starts on the epicardial side and extends as a wave-front from the epicardium towards the endocardium.20 Clinical diagnosis of AC is made using International Task Force (revised) criteria, including structural (MRI and echocardiogram), histological (e.g. endomyocardial biopsy), electrocardiographic, arrhythmic and genetic features.2 The sensitivity of endomyocardial biopsies from the right ventricular septum in AC is low, according to our findings this might also be the case in PLN mutation associated cardiomyopathies.21 The method of fibrosis quantification presented here can be used for several applications. First, determination of the fibrosis pattern in the heart could provide an important link for genotypephenotype relationships in genetic cardiomyopathies. Previous studies have proposed morphometric evaluation of either fibrosis or adipocytes in different tissues.22-24 In our analysis fibrosis and fibrofatty replacement can be assessed simultaneously in layer specific detail. We
6
defined different regions of interest to divide the heart in an epicardial, compact and trabeculated layer. In addition, the pattern of fibrosis can be studied in the different regions of both ventricles. By studying the exact fibrosis pattern throughout the heart, patterns of disease might be discovered thereby elucidating mechanisms of pathophysiology. Numerous disease-causing genes for different cardiomyopathies have been identified during the past two decades and the challenge for the future is to link these genetic mutations to specific patterns of disease in the heart.20, 25 In future, AC with different underlying causal mutations could be compared to the PLN p.Arg14del mutation carriers with AC. The second application could be validation of myocardial imaging techniques. The reference noninvasive standard for fibrosis detection is late gadolinium enhancement on MRI.3 Adequate correlation to the gold standard of histology is important. Thus far, correlation studies are mostly done with small endomyocardial biopsies that only represent a fraction of the total myocardium or with triphenyl tetrazolium chloride (TTC) stained heart slides.26 Recently, several novel techniques for fibrosis detection have been proposed, including T1-mapping, that also require adequate correlation to histology.3, 27 Cardiac MRI images obtained before autopsy or heart transplantation, indicating fibrosis could be divided in similar segments to produce a bullseye for comparison with histopathological quantification. However, some heart failure patients are ineligible for MRI because of implanted devices, such as implantable cardioverter-defibrillator, cardiac resynchronization therapy and left ventricular assist devices. A third potential application could be systematic fibrosis quantification in animal models, for example in models of ischemic28 or non-ischemic cardiomyopathy29. Effects of novel therapies on myocardial fibrosis can be examined in randomized preclinical trials, by systematically comparing the amount of fibrosis on histology. A standardized preclinical model with induced myocardial infarction can be used to study new therapeutics, such as cell therapy, as a strategy to attenuate cardiac fibrosis and stop progression towards heart failure.30, 31
101
PART TWO CHAPTER 6
In conclusion, in PLN p.Arg14del mutation associated cardiomyopathy myocardial fibrosis is predominantly present in the left posterolateral wall, whereas fatty changes are more pronounced in the wall of the right ventricle. In the analysed heart slices from PLN p.Arg14del mutation carriers with non-ischemic cardiomyopathy we found an overlap in distribution pattern between patients with DCM and AC and a significant higher percentage of adipose tissue in AC compared to control hearts (p=0.028). We developed a novel method for systematic high resolution digital histological quantification of fibrosis and fatty tissue in the heart. This method can be used to assess cardiac fibrosis and fatty tissue in a broad range of human cardiomyopathies, animal models and can serve as gold standard for noninvasive imaging techniques. Acknowledgments We thank Petra van der Kraak-Homoet for technical assistance. We are also grateful to R.K. Stellato, MSc for statistical advice.
102
CARDIAC DIGITAL HISTOLOGICAL QUANTIFICATION
REFERENCES 1. 2.
3. 4. 5. 6. 7.
8.
9.
10. 11. 12. 13.
14.
15.
16. 17. 18.
19.
20. 21.
Weber KT, Sun Y, Bhattacharya SK, Ahokas RA, Gerling IC. Myofibroblast-mediated mechanisms of pathological remodelling of the heart. Nat Rev Cardiol. 2013;10:15-26. Marcus FI, McKenna WJ, Sherrill D, Basso C, Bauce B, Bluemke DA, et al. Diagnosis of arrhythmogenic right ventricular cardiomyopathy/dysplasia: proposed modification of the task force criteria. Circulation. 2010;121:1533-41. Mewton N, Liu CY, Croisille P, Bluemke D, Lima JA. Assessment of myocardial fibrosis with cardiovascular magnetic resonance. J Am Coll Cardiol. 2011;57:891-903. de Jong S, van Veen TA, de Bakker JM, van Rijen HV. Monitoring cardiac fibrosis: a technical challenge. Neth Heart J. 2012;20:44-8. Schmitt JP, Kamisago M, Asahi M, Li GH, Ahmad F, Mende U, et al. Dilated cardiomyopathy and heart failure caused by a mutation in phospholamban. Science. 2003;299:1410-3. MacLennan DH, Kranias EG. Phospholamban: a crucial regulator of cardiac contractility. Nat Rev Mol Cell Biol. 2003;4:566-77. Haghighi K, Kolokathis F, Pater L, Lynch RA, Asahi M, Gramolini AO, et al. Human phospholamban null results in lethal dilated cardiomyopathy revealing a critical difference between mouse and human. J Clin Invest. 2003;111:869-76. Haghighi K, Kolokathis F, Gramolini AO, Waggoner JR, Pater L, Lynch RA, et al. A mutation in the human phospholamban gene, deleting arginine 14, results in lethal, hereditary cardiomyopathy. Proc Natl Acad Sci U S A. 2006;103:1388-93. van der Zwaag PA, van Rijsingen IA, Asimaki A, Jongbloed JD, van Veldhuisen DJ, Wiesfeld AC, et al. Phospholamban R14del mutation in patients diagnosed with dilated cardiomyopathy or arrhythmogenic right ventricular cardiomyopathy: evidence supporting the concept of arrhythmogenic cardiomyopathy. Eur J Heart Fail. 2012;14:1199-207. Huisman A, Looijen A, van den Brink SM, van Diest PJ. Creation of a fully digital pathology slide archive by high-volume tissue slide scanning. Hum Pathol. 2010;41:751-7. Ruifrok AC, Johnston DA. Quantification of histochemical staining by color deconvolution. Anal Quant Cytol Histol. 2001;23:291-9. Pratt WK. Digital image processing: PIKS Scientific inside. 4th ed. Hoboken, New Jersey: Wiley; 2007. Gustavsson M, Verardi R, Mullen DG, Mote KR, Traaseth NJ, Gopinath T, et al. Allosteric regulation of SERCA by phosphorylation-mediated conformational shift of phospholamban. Proc Natl Acad Sci U S A. 2013;110:17338-43. La Gerche A, Burns AT, Mooney DJ, Inder WJ, Taylor AJ, Bogaert J, et al. Exercise-induced right ventricular dysfunction and structural remodelling in endurance athletes. Eur Heart J. 2012;33:998-1006. Masci PG, Barison A, Aquaro GD, Pingitore A, Mariotti R, Balbarini A, et al. Myocardial delayed enhancement in paucisymptomatic nonischemic dilated cardiomyopathy. Int J Cardiol. 2012;157:43-7. Basso C, Corrado D, Marcus FI, Nava A, Thiene G. Arrhythmogenic right ventricular cardiomyopathy. Lancet. 2009;373:1289-300. Marcus FI, Fontaine GH, Guiraudon G, Frank R, Laurenceau JL, Malergue C, et al. Right ventricular dysplasia: a report of 24 adult cases. Circulation. 1982;65:384-98. Corrado D, Basso C, Thiene G, McKenna WJ, Davies MJ, Fontaliran F, et al. Spectrum of clinicopathologic manifestations of arrhythmogenic right ventricular cardiomyopathy/dysplasia: a multicenter study. J Am Coll Cardiol. 1997;30:1512-20. Te Riele AS, James CA, Philips B, Rastegar N, Bhonsale A, Groeneweg JA, et al. Mutationpositive arrhythmogenic right ventricular dysplasia/cardiomyopathy: the triangle of dysplasia displaced. J Cardiovasc Electrophysiol. 2013;24:1311-20. Basso C, Bauce B, Corrado D, Thiene G. Pathophysiology of arrhythmogenic cardiomyopathy. Nat Rev Cardiol. 2012;9:223-33. Basso C, Ronco F, Marcus F, Abudureheman A, Rizzo S, Frigo AC, et al. Quantitative assessment of endomyocardial biopsy in arrhythmogenic right ventricular cardiomyopathy/dysplasia: an in vitro validation of diagnostic criteria. Eur Heart J. 2008;29:2760-71.
6
103
PART TWO CHAPTER 6
22. Krajewska M, Smith LH, Rong J, Huang X, Hyer ML, Zeps N, et al. Image analysis algorithms for immunohistochemical assessment of cell death events and fibrosis in tissue sections. J Histochem Cytochem. 2009;57:649-63. 23. Farris AB, Adams CD, Brousaides N, Della Pelle PA, Collins AB, Moradi E, et al. Morphometric and visual evaluation of fibrosis in renal biopsies. J Am Soc Nephrol. 2011;22:176-86. 24. Osman OS, Selway JL, Kepczynska MA, Stocker CJ, Oâ&#x20AC;&#x2122;Dowd JF, Cawthorne MA, et al. A novel automated image analysis method for accurate adipocyte quantification. Adipocyte. 2013;2:160-4. 25. Jacoby D, McKenna WJ. Genetics of inherited cardiomyopathy. Eur Heart J. 2012;33:296-304. 26. Schalla S, Bekkers SC, Dennert R, van Suylen RJ, Waltenberger J, Leiner T, et al. Replacement and reactive myocardial fibrosis in idiopathic dilated cardiomyopathy: comparison of magnetic resonance imaging with right ventricular biopsy. Eur J Heart Fail. 2010;12:227-31. 27. Moon JC, Messroghli DR, Kellman P, Piechnik SK, Robson MD, Ugander M, et al. Myocardial T1 mapping and extracellular volume quantification: a Society for Cardiovascular Magnetic Resonance (SCMR) and CMR Working Group of the European Society of Cardiology consensus statement. J Cardiovasc Magn Reson. 2013;15:92. 28. Pop M, Ghugre NR, Ramanan V, Morikawa L, Stanisz G, Dick AJ, et al. Quantification of fibrosis in infarcted swine hearts by ex vivo late gadolinium-enhancement and diffusion-weighted MRI methods. Phys Med Biol. 2013;58:5009-28. 29. Gho JM, Kummeling GJ, Koudstaal S, Jansen of Lorkeers SJ, Doevendans PA, Asselbergs FW, et al. Cell therapy, a novel remedy for dilated cardiomyopathy? A systematic review. J Card Fail. 2013;19:494-502. 30. Elnakish MT, Kuppusamy P, Khan M. Stem cell transplantation as a therapy for cardiac fibrosis. J Pathol. 2013;229:347-54. 31. Koudstaal S, Jansen of Lorkeers S, Gho JM, van Hout GP, Jansen MS, Grundeman PF, et al. Myocardial infarction and functional outcome assessment in pigs. J Vis Exp. 2014.
104
CARDIAC DIGITAL HISTOLOGICAL QUANTIFICATION
SUPPLEMENTARY INFORMATION
6
Supplementary Figure 1. Schematic overview of Case 6 Arrhythmogenic Cardiomyopathy with representative images. A,B, schematic heart slice overview of mean fibrosis (A) and adipose tissue (B). Corresponding slide numbers are shown inside the inner rings (n = 11). Mean percentage of fibrosis (A) or adipose tissue (B) per area have been superimposed on the overview (values rounded to the nearest whole number, values <20% are shown in white to improve readability). C,E,G, raw microscopic slide images after Massonâ&#x20AC;&#x2122;s trichrome staining, respectively corresponding to slide 1, 5 and 7. D,F,H, corresponding slides after digital processing. Red: cardiomyocytes. Blue: connective tissue. Pseudo green: adipose tissue. The slides depicted in Figure 1C-F are also derived from the same heart, 1C,D correspond to slide 3 and 1E,F correspond to slide 10.
105
PART TWO CHAPTER 6
Supplementary Table 1. Mean percentage of fibrosis and adipose tissue per region and condition Control (n=3) Fibrosis
Adipose tissue
Region 1
AC (n=3)
DCM (n=5)
Mean (%)
SD
Mean (%)
SD
Mean (%)
SD
2.5
0.3
40.1
8.8
38.9
10.5
Region 2
2.2
0.6
25.4
9.7
31.2
11.3
Region 3
2.3
0.4
17.7
4.3
33.8
12.0
Region 4
1.8
0.7
18.5
10.8
34.0
14.4
Region 5
2.4
0.8
26.4
8.2
30.6
10.2
Region 6
2.4
0.4
37.6
2.3
34.5
13.5
Region 7
4.6
1.6
24.9
3.6
23.1
9.3
Region 8
5.1
2.0
19.7
1.7
19.3
9.2
Region 1
0.7
0.7
6.9
4.7
5.8
6.0
Region 2
0.6
0.6
10.1
10.8
5.7
9.7
Region 3
0.7
0.4
3.6
3.9
1.8
1.1
Region 4
2.0
2.5
4.9
2.7
2.5
2.2
Region 5
0.7
0.4
6.4
7.5
1.9
0.7
Region 6
0.7
0.2
7.7
5.6
2.4
1.1
Region 7
7.8
6.1
37.2
14.1
26.1
20.0
Region 8
14.4
5.9
28.9
4.1
24.3
12.0
Regions correspond to the depicted regions in Figure 6C. AC, Arrhythmogenic Cardiomyopathy; DCM, Dilated Cardiomyopathy; SD, standard deviation.
106
107
PART TWO
|
Finding Fibrosis Patterns
Chapter
7
The Distribution Pattern of Fibrosis in Genetic Cardiomyopathy is Related to the Type of Pathogenic Mutation In preparation
Johannes M.I.H. Gho1*, Shahrzad Sepherkhouy2*, RenĂŠ van Es1, Magdalena Harakalova1,2, Nicolaas de Jonge1, Jasper J. van der Smagt3, Roel Goldschmeding2, Roel A. de Weger2, Folkert W. Asselbergs1, Aryan Vink2 *Both authors contributed equally
1
Department of Cardiology, Division Heart and Lungs, University Medical Center Utrecht, Utrecht, the Netherlands
2
Department of Pathology, University Medical Center Utrecht, Utrecht, the Netherlands
3
Department of Medical Genetics, University Medical Center Utrecht, Utrecht, the Netherlands
PART TWO CHAPTER 7
ABSTRACT Aims Genetic cardiomyopathies form a heterogeneous group in which myocardial fibrosis is frequently observed. We aimed to unravel the exact distribution pattern of cardiac fibrosis in cardiomyopathies caused by different specific pathogenically mutated genes. Methods and results A complete transversal slice was obtained from hearts of 28 patients with end stage heart failure due to non-ischemic cardiomyopathy and a known mutation and 4 controls. Fibrosis and fatty changes were quantitatively analyzed using digital microscopy. Hearts with mutations in the nuclear envelope protein lamin A/C (n=4) and hearts with sarcomeric mutations (n=10) showed circumferential subendocardial and midmyocardial fibrosis in left and right ventricle, except for 2 hearts from patients that initially had a hypertrophic cardiomyopathy. In hearts with a desmosomal (n=3) or a calcium pump (phospholamban; n=8) mutation fibrosis and fibrofatty replacement were observed in the left ventricle outer myocardium, mainly in the posterolateral wall, and in the right ventricle. In phospholamban mutated hearts the left ventricle was significantly more affected than in the desmosomal group. In hearts with a desminopathy (n=3) the fibrosis was mainly located in the outer myocardium of the left ventricle in combination with fibrosis and minor fibrofatty replacement in the right ventricle. Conclusion Distribution of fibrosis in genetic cardiomyopathies is related to the type of pathogenic mutation. These specific patterns may provide a roadmap for cardiac imaging interpretation and add to our understanding of disease patterns.
110
FIBROSIS IN GENETIC CARDIOMYOPATHIES
INTRODUCTION Genetic cardiomyopathies are disorders leading to declined functionality of the heart muscle and have been associated with different pathogenic mutations.1 Currently patients with cardiomyopathy are categorized based on their clinical presentation and clinical phenotype. However, the clinical presentation and clinical phenotype of a cardiomyopathy caused by a specific pathogenic mutation can vary enormously among patients. For example the phospholamban (PLN) p.Arg14del mutation can lead to a clinical presentation of heart failure, but rhythm disturbances can also be the primary presentation.2 In genetic cardiomyopathies the clinical presentation can also vary greatly due to variable penetrance, even in siblings with the same specific gene mutation. 3 This heterogenous clinical presentation of genetic cardiomyopathies, even among patients with the same genetic mutation, makes it challenging for clinicians to adequately classify these patients. Therefore more tools are needed to distinguish the different genetic cardiomyopathies and their underlying mechanisms. Although genetic cardiomyopathies are heterogeneous, histologically observed cardiac fibrosis is a hallmark feature seen in most cardiomyopathies. A multitude of mechanisms accompanied with stress and injury can cause changes in the extracellular matrix leading to increased production of collagen by interstitial fibroblasts. Fibrosis is a major cause of cardiac stiffness, diastolic and systolic impairment, arrhythmias and can eventually lead to heart failure.4 Knowledge about the pattern of fibrosis in relation to the pathogenic mutation will improve diagnosis and may eventually lead to better understanding of disease mechanisms. We recently demonstrated that PLN p.Arg14del cardiomyopathy has a specific pattern of fibrosis in the heart that is
7
independent of clinical presentation.5 The aim of this study was to unravel the amount and distribution pattern of cardiac fibrosis in cardiomyopathies caused by different specific pathogenic mutations.
METHODS The study met the criteria of the code of proper use of human tissue that is used in the Netherlands. The study was approved by the scientific advisory board of the biobank of the University Medical Center Utrecht, Utrecht, the Netherlands (protocol no. 12/387). Thirty-one hearts with cardiomyopathy and a known cardiac pathogenic mutation were obtained at transplantation (n=25) or autopsy (n=6). At the time of transplantation or death all patients had severe heart failure with reduced ejection fraction. Initially, at the time of clinical presentation, the cardiomyopathy of these patients had been classified as dilated (DCM), arrhythmogenic (ACM) or hypertrophic (HCM) cardiomyopathy by their cardiologist. For the initial diagnosis ACM the Revised Task Force criteria were used.6 Four non-cardiomyopathy hearts served as control group: three donor hearts rejected for transplantation and one heart obtained at autopsy of a patients with a non-cardiac cause of death. We used a systematic methodology for high resolution digital cardiac fibrosis quantification as previously described.5 Hearts were cut in transverse (short-axis) slices of 1 cm thick starting at the apex including both ventricles. A complete transverse heart slice of 1 cm thick was obtained
111
PART TWO CHAPTER 7
at the mid-ventricular level (halfway between the atrioventricular valves and the apex) and fixed in formalin. This heart slice was cut in smaller pieces and embedded in paraffin. A map of the heart slice was drawn to annotate the origin of each tissue specimen. Microscopic slides were stained with Masson’s trichrome (Artisan Link Pro, Dako, Glostrup, Denmark) and the slides were scanned at 20x magnification (Aperio, Leica Biosystems, Nussloch, Germany). The cardiac tissue was digitally divided into three layers: trabeculated myocardium, inner compact myocardium and outer compact myocardium. The epicardial fatty tissue was not analyzed. In case of fibrofatty replacement and in the right ventricle (RV), the border between epicardial fatty tissue and the myocardium was drawn by a line that connected the most outer parts (remnants) of the myocardium. For the interventricular septum and in some cases for the RV it was not possible to divide the myocardium into an inner and outer portion so in that case the whole compact myocardium was analyzed as one layer. In all three layers the percentage of connective tissue (blue), cardiomyocytes (red) and adipose tissue (cells with non-stained cytoplasm) was digitally quantified using MATLAB (Release R2012a, The MathWorks, Inc., Natick, Massachusetts, United States; http://sourceforge.net/projects/fibroquant). The annotated map of the transversal slice is automatically transformed to a standardized schematic overview and results in percentage fibrosis and fatty tissue are displayed using a color scale (Addendum Figure 1). For the comparison between the different mutation groups the heart was divided into 6 regions for the left ventricle (posterolateral, posterior, septal, anterior, anterolateral, and lateral part) and 2 regions for the right ventricle (posterior and anterior). Statistics were performed using IBM SPSS Statistics (Version 20.0, IBM Corporation, Armonk, New York, United States). Categorical data were compared using a Fisher’s exact test. A MannWhitney test was used to compare continuous variables between 2 groups. A Kruskal-Wallis test was used to compare continuous variables between multiple groups. Data are presented as median [interquartile range, IQR] or mean ± standard error of the mean.
RESULTS Patients were grouped in functional groups of gene mutations: sarcomeric, nuclear envelope, desmin filament network, calcium pump and desmosomal mutations (Table 1). Three patients had two pathogenic mutations in different functional groups and were excluded for further analysis. Patient characteristics are summarized in Table 1. The mean age was 36 ± 14 years; 16 patients (57%) were male. During heart transplantation or death all patients had severe reduced ejection fraction heart failure. The initial clinical diagnosis that was noted in the medical file by the treating cardiologist at the time of the diagnosis of the cardiomyopathy varied among groups. All 4 lamin A/C mutation patients presented with DCM and all 3 desmosomal mutation patients presented with ACM. The sarcomeric patients presented with DCM or HCM. Sarcomeric patients with initial HCM phenotype that progressed to a DCM phenotype had the following mutations: myosin heavy chain 7 and myosin binding protein C3. In the desminopathy and calcium pump groups patients presented with DCM or ACM. Half of the patients had previous implantation of a left ventricle assist device (LVAD). On routine coronary angiography for pre-heart transplant / LVAD
112
FIBROSIS IN GENETIC CARDIOMYOPATHIES
implantation evaluation two patients had coronary artery disease for which single-vessel percutaneous coronary intervention had been performed (1 phospholamban mutation and 1 lamin A/C mutation). The total amount of fibrosis The total amount of fibrosis at the midventricular level per condition in left and right ventricle is shown in Figure 1. The four controls revealed almost no fibrosis (median [IQR] 3% [2-4]). On average the hearts with a calcium pump mutation (25% [21-38]) and the hearts with a desminopathy (25% [24-26]) had the most fibrosis, followed by the sarcomeric gene mutations (23% [16-28]), desmosomal mutations (14% [14-17]) and the nuclear envelope gene mutations (11% [7-24]; p=0.022 between mutation groups).
Table 1. Functional groups of mutated genes
Functional group of mutated genes
Gene mutations in group
Sarcomeric
5 titin, 2 myosin binding protein C, 1 myosin heavy chain 7, 1 troponin T, 1 troponin I3
Nuclear envelope
4 lamin A/C
Desmin filament network
2 desmin, 1 alpha-B-crystallin
Calcium pump
8 phospholamban
Desmosomal
2 plakophilin-2, 1 desmoplakin
7
Figure 1. Boxplots of mean percentage of fibrosis in both ventricles in the different mutation groups. P=0.022 between mutation groups.
113
PART TWO CHAPTER 7
Distribution of fibrosis and adipose tissue Examples of Massonâ&#x20AC;&#x2122;s trichrome stains of the myocardium of the left ventricle (LV) per condition are shown in Figure 2. The stains in Figure 2 show that the fibrosis was not evenly distributed throughout the wall of the left ventricle. Schematic overviews of the distribution of fibrosis and adipose tissue for each layer (trabecular layer, inner compact myocardium, outer compact myocardium) per condition are shown in Figure 3. The average percentage of fibrosis per layer in the different mutation groups is shown in Addendum Figure 2.
Figure 2. Histological findings of the posterolateral wall in the control heart and pathogenic mutations. The slides were stained with Massonâ&#x20AC;&#x2122;s trichrome: control heart with almost no fibrosis; hearts with a desminopathy, calcium pump (phospholamban) mutation and desmosomal mutation with fibrosis in the outer layer of the myocardium; hearts with a mutation in the genes encoding nuclear envelope (lamin A/C) and sarcomeric proteins with subendocardial fibrosis and fibrosis in the inner layer of the compact myocardium.
114
FIBROSIS IN GENETIC CARDIOMYOPATHIES
A
B
Figure 3. Schematic overview of fibrosis (A) and adipose tissue (B) in the myocardium of heart slices of control hearts and the different groups of mutations. The results of the digital quantification in percentage of fibrosis and adipose tissue are shown using a colorscale. LV, left ventricle; RV, right ventricle; Ant., anterior; Post., posterior.
7
Three different distribution patterns were observed: Pattern 1: classical arrhythmogenic pattern with left ventricular involvement •
Left ventricle: fibrosis and some fatty changes where the outer compact myocardium of the posterolateral wall is most affected.
•
Right ventricle: replacement of myocardium by adipose tissue accompanied with fibrosis
•
(fibrofatty replacement). Mutation groups: calcium pump (PLN) and desmosomal mutations. Left ventricular fibrosis was more pronounced in the PLN mutation group (27% [22-39]) than in the desmosomal group (15% [14-17], p=0.024). A trend towards more adipose tissue in the RV of the desmosomal mutation group (46% [43-53]) than in the PLN group (28% [26-39], p=0.07) was found.
Pattern 2: circumferential subendocardial fibrosis LV and RV •
Left ventricle: predominantly circumferential trabecular (subendocardial) and midmyocardial fibrosis without adipose tissue.
•
Right ventricle: minor to moderate fibrosis without increased fatty infiltration.
•
Mutation groups: sarcomeric and nuclear envelope mutations. All lamin A/C mutation patients revealed pattern 2 where the RV fibrosis was most pronounced in the ventral wall.
115
PART TWO CHAPTER 7
The sarcomeric group is a heterogenous group consisting of patients presenting with DCM and HCM. Patients in the sarcomeric group with DCM (mutations in titin (n=5), troponin T2 (n=1) and troponin I3 (n=1)) all had pattern 2. In the titin subgroup 2/5 hearts also parts of the outer compact myocardium revealed fibrosis. Sarcomeric mutation patients presenting with HCM: the patient with a myosin heavy chain 7 mutation (n=1) also presented as pattern 2 with severe replacement fibrosis of the septum, whereas both patients with a myosin binding protein C3 mutation (n=2) showed a more diffuse distribution with both interstitial fibrosis and replacement fibrosis (Addendum Figure 3). Pattern 3: variable predominantly subepicardial fibrosis LV with fibrosis and minor fatty changes RV •
Left ventricle: fibrosis predominantly most pronounced in the outer compact myocardium, location varying per patient with involvement of the anterior, lateral and posterior wall. Minor fatty changes in the LV of one patient.
•
Right ventricle: moderate fibrosis and minor fibrofatty changes.
•
Mutation group: desminopathies.
Posterolateral wall of the left ventricle The posterolateral wall of the left ventricle was the most distinctive area when trabecular (endocardial), inner compact and outer compact (epicardial) myocardium were compared. Therefore subgroup analysis was performed in this region where these three layers of the wall were compared and divided into 3 groups: fibrosis/fatty changes epicardial > endocardial, epicardial < endocardial and epicardial = endocardial (difference at least 5%). Strikingly, all hearts in the desmosomal (3/3) and PLN (8/8) groups showed more fibrosis/fatty changes in the epicardial than in the endocardial area. In all hearts of the nuclear envelope group (4/4) the opposite pattern was observed with more fibrosis in the endocardial area than in the epicardial area. 6/8 (75%) of the hearts with a sarcomeric mutation also had more endocardial than
Sarcomeric n=10
Lamin A/C n=4
Desminopathy n=3
Calcium pump n=8
Desmosomal n=3
Total n=28
Age at diagnosis (yrs)
Control n=4
Table 2. Patient characteristics per mutation group
Unknown
29±15
48±8
40±2
36±13
34±7
36±14
Sex (m / f)
Unknown
6/4
3/1
2/1
3/5
2/1
16/12
Initial clinical diagnosis
-
7 DCM 3 HCM
4 DCM
2 DCM 1 ACM
6 DCM 2 ACM
3 ACM
19 DCM 6 ACM 3 HCM
LVAD
-
4/10
3/4
2/3
5/8
0/3
14/28
PM/ICD/CRT-D
-
6/10
3/4
3/3
7/8
3/3
22/28
DCM = dilated cardiomyopathy, HCM = hypertrophic cardiomyopathy, ACM = arrhythmogenic cardiomyopathy, LVAD = left ventricular assist device, PM = pacemaker, ICD = Implantable Cardioverter Defibrillator, CRT-D = Cardiac Resynchronization Therapy Defibrillator.
116
FIBROSIS IN GENETIC CARDIOMYOPATHIES
epicardial fibrosis. In the desminopathy group 1/3 (33%) hearts showed more epicardial than endocardial alterations, whereas the other 2 hearts did not reveal differences between both wall layers. The latter can be explained by the fact that fibrosis in the desminopathy is more pronounced in the anterior and posterior wall than in the posterolateral wall. Differences among mutation groups were significant (p<0.001), results are shown in Figure 4.
7 Figure 4. Pattern of fibrosis and / or adipose tissue in the posterolateral wall of the left ventricle is associated with mutation group. Trabecular (endocardial), inner compact and outer compact (epicardial) myocardium were compared and divided into 3 groups: fibrosis/fatty changes epicardial > endocardial, epicardial < endocardial and epicardial = endocardial (difference at least 5%).
DISCUSSION By applying our previously developed method for high resolution systematic digital histological quantification of cardiac fibrosis and adipose tissue, we aimed to unravel the exact distribution pattern of cardiac fibrosis in cardiomyopathies caused by different specific pathogenically mutated genes. We found that distribution pattern of fibrosis in genetic cardiomyopathies is related to the type of pathogenic genetic mutation. Especially the posterolateral wall of the left ventricle was found to be highly discriminating between groups of mutations. Desmosomal and calcium pump mutations Desmosomes are located in the intercalated disks that connect cardiomyocytes. Desmosomal mutations have been associated with ACM that was first described as fibrofatty replacement of the right ventricle in the classical triangle of dysplasia consisting of the right ventricle inflow tract, outflow tract and apex. Later it was recognized that ACM is not limited to the right ventricle
117
PART TWO CHAPTER 7
and that the left ventricle is frequently involved not only in end stage disease, but also in earlier stages.7 In the present study patients with desmosomal mutations (2 plakophilin-2 and 1 desmoplakin mutation) revealed right ventricular fibrofatty changes and fibrosis with fatty changes in the left ventricular wall, most pronounced in the posterolateral wall. Our observations in the desmosomal group are in line with earlier observations in autopsy studies from ACM patients with unknown mutations.8, 9 In addition our results in the left ventricle confirm late gadolinium enhancement in cardiovascular magnetic resonance studies that typically involves the subepicardial and midwall layers of the inferolateral and inferoseptal regions of the left ventricle in ACM.10 We recently demonstrated that hearts from patients with a PLN p.Arg14del mutation also have a pattern of right ventricle fibrofatty replacement and left ventricle fibrosis with fatty changes mostly in the posterior wall, independently of clinical presentation.5 To the best of our knowledge the present study is the first histological study that shows that PLN hearts have significantly more fibrosis in the left ventricle and a trend towards less fibrofatty replacement in the right ventricle as compared to hearts with desmosomal mutations. Our results confirm recent observations in a cohort of 142 Dutch AC patients and in a combined USA and Dutch cohort of 577 patients. In these cohorts also more left ventricular involvement in the PLN mutated patients was found compared to desmosomal mutations using electrocardiographic and cardiovascular magnetic resonance criteria.11, 12 Nuclear envelope mutations The lamin A/C gene encodes by alternative splicing for the nuclear envelope proteins lamin A and C, which are major components of the nuclear lamina that contribute to the structural integrity of the nuclear envelope and provide mechanical support for the nucleus. Mutations in the lamin A/C gene can cause several disorders including dilated cardiomyopathy accompanied by conduction abnormalities/arrhythmias.13 We observed a pattern of left ventricular subendocardial and middle layer fibrosis and subtle right ventricular fibrosis more in the ventral than dorsal wall of the right ventricle. Previous histological postmortem examinations of fullthickness myocardium in 5 patients demonstrated extensive areas of interstitial and replacement fibrosis throughout the left and right ventricular myocardium and extensive fibrosis of the cardiac conduction system without further details of the exact location of the fibrosis.14 In another study of a large Japanese family with DCM and AV block due to a lamin A/C mutation, fibrosclerosic degeneration in the AV node and interstitial fibrosis in RV endocardial biopsies were described.15 In our study, hearts with lamin A/C mutations revealed the least amount of fibrosis at the mid ventricular level as compared to the other mutations groups. This might be partly explained by the observations in a previous cardiovascular magnetic resonance study of lamin A/C mutation carriers where late gadolinium enhancement was predominantly located in the basal or midventricular septal wall, including the area around the AV node, and diffuse enhancement of LV myocardium was observed in advanced cases of the disease.16 So we add to previous literature that the fibrosis in lamin A/C mutation induced cardiomyopaty is predominantly present in the subendocardial and midmyocardial layer of the left ventricle and in the ventral wall of the right ventricle in addition to the previously established fibrosis in the AV node area.
118
FIBROSIS IN GENETIC CARDIOMYOPATHIES
Sarcomeric mutations including titin Titin is the largest human protein and two titin molecules together span the entire sarcomere and are anchored at the Z-line and M-line.17 It was reported 2012 that truncating mutations in TTN, the gene encoding titin, account for a significant portion of about 25% of the genetic etiology in familial DCM.18 Since this report there has been some debate about the pathogenicity due to observations of TTN mutations in apparently healthy controls and in the general population.19 We observed that TTN mutations induce predominantly subendocardial and midmyocardial left ventricular fibrosis and in 2/5 hearts also some fibrosis in the outer myocardial layer next fibrosis in the right ventricular wall. In a previous case report of one explanted heart with a novel mutation, c.58880insA, in exon 326 of the titin gene, some interstitial fibrosis was described without further specification of the location of the fibrosis.20 Another study reported that histopathological characteristics of cardiac specimens from subjects with TTN truncating mutations were typical of idiopathic dilated cardiomyopathy, without further specification of the fibrosis pattern.18 In a titin knock-in mouse model mimicking the c.43628insAT allele of the TTN 2-bp insertion mutation diffuse myocardial fibrosis was observed, also without further specification.21 Sarcomeric cardiomyopathies are caused by a number of mutated sarcomeric genes and can lead to a variety of phenotypes, including HCM, DCM and restrictive cardiomyopathy.22 In our series we have hearts from 3 patients that initially presented as HCM and progressed to DCM. Two of these patients had a MYBC3 mutation and hearts from these patients showed fibrosis in the wall of both ventricles without specific pattern consisting of both interstitial fibrosis and replacement type of fibrosis. One patient with a MYH7 mutation showed severe fibrosis of the septum with subendocardial and midventricular fibrosis in the left ventricle. These observations
7
are in line with previous observations in HCM patients with unknown genetic mutations where in addition to interstitial fibrosis areas of replacement fibrosis were observed, probably induced by ischemic myocardial damage due to insufficient oxygen supply to the hypertrophic myocardium.23 Two hearts in our series had a mutation in the thin sarcomeric filaments troponin I3 and troponin T2. In both hearts the fibrosis was predominantly present in the subendocardial and midmyocardial layers of both ventricles, but in the TNNT2 heart changes were only subtle. In a recent case report of hearts from two cousins with the same p.Arg186Gln (CGG>GAG)(R186Q) mutation in exon 8 of the troponin I (TNN13) gene both explanted hearts revealed increased interstitial fibrosis without further specification of the pattern and one of the two hearts showed severe scars throughout the wall of the left ventricle.24 In another report of two cases of elderly individuals (74 and 92 years old) with the Lys183del mutation in the TNNI3 gene transmural diffuse scarring of the left ventricle was observed in one case and in the other case the scarring was predominantly present in the septum and anterior wall.25 In two cases reports of the Arg92Trp mutation in the TNNT gene with clinical DCM-like features the anterior walls and interventricular septa of the hearts were replaced with extensive fibrosis and showed thinning. 26 In these case reports diffuse interstitial fibrosis was not mentioned. Thus we observed fibrosis mainly in the subendocardial and midmyocardial areas of the left and right ventricles in hearts with mutations in the sarcomere. In hearts with first presentation of HCM this pattern was not found, which was caused by large areas of replacement fibrosis.
119
PART TWO CHAPTER 7
Desminopathies Desmin is a muscle-specific intermediate filament essential for proper muscular structure and function. Mutations affecting desmin expression or promoting its aggregation lead to skeletal myopathies and/or cardiomyopathies. We observed fibrosis most pronounced in the outer compact myocardium of the left ventricle of which the location varied per patient and fibrosis and minor fibrofatty changes in the right ventricle. Our results are in line with findings of delayed enhancement in a cardiovascular MR imaging study in patients with desmin mutations with skeletal muscle weakness in which delayed enhancement was detected in a midwall position of the left ventricle without signs of fibrosis in the subendocardial region.27 In this imaging study the location of fibrosis varied per patient and involved the anterolateral, lateral and the inferoseptal LV wall. In our study the patient with the most fibrofatty replacement in the RV initially presented as AMC. It is known that atrioventricular conduction abnormalities are a frequent feature of desminopathy attributed to the fact that the heart conduction system is rich in desmin.28 However ventricular tachycardiaâ&#x20AC;&#x2122;s have also been described, which might be partly explained by the fibrofatty changes that we observed in the RV.29 Clinical implications and limitations Our results may have clinical implications. Techniques of CMR with late gadolinium enhancement have improved dramatically in recent years. In addition novel MR techniques with or without using contrast agents look promising for cardiac fibrosis detection.30 We observed that in the LV posterolateral wall the pattern of fibrosis and fatty changes is highly distinctive for groups of mutations. Therefore we hope the present study will provide a roadmap for further in vivo cardiac fibrosis assessment in patients. Our study is limited by the fact that groups of hearts with mutated genes are small. However, our study is a start of systematic analysis of hearts with known mutations. We think that in coming years cardiovascular pathologists around the world can work in collaboration with cardiologists and clinical geneticists to create databases and biobanks of cardiomyopathy patients that include clinical variables, pathogenic mutations and detailed examination of cardiac tissue. For this it is important that pathologists work according to standardized protocols when examining hearts and initiatives to create such a protocol, e.g., under the auspices of the Association of European Cardiovascular Pathology are urgently needed. In conclusion, we observed that the pattern of fibrosis and fatty changes in cardiomyopathy hearts with known pathogenic mutations as assessed by digital histological analysis is associated with the mutation group. The posterolateral wall of the LV appeared to be the most discriminating area for the identification of the different groups of mutations.
120
FIBROSIS IN GENETIC CARDIOMYOPATHIES
REFERENCES 1. 2.
3. 4. 5.
6.
7. 8. 9.
10.
11.
12.
13.
14.
15.
16.
17. 18. 19. 20.
Watkins H, Ashrafian H, Redwood C. Inherited cardiomyopathies. N Engl J Med. 2011;364:164356. van der Zwaag PA, van Rijsingen IA, Asimaki A, Jongbloed JD, van Veldhuisen DJ, Wiesfeld AC, et al. Phospholamban R14del mutation in patients diagnosed with dilated cardiomyopathy or arrhythmogenic right ventricular cardiomyopathy: evidence supporting the concept of arrhythmogenic cardiomyopathy. Eur J Heart Fail. 2012;14:1199-207. Jacoby D, McKenna WJ. Genetics of inherited cardiomyopathy. Eur Heart J. 2012;33:296-304. Weber KT, Sun Y, Bhattacharya SK, Ahokas RA, Gerling IC. Myofibroblast-mediated mechanisms of pathological remodelling of the heart. Nat Rev Cardiol. 2013;10:15-26. Gho JM, van Es R, Stathonikos N, Harakalova M, te Rijdt WP, Suurmeijer AJ, et al. High resolution systematic digital histological quantification of cardiac fibrosis and adipose tissue in phospholamban p.Arg14del mutation associated cardiomyopathy. PloS One. 2014;9:e94820. Marcus FI, McKenna WJ, Sherrill D, Basso C, Bauce B, Bluemke DA, et al. Diagnosis of arrhythmogenic right ventricular cardiomyopathy/dysplasia: proposed modification of the Task Force Criteria. Eur Heart J. 2010;31:806-14. Basso C, Bauce B, Corrado D, Thiene G. Pathophysiology of arrhythmogenic cardiomyopathy. Nat Rev Cardiol. 2012;9:223-33. Basso C, Thiene G, Corrado D, Angelini A, Nava A, Valente M. Arrhythmogenic right ventricular cardiomyopathy. Dysplasia, dystrophy, or myocarditis? Circulation. 1996;94:983-91. Corrado D, Basso C, Thiene G, McKenna WJ, Davies MJ, Fontaliran F, et al. Spectrum of clinicopathologic manifestations of arrhythmogenic right ventricular cardiomyopathy/dysplasia: a multicenter study. J Am Coll Cardiol. 1997;30:1512-20. Sen-Chowdhry S, Syrris P, Prasad SK, Hughes SE, Merrifield R, Ward D, et al. Left-dominant arrhythmogenic cardiomyopathy: an under-recognized clinical entity. J Am Coll Cardiol. 2008;52:2175-87. Groeneweg JA, van der Zwaag PA, Olde Nordkamp LR, Bikker H, Jongbloed JD, Jongbloed R, et al. Arrhythmogenic right ventricular dysplasia/cardiomyopathy according to revised 2010 task force criteria with inclusion of non-desmosomal phospholamban mutation carriers. Am J Cardiol. 2013;112:1197-206. Bhonsale A, Groeneweg JA, James CA, Dooijes D, Tichnell C, Jongbloed JD, et al. Impact of genotype on clinical course in arrhythmogenic right ventricular dysplasia/cardiomyopathyassociated mutation carriers. Eur Heart J. 2015;36:847-55. Fatkin D, MacRae C, Sasaki T, Wolff MR, Porcu M, Frenneaux M, et al. Missense mutations in the rod domain of the lamin A/C gene as causes of dilated cardiomyopathy and conductionsystem disease. N Engl J Med. 1999;341:1715-24. van Tintelen JP, Tio RA, Kerstjens-Frederikse WS, van Berlo JH, Boven LG, Suurmeijer AJ, et al. Severe myocardial fibrosis caused by a deletion of the 5' end of the lamin A/C gene. J Am Coll Cardiol. 2007;49:2430-9. Otomo J, Kure S, Shiba T, Karibe A, Shinozaki T, Yagi T, et al. Electrophysiological and histopathological characteristics of progressive atrioventricular block accompanied by familial dilated cardiomyopathy caused by a novel mutation of lamin A/C gene. J Cardiovasc Electrophysiol. 2005;16:137-45. Holmstrom M, Kivisto S, Helio T, Jurkko R, Kaartinen M, Antila M, et al. Late gadolinium enhanced cardiovascular magnetic resonance of lamin A/C gene mutation related dilated cardiomyopathy. J Cardiovasc Magn Reson. 2011;13:30. Liversage AD, Holmes D, Knight PJ, Tskhovrebova L, Trinick J. Titin and the sarcomere symmetry paradox. J Mol Biol. 2001;305:401-9. Herman DS, Lam L, Taylor MR, Wang L, Teekakirikul P, Christodoulou D, et al. Truncations of titin causing dilated cardiomyopathy. N Engl J Med. 2012;366:619-28. Golbus JR, Puckelwartz MJ, Fahrenbach JP, Dellefave-Castillo LM, Wolfgeher D, McNally EM. Population-based variation in cardiomyopathy genes. Circ Cardiovasc Genet. 2012;5:391-9. Yoskovitz G, Peled Y, Gramlich M, Lahat H, Resnik-Wolf H, Feinberg MS, et al. A novel titin mutation in adult-onset familial dilated cardiomyopathy. Am J Cardiol. 2012;109:1644-50.
7
121
PART TWO CHAPTER 7
21. Gramlich M, Michely B, Krohne C, Heuser A, Erdmann B, Klaassen S, et al. Stress-induced dilated cardiomyopathy in a knock-in mouse model mimicking human titin-based disease. J Mol Cell Cardiol. 2009;47:352-8. 22. Olivotto I, d'Amati G, Basso C, Van Rossum A, Patten M, Emdin M, et al. Defining phenotypes and disease progression in sarcomeric cardiomyopathies: contemporary role of clinical investigations. Cardiovasc Res. 2015;105:409-23. 23. Basso C, Thiene G, Corrado D, Buja G, Melacini P, Nava A. Hypertrophic cardiomyopathy and sudden death in the young: pathologic evidence of myocardial ischemia. Hum Pathol. 2000;31:988-98. 24. Roberts WC, Roberts CC, Ko JM, Grayburn PA, Tandon A, Kuiper JJ, et al. Dramatically different phenotypic expressions of hypertrophic cardiomyopathy in male cousins undergoing cardiac transplantation with identical disease-causing gene mutation. Am J Cardiol. 2013;111:1818-22. 25. Funada A, Masuta E, Fujino N, Hayashi K, Ino H, Kita Y, et al. Heterogeneity of clinical manifestation of hypertrophic cardiomyopathy caused by deletion of lysine 183 in cardiac troponin I gene. Int Heart J. 2010;51:214-7. 26. Shimizu M, Ino H, Yamaguchi M, Terai H, Uchiyama K, Inoue M, et al. Autopsy findings in siblings with hypertrophic cardiomyopathy caused by Arg92Trp mutation in the cardiac troponin T gene showing dilated cardiomyopathy-like features. Clin Cardiol. 2003;26:536-9. 27. Strach K, Sommer T, Grohe C, Meyer C, Fischer D, Walter MC, et al. Clinical, genetic, and cardiac magnetic resonance imaging findings in primary desminopathies. Neuromuscul Disord. 2008;18:475-82. 28. Goldfarb LG, Dalakas MC. Tragedy in a heartbeat: malfunctioning desmin causes skeletal and cardiac muscle disease. J Clin Invest. 2009;119:1806-13. 29. Luethje LG, Boennemann C, Goldfarb L, Goebel HH, Halle M. Prophylactic implantable cardioverter defibrillator placement in a sporadic desmin related myopathy and cardiomyopathy. Pacing Clin Electrophysiol. 2004;27:559-60. 30. van Oorschot JW, Gho JM, van Hout GP, Froeling M, Jansen Of Lorkeers SJ, Hoefer IE, et al. Endogenous contrast MRI of cardiac fibrosis: beyond late gadolinium enhancement. J Magn Reson Imaging. 2015;41:1181-9.
122
FIBROSIS IN GENETIC CARDIOMYOPATHIES
ADDENDUM
7
Addendum Figure 1. Transversal heart slice stained with Massonâ&#x20AC;&#x2122;s trichrome (A) is automatically transformed to a standardized schematic overview of fibrosis (B) and adipose tissue (C) to determine the exact pattern of fibrosis and fatty changes. The results of the digital quantification in percentage of fibrosis (B) or adipose tissue (C) is shown using a color scale. Ant. = anterior; LV = left ventricle; Post. = posterior; RV = right ventricle.
Addendum Figure 2. Average value of fibrosis of the left ventricular free wall in the trabecular, inner compact and outer compact myocardial layer.
123
PART TWO CHAPTER 7
Addendum Figure 3. Schematic overview of fibrosis in the different mutated genes in the sarcomeric group. The results of the digital quantification in percentage of fibrosis is shown using a color scale. Ant. = anterior; LV = left ventricle; Post. = posterior; RV = right ventricle.
124
125
PART THREE
|
Elucidating (Epi)genetic and Translating
Therapeutic Pathways
Chapter
8
Chromatin Regulation in Phospholamban R14del Mutation Associated Cardiomyopathy In preparation
Johannes M.I.H. Gho1*, Magdalena Harakalova1,2,4*, Shahrzad Sepehrkhouy2, Joyce van Kuik2, Noortje A.M. van den Dungen1, Erica Sierra-de Koning2, RenĂŠ van Es1, Arjan H. Schoneveld1, Pieter A. Doevendans1, Nicolaas de Jonge1, Edward E.S. Nieuwenhuis3, Hester M. den Ruijter1, Gerard Pasterkamp1, Roel A. de Weger2, Aryan Vink2, Michal Mokry3*, Folkert W. Asselbergs1,4,5* *equal contribution
1
Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands
2
Department of Pathology, University Medical Center Utrecht, Utrecht, The Netherlands
3
Laboratory of translational immunology, Wilhelmina Childrenâ&#x20AC;&#x2122;s Hospital, Utrecht, The Netherlands
4
ICIN-Netherlands Heart Institute, Utrecht, The Netherlands
5
Institute of Cardiovascular Science, Faculty of Population Health Sciences, University College London, London, United Kingdom
PART THREE CHAPTER 8
ABSTRACT Background The phospholamban (PLN) R14del founder mutation is associated with dilated and arrhythmogenic cardiomyopathy with high mortality and morbidity in hundreds of Dutch mutation carriers. DNA occupancy of H3K27ac histone mark gives valuable information about transcriptional regulation and can be used for candidate gene or biomarker discovery. For the first time we used ChIP-seq to assay disease-specific fingerprint of H3K27ac in cardiac tissue from PLN R14del end-stage cardiomyopathy to better understand the underlying pathophysiological mechanisms. Methods and results Firstly, we catalogued DNA regulatory elements by producing H3K27ac ChIP-seq data in cardiac tissue from PLN R14del cardiomyopathy carriers (n=6) compared to healthy controls (n=4). Increased H3K27ac occupancy was detected within a 5kb window near genes involved in fibrosis, developmental programs and chromatin assembly, while decreased H3K27ac occupancy was linked to genes involved in mitochondrial function and beta-oxidation of fatty acids. Secondly, we performed tissue specificity analysis of differentially regulated genes in PLN R14del cardiomyopathy using RNA data from various tissues. We observed that expression of all genes with differential H3K27ac occupancy was enriched in fetal and adult heart and adult skeletal muscle. Thirdly, in order to exclude general pathways linked to end-stage heart failure, we included ChIP-seq data from patients with other genetic types of non-ischemic cardiomyopathy (n=6) and ischemic cardiomyopathy (n=4). K-mean cluster analysis revealed 208 genes annotated to PLN R14del-specific regulatory regions. Among these, 55 genes are involved in (phospho) lipid and lipoprotein synthesis and metabolism and can be considered for candidate genes explaining the lipid storage evident in PLN R14del hearts. Conclusion Using an integrative chromatin analysis we identified the major effector genes and pathways involved in PLN R14del cardiomyopathy. Ubiquitously expressed genes with increased H3K27ac occupancy in patients are in line with general fibrotic pathways and developmental programs in end-stage organ failure. The suppression of mitochondrial lipid metabolism needed for muscle function is underlined by cardiac- and skeletal-muscle specific genes with increased H3K27ac occupancy. The detection of 208 PLN R14del-specific genes might provide valuable clues to recognize carriers at a high risk of developing cardiomyopathy using a plasma detectable biomarker.Â
128
CHROMATIN REGULATION IN CARDIOMYOPATHIES
INTRODUCTION Heart failure (HF) is a major medical problem in the western world, with a high morbidity and mortality risk.1 Cardiomyopathies are a leading cause of HF and numerous disease causing genes have been found.2, 3 The phospholamban (PLN) R14del mutation has been associated with dilated and arrhythmogenic right ventricular cardiomyopathies and is often accompanied with life threatening ventricular arrhythmias.4, 5 Phospholamban is a regulator of the sarcoplasmic reticulum Ca2+ ATPase (SERCA2a) pump, important for maintaining Ca2+ homeostasis and crucial for cardiac contractility. Within the Netherlands a relatively large population of patients exist that have the PLN R14del mutation.6 Previous experimental research showed possible superinhibition by mutant PLN leading to disturbed Ca2+ cycling.4 However, the precise pathophysiological mechanism in PLN R14del mutation carriers leading to cardiac fibrosis and HF remains unknown and age of onset and symptom severity differs greatly between patients. Previously, in a human histopathological study of PLN R14del cardiomyopathy, we found that myocardial fibrosis was predominantly located in the left posterolateral wall and fatty changes were more pronounced in the right ventricular wall.7 Induced pluripotent stem cells (iPSCs) derived cardiomyocytes from a patient with the R14del mutation showed Ca2+ handling abnormalities, abnormal cytoplasmic distribution of PLN protein and increased expression of cardiac hypertrophy markers (ANF, BNP) and a decreased MYH6/MYH7 ratio.8 Recently, next-generation sequencing (NGS) technologies have emerged to profile the DNA regulatory elements controlling the timing, localization and the level of gene transcription.9 Using a hypothesis free genome-wide approach associations with disease can be found by profiling histone marks related to active promoters and enhancers. We hypothesize that changes of the profiles of these epigenetic programs can serve as an indicator of disease pathogenesis.10 The
8
aim of this project is to profile the active chromatin histone mark H3K27ac assayed by chromatin immunoprecipitation and sequencing (ChIP-seq) to detect the regulome of human cardiac tissue from PLN R14del cardiomyopathy patients and healthy controls. To exclude chromatin activity related to end-stage heart failure and cardiomyopathies in general, we included also two distinct types of cardiomyopathies, such as ischemic cardiomyopathy and non-ischemic cardiomyopathy based on mutations in genes encoding the proteins of the sarcomere. Identification of differentially active regions in cardiac disease has the potential to lead to a more fundamental understanding of the underlying pathophysiological mechanisms related to HF and identify new treatment targets and diagnostic biomarkers.
METHODS Study design and samples This study was approved by the Biobank Research Ethics Committee, University Medical Center Utrecht, Utrecht, the Netherlands (protocol number 12/387). Written informed consent was obtained or in certain cases waived by the ethics committee when obtaining informed consent was not possible due to death of the subject. Heart samples collected at autopsy or explantation
129
PART THREE CHAPTER 8
were obtained from patients with the PLN R14del mutation (n=6). Control hearts were used as a reference (n=4). To further elaborate on R14del-specific changes hearts obtained from patients with ischemic cardiomyopathy (n=4) and from non-ischemic cardiomyopathy based on mutations in genes encoding the protein of the sarcomere (n=6) were also included. Samples used for ChIP-seq were obtained from the posterolateral part of the left ventricle or from the septum in a region halfway between the atrioventricular valves and the apex, and were stored at -80°C (Table 1, Table S1). Histopathological characterization For each individual, an adjacent part of tissue block was paraffin-embedded and stained with Masson’s trichrome. In case sections from whole heart transversal slices were available, we used high resolution systematic digital histological quantification of fibrosis and fatty tissue in trichrome stained slides to create schematic overviews showing mean fibrosis or adipose tissue per group.7 The digital histological images from the paraffin-embedded slides stained with Masson’s trichrome of adjacent regions used for ChIP-seq can be seen in Figure S1. The results of the mean percentage of fibrosis or adipose tissue in whole heart slices per group are shown in schematic overviews (Figure S2). Chromatin H3K27ac immunoprecipitation and sequencing For each sample, 10 x 10 µm thick frozen cardiac tissue slices were used for chromatin isolation using the MAGnify™ Chromatin Immunoprecipitation System kit (Life Technologies) according to manufacturer’s instructions. In short, the tissue was crosslinked with 1% formaldehyde and the crosslinking was stopped by adding 1.25 M glycine. Cells were lysed using the kit-provided lysis buffer and nuclei were sonicated using Covaris microTUBE (duty cycle 5%, intensity 2, 200 cycles per burst, 60s cycle time, 10 cycles). Sheared chromatin was diluted based on the expected number of isolated cells and was incubated with antihistone H3 acetyl K27 antibody (H3K27ac; ab4729, Abcam) pre-coupled to magnetic beads for 2 hours at 4°C. Beads were extensively washed and crosslinking was reversed by the kit-provided reverse crosslinking buffer and 20 mg/mL Proteinase K. DNA was purified using ChIP DNA Clean & Concentrator kit (Zymo Research). Isolated DNA was additionally sheared, end-repaired, sequencing adaptors were ligated and the library was amplified by PCR using primers with sample-specific barcodes according to our modification to manufacturer’s recommendations.11 After PCR, the library was purified and checked for the proper size range and for the absence of adaptor dimers on a 2% agarose gel and sequenced on SOLiD Wildfire sequencer. ChIP-seq data processing Sequencing reads were mapped against the reference genome (hg19 assembly, NCBI37) using BWA package12 (-c, -l 25, -k 2, -n 10). Multiple reads mapping to same location and strand have been collapsed to single read and only uniquely placed reads were used for peak-calling. Peaks were called using Cisgenome 2.013 ( -e 150 -maxgap 200 -minlen 200).
130
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Table 1. Detailed clinical overview of the included individuals PLN R14del (n=6)
Sarcomeric (n=6)
ICM (n=4)
Controls (n=4)
Age at explant, mean±SD
43±7.6*
35±4.5
56±7.9
unknown
Female gender, count [%]
2 (33.3%)
4 (66.7%)
1 (25%)
unknown
Cardiac device, count [%] - ICD
6 (100%)
4 (66.7%)
2 (50%)
unknown
- CRT-D
2 (33.3%)
1 (16.7%)
2 (50%)
unknown
- LVAD
3 (50%)
2 (33.3%)
1 (25%)
unknown
Last known EF, mean
21%#
21%#
20%
unknown
Heart weight [g], mean±SD
403±76.6*
349±135.6
470±165.9§
339±112.5§
*n=5; #n=4; §n=3. Abbreviations: PLN = phospholamban, ICM = ischaemic cardiomyopathy, SD = standard deviation, ICD = Implantable Cardioverter Defibrillator, CRT-D = Cardiac Resynchronization Therapy Defibrillator, LVAD = left ventricular assist device, EF = ejection fraction.
Identification of regions with differential H3K27ac occupancy (Analysis 1) Peak coordinates from all PLN and Control samples were stretched to at least 2000 base pairs and collapsed into a single common list. Overlapping peaks were merged based on their outmost coordinates. Only peaks supported by at least 2 independent datasets were further analyzed. Sequencing reads from each ChIP-seq sample were overlapped back with the common peak list to set the H3K27ac occupancy for every peak-sample pair. Sex chromosomes were excluded from analysis. Peaks with differential H3K27ac occupancy between PLN samples and controls were identified using DESeq2 using the standard settings14 (p < 0.05). Supervised hierarchical clustering was performed with quantile normalized (limma::normalizeQiantiles() function in R),
8
log2 transformed and median centered read counts per common peak. To avoid log2 transformation of zero values, one read was added to each peak. Gene annotation was performed for peaks with the peak summit located within 5kb from the transcription start site (TSS) of a gene. ToppGene Suite tool ToppFun was used for gene list enrichment analysis and candidate gene prioritization based on functional annotations and protein interactions network.15 The list of hyper- and hypoacetylated genes was tested separately using Bonferroni correction, p-value cut-off of 0.05 and gene limit of 1-2,000 genes per pathway. Tissue specificity of annotated genes (Analysis 2) To assess the tissue specificity of collected genes, publically available RNA sequencing (RNAseq) data were downloaded from two sources. Expression values represented by reads per kilobase per million (RPKM), were obtained from the RNA-Seq Atlas, a reference database for gene expression profiling in eleven healthy human tissue types (adipose, colon, heart, hypothalamus, kidney, liver, lung, ovary, skeletal muscle, spleen and testes) pooled from multiple donors.16 In addition, raw RNA-seq data from 91 and 120 days old fetal hearts were downloaded from The NIH Roadmap Epigenomics Mapping Consortium under accession numbers GSM1059495 and GSM1059494, respectively. Data were subsequently mapped to the human reference genome GRCh37/hg19 using BWA mapper and RPKM values were calculated. If genes
131
PART THREE CHAPTER 8
had more than one annotated transcript, RPKM values were averaged out. RPKM values from both sources (adult and fetal tissues) were quantile normalized, log2 transformed and mediancentered per gene. A small number (0.1) was added to each RPKM to avoid logarithmic transformation of zero values. Relative expression values higher than 4.5 and lower than -4.5 were set to 4.5 and -4.5, respectively. k-mean clustering (Analysis 3) Sequencing reads from each ChIP-seq sample (PLN (n=6), controls (n=4), ischemic (n=4) and Sarcomeric (n=6)) were overlapped back with the differentially acetylated peak list to set the H3K27ac occupancy for every peak-sample pair. Raw read counts were quantile normalized (limma::normalizeQiantiles() function in R), log2 transformed and median centered (to avoid log2 transformation of zero values, one read was added to each peak). Median value from each sample group was used to construct an n x k table where n = 4 (one value per each sample type) and k represent the number of differentially acetylated regions. The k-means (nstart = 200) function in R was user to partition the peaks into 12 different clusters. To enable the reproducibility of identified clusters set.seed(10) R command was called before the clustering. Clusters with PLN R14del-specific patters were used for gene functional annotations using ToppGene Suite (ToppFun) and GeneMANIA prediction server.17 The list of hyper- and hypoacetylated genes was tested separately using Bonferroni correction, a p-value cut-off of 0.05 and gene limit of 1-2,000 genes per pathway (ToppFun) and default automatically selected weighting settings (GeneMANIA).
RESULTS Regions with differential H3K27ac occupancy in PLN group versus control group (Analysis 1) On average, we identified 28,149±9,538 and 25,721±8,460 peaks in the PLN and the control group, respectively (Table S2). When comparing both groups 2107 peaks were differentially regulated (Figure 1 and Figure S3). Out of those, 958 peaks were hyperacetylated in the PLN group and 225 genes were annotated in their proximity including 212 protein coding, 1 antisense RNA, 3 miRNA, 2 lncRNA and 5 snoRNA genes, and 2 pseudogenes (Table S3). Gene enrichment analysis resulted in several enriched GO biological processes related to fibrosis, (cardiac) development and chromatin assembly. Significantly enriched KEGG and REACTOME pathways as well as the GO processes related to chromatin assembly were mostly based on the histone H1 cluster on chromosome 6 based on a single significantly hyperacetylated peak. Two GO cellular components were overrepresented: extracellular matrix and actin cytoskeleton (Table S4). Hyperacetylated H3K27ac peak was detected within 5kb from TSS of the PDLIM3 gene previously linked to dilated cardiomyopathy.18 Hypoacetylation in the PLN group was detected in 1149 peaks and annotated to 482 genes including 454 protein coding, 3 antisense RNA, 6 miRNA, and 6 lncRNA genes, and 2 intronic transcripts (Table S5). Gene enrichment analysis resulted in two GO biological processes related to acyltransferase activity. GO cellular components showed enrichment related to the mitochondrial matrix. Three pathways were significantly enriched: isoleucine degradation
132
CHROMATIN REGULATION IN CARDIOMYOPATHIES
(BIOCYC), metabolism (REACTOME) and fatty acid beta oxidation (WikiPathways). Hypoacetylated H3K27ac peaks in the PLN group were detected within 5kb from TSS of CDKN1C, CHKB, DES, HADHA, HADHB, KCNQ1, MLYCD, MYH7, MYL3, NDUFV1, NEU1, PIGT, PNPLA2, SLC25A20 (Human Phenotype Ontology, Cardiomyopathy, HP:0001638). RNA specificity of genes located near peaks with differential acetylation (Analysis 2) Secondly, we performed tissue specificity analysis of differentially regulated genes in PLN R14del cardiomyopathy (Table S3 and Table S5) using RNA data from various tissues. While there was no significant tissue specificity observed in the group of genes with increased H3K27ac occupancy (Figure 2A), genes with decreased H3K27ac occupancy were more expressed in cardiac and skeletal muscle, including fetal heart (Figure 2B). Genes with increased H3K27ac occupancy have on average higher expression than the mean expression of all genes expressed in human adult heart while this was not the case for the hypoacetylated genes. PLN-specific fingerprint of chromatin regulation (Analysis 3) In order to exclude general pathways linked to end-stage heart failure, we included ChIP-seq data from patients with other genetic types of non-ischemic cardiomyopathy (n=6) and ischemic cardiomyopathy (n=4). K-mean analysis revealed PLN R14del-specific clusters of genes, among others, with PLN R14del-specific up- and down acetylation or clusters of peaks linked to cardiomyopathies in general (Figure S4). Next, we focused on the main clusters showing a PLN
8
Figure 1. Supervised clustering of differentially acetylated regions in PLN patients and healthy controls. The clustering is based on the mean of H3K27ac signal over differentially acetylated regions (p<0.05) in the heart.
133
PART THREE CHAPTER 8
R14del-specific pattern. We have identified two clusters with hyperacetylation (cluster 1 and 12) including 50 genes and two clusters with hypoacetylation (cluster 3 and 4) resulting in 158 genes in PLN R14del when compared to the other groups (Figure 3, Table 2). Interestingly, 13 genes from the hyperacetylated group and 55 genes from the hypoacetylated group are known to be involved in various pathways related to (phospho)lipid and lipoprotein synthesis and metabolism (Table 2).
Figure 2. Box plot representing tissue specificity of detected genes. Public RNA sequencing data were used to assess the tissue specificity of genes located in the proximity of hyperacetylated peaks (A) and hypoacetylated peaks (B). Only the regions differentially acetylated between healthy controls and PLN patients (p<0.05) are used in this analysis. The first column indicates the mean expression of all genes expressed in adult cardiac tissue. Columns 2-13 indicate the selection of genes based on this study. Y-axis represents log2 RPKM (reads per kilobase per million).
DISCUSSION This is the first study using ChIP-seq to assay disease-specific fingerprint of H3K27ac in cardiac tissue from PLN R14del end-stage cardiomyopathy to better understand the underlying pathophysiological mechanisms. We found differentially regulated gene regions in PLN R14del hearts mainly involved in fibrosis, (cardiac) development and chromatin assembly and PLN R14del cardiomyopathy specific pathways related to lipid metabolism. Epigenetic modification (DNA methylation and histone modifications) can make regulatory elements of genes more or less permissive to transcription factors.19 Several noncoding RNAs (e.g., microRNA and long noncoding RNA) are involved in the regulation of gene expression integrated in a network with epigenetic modifications and disregulation has been associated
134
CHROMATIN REGULATION IN CARDIOMYOPATHIES
8
Figure 3. Comparison PLN R14del vs. other types of cardiomyopathies. K-mean clustering of PLN R14del cardiomyopathy (1-6), healthy controls (7-10), ischemic cardiomyopathy (11-14) and sarcomeric non-ischemic cardiomyopathy (15-20) based on H3K27ac signal was user to partition the peaks into 12 different clusters. Selected were clusters considered to be specifically upregulated in PLN R14del (A) and downregulated in PLN R14del (B) as compared to other conditions. (n=number of detected peaks)
with cardiac failure. In a previous study from Yang et al. RNA sequencing detected differential expression in heart failure and long noncoding RNA (lncRNA) could distinguish ischemic cardiomyopathy from non-ischemic cardiomyopathy.20 Furthermore, pathological expression of lncNRAs improved in response to LVAD support. Herrer et al. showed altered cytoskeletal processes in human tissue RNA-Seq data of ischemic and dilated cardiomyopathy. They also found changes in MYLK4, RHOU, and ANKRD1 with qRT-PCR.21 The transcriptional activator p300 has been associated with hypertrophy and heart failure through transcription by myocyte enhancer factor-2 and GATA-4.22, 23 In addition, several stress related pathways have been implicated in DCM, including biomechanical, cytoskeletal, myocyte survival, apoptosis and sarcoplasmic reticulum calcium cycling pathways.24
135
PART THREE CHAPTER 8
Table 2. Genes specific for PLN R14del cardiomyopathy Hyperacetylated genes (cluster 1,12)
136
Gene symbol Gene name
Gene symbol Gene name
ABL1
ABL proto-oncogene 1, non-receptor tyrosine kinase
ITPRIPL2
inositol 1,4,5-trisphosphate receptor interacting protein-like 2
ARHGEF28
Rho guanine nucleotide exchange factor (GEF) 28
KCNE4
potassium channel, voltage gated subfamily E regulatory beta subunit 4
BFSP1
beaded filament structural protein 1, filensin
L3MBTL3
l(3)mbt-like 3 (Drosophila)
CCDC80
coiled-coil domain containing 80
LUM
lumican
CMKLR1
chemerin chemokine-like receptor 1
LY96
lymphocyte antigen 96
CNN3
calponin 3, acidic
MAML2
mastermind-like 2 (Drosophila)
COL14A1
collagen, type XIV, alpha 1
MMP2
matrix metallopeptidase 2
COL6A3
collagen, type VI, alpha 3
MSMO1
methylsterol monooxygenase 1
COL8A1
collagen, type VIII, alpha 1
NNMT
nicotinamide N-methyltransferase
CREB3L2
cAMP responsive element binding protein 3-like 2
NOD1
nucleotide-binding oligomerization domain containing 1
CYP1B1
cytochrome P450, family 1, subfamily B, polypeptide 1
NOX4
NADPH oxidase 4
EDNRA
endothelin receptor type A
NT5E
5â&#x20AC;&#x2122;-nucleotidase, ecto (CD73)
EMP1
epithelial membrane protein 1
PDGFD
platelet derived growth factor D
EPSTI1
epithelial stromal interaction 1 (breast)
PECAM1
platelet/endothelial cell adhesion molecule 1
ESR1
estrogen receptor 1
PPAP2A
phosphatidic acid phosphatase type 2A
FIBIN
fin bud initiation factor homolog (zebrafish)
PRICKLE2
prickle homolog 2 (Drosophila)
FLRT2
fibronectin leucine rich transmembrane protein 2
RNF138P1
ring finger protein 138, E3 ubiquitin protein ligase pseudogene 1
GLIPR1
GLI pathogenesis-related 1
SEC22B
SEC22 vesicle trafficking protein homolog B (S. cerevisiae) (gene/ pseudogene)
HIST1H1B
histone cluster 1, H1b
SLCO3A1
solute carrier organic anion transporter family, member 3A1
HIST1H2AM
histone cluster 1, H2am
SNORA74B
small nucleolar RNA, H/ACA box 74B
HIST1H2BO
histone cluster 1, H2bo
SNTB1
syntrophin, beta 1 (dystrophinassociated protein A1, 59kDa, basic component 1)
HIST1H3I
histone cluster 1, H3i
SPRY4
sprouty homolog 4 (Drosophila)
HIST1H3J
histone cluster 1, H3j
UACA
uveal autoantigen with coiled-coil domains and ankyrin repeats
HIST1H4L
histone cluster 1, H4l
WIPF1
WAS/WASL interacting protein family, member 1
IFFO2
intermediate filament family orphan 2
ZNF217
zinc finger protein 217
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Table 2. Continued Hypoacetylated genes (cluster 3,4) Gene symbol Gene name
Gene symbol Gene name
ABCB6
ATP-binding cassette, sub-family B (MDR/TAP), member 6 (Langereis blood group)
MADD
MAP-kinase activating death domain
ABHD6
abhydrolase domain containing 6
MAP4K1
mitogen-activated protein kinase kinase kinase kinase 1
ACBD4
acyl-CoA binding domain containing 4
MAPK7
mitogen-activated protein kinase 7
ADSSL1
adenylosuccinate synthase like 1
MB
myoglobin
AGPAT2
1-acylglycerol-3-phosphate O-acyltransferase 2
MFN2
mitofusin 2
ALS2CL
ALS2 C-terminal like
MFSD3
major facilitator superfamily domain containing 3
ANGPTL4
angiopoietin-like 4
MIR486-1
microRNA 486-1
ANK1
ankyrin 1, erythrocytic
MLPH
melanophilin
AP5S1
adaptor-related protein complex 5, sigma 1 subunit
MLYCD
malonyl-CoA decarboxylase
AQP1
aquaporin 1 (Colton blood group)
MRPL21
mitochondrial ribosomal protein L21
AQP7
aquaporin 7
MYEOV
myeloma overexpressed
ARAP1
ArfGAP with RhoGAP domain, ankyrin repeat and PH domain 1
MYH7
myosin, heavy chain 7, cardiac muscle, beta
ARHGAP23
Rho GTPase activating protein 23
MYL3
myosin, light chain 3, alkali; ventricular, skeletal, slow
ARHGAP27
Rho GTPase activating protein 27
MYL7
myosin, light chain 7, regulatory
ASAP3
ArfGAP with SH3 domain, ankyrin repeat and PH domain MYLK3 3
myosin light chain kinase 3
ASB18
ankyrin repeat and SOCS box containing 18
MYLPF
myosin light chain, phosphorylatable, fast skeletal muscle
ASS1
argininosuccinate synthase 1
MZT2A
mitotic spindle organizing protein 2A
ATG2A
autophagy related 2A
NCBP2
nuclear cap binding protein subunit 2, 20kDa
ATP6V1F
ATPase, H+ transporting, lysosomal 14kDa, V1 subunit F NEU1
sialidase 1 (lysosomal sialidase)
BAHD1
bromo adjacent homology domain containing 1
NMRK2
nicotinamide riboside kinase 2
BAP1
BRCA1 associated protein-1 (ubiquitin carboxy-terminal hydrolase)
NUDT8
nudix (nucleoside diphosphate linked moiety X)-type motif 8
BCL7C
B-cell CLL/lymphoma 7C
P2RY2
purinergic receptor P2Y, G-protein coupled, 2
BRICD5
BRICHOS domain containing 5
PACSIN3
protein kinase C and casein kinase substrate in neurons 3
C16orf86
chromosome 16 open reading frame 86
PBXIP1
pre-B-cell leukemia homeobox interacting protein 1
C1orf228
chromosome 1 open reading frame 228
PCBP4
poly(rC) binding protein 4
C1orf86
chromosome 1 open reading frame 86
PDCD6
programmed cell death 6
CABLES2
Cdk5 and Abl enzyme substrate 2
PDE4A
phosphodiesterase 4A, cAMP-specific
CASP2
caspase 2, apoptosis-related cysteine peptidase
PDK2
pyruvate dehydrogenase kinase, isozyme 2
CCDC57
coiled-coil domain containing 57
PGP
phosphoglycolate phosphatase
CDK18
cyclin-dependent kinase 18
PHF7
PHD finger protein 7
8
137
PART THREE CHAPTER 8
Table 2. Continued
138
CHAC1
ChaC glutathione-specific gammaglutamylcyclotransferase 1
PIGT
phosphatidylinositol glycan anchor biosynthesis, class T
CLIP2
CAP-GLY domain containing linker protein 2
PITPNM2
phosphatidylinositol transfer protein, membrane-associated 2
CLRN2
clarin 2
PLCD3
phospholipase C, delta 3
CLUH
clustered mitochondria (cluA/CLU1) homolog
PPAPDC3
phosphatidic acid phosphatase type 2 domain containing 3
CLUHP3
clustered mitochondria (cluA/CLU1) homolog pseudogene 3
PRR26
proline rich 26
CNIH4
cornichon family AMPA receptor auxiliary protein 4
PSME2
proteasome (prosome, macropain) activator subunit 2 (PA28 beta)
CRIP3
cysteine-rich protein 3
PTDSS2
phosphatidylserine synthase 2
CTF1
cardiotrophin 1
PTGDS
prostaglandin D2 synthase 21kDa (brain)
CTNNBIP1
catenin, beta interacting protein 1
PWWP2B
PWWP domain containing 2B
DES
desmin
PYROXD2
pyridine nucleotide-disulphide oxidoreductase domain 2
DGKZ
diacylglycerol kinase, zeta
QARS
glutaminyl-tRNA synthetase
DIRAS1
DIRAS family, GTP-binding RAS-like 1
QDPR
quinoid dihydropteridine reductase
DNAAF1
dynein, axonemal, assembly factor 1
RANBP1
RAN binding protein 1
DNHD1
dynein heavy chain domain 1
RGS3
regulator of G-protein signaling 3
DYSF
dysferlin
RNF31
ring finger protein 31
EIF3K
eukaryotic translation initiation factor 3, subunit K
SELENBP1
selenium binding protein 1
ELP6
elongator acetyltransferase complex subunit 6
SGSM2
small G protein signaling modulator 2
ENKD1
enkurin domain containing 1
SLC25A20
solute carrier family 25 (carnitine/ acylcarnitine translocase), member 20
FAHD1
fumarylacetoacetate hydrolase domain containing 1
SLC25A34
solute carrier family 25, member 34
FAM110D
family with sequence similarity 110, member D
SLC4A3
solute carrier family 4 (anion exchanger), member 3
FAM65A
family with sequence similarity 65, member A
SLCO2B1
solute carrier organic anion transporter family, member 2B1
FDXR
ferredoxin reductase
SPAG7
sperm associated antigen 7
FPGS
folylpolyglutamate synthase
SPAG8
sperm associated antigen 8
GBAS
glioblastoma amplified sequence
SRF
serum response factor (c-fos serum response element-binding transcription factor)
GPSM1
G-protein signaling modulator 1
STK11
serine/threonine kinase 11
GPT
glutamic-pyruvate transaminase (alanine aminotransferase)
STXBP6
syntaxin binding protein 6 (amisyn)
GTF3C2
general transcription factor IIIC, polypeptide 2, beta 110kDa
SYNPO2L
synaptopodin 2-like
HADHA
hydroxyacyl-CoA dehydrogenase/3-ketoacyl-CoA TBC1D10B thiolase/enoyl-CoA hydratase (trifunctional protein), alpha subunit
TBC1 domain family, member 10B
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Table 2. Continued HADHB
hydroxyacyl-CoA dehydrogenase/3-ketoacyl-CoA thiolase/enoyl-CoA hydratase (trifunctional protein), beta subunit
TINAGL1
tubulointerstitial nephritis antigen-like 1
HAGH
hydroxyacylglutathione hydrolase
TMEM120B
transmembrane protein 120B
HES4
hes family bHLH transcription factor 4
TMEM139
transmembrane protein 139
HINT2
histidine triad nucleotide binding protein 2
TMEM154
transmembrane protein 154
HRH2
histamine receptor H2
TMEM200B
transmembrane protein 200B
HSD17B1
hydroxysteroid (17-beta) dehydrogenase 1
TMEM229B
transmembrane protein 229B
HSDL1
hydroxysteroid dehydrogenase like 1
TMEM53
transmembrane protein 53
IGHMBP2
immunoglobulin mu binding protein 2
TMEM82
transmembrane protein 82
ILVBL
ilvB (bacterial acetolactate synthase)-like
TNFAIP8L1
tumor necrosis factor, alpha-induced protein 8-like 1
INPP5J
inositol polyphosphate-5-phosphatase J
TP73-AS1
TP73 antisense RNA 1
IPPK
inositol 1,3,4,5,6-pentakisphosphate 2-kinase
TRMT2A
tRNA methyltransferase 2 homolog A (S. cerevisiae)
KCNJ4
potassium channel, inwardly rectifying subfamily J, member 4
TSR1
TSR1, 20S rRNA accumulation, homolog (S. cerevisiae)
KLHDC8B
kelch domain containing 8B
TTYH3
tweety family member 3
KLHL42
kelch-like family member 42
TUB
tubby bipartite transcription factor
LAMB2
laminin, beta 2 (laminin S)
UQCRC1
ubiquinol-cytochrome c reductase core protein I
LCNL1
lipocalin-like 1
USP19
ubiquitin specific peptidase 19
LIN9
lin-9 DREAM MuvB core complex component
VWA7
von Willebrand factor A domain containing 7
LOC150776
sphingomyelin phosphodiesterase 4, neutral membrane (neutral sphingomyelinase-3) pseudogene
ZNF629
zinc finger protein 629
LOC645166
lymphocyte-specific protein 1 pseudogene
ZNF747
zinc finger protein 747
LPAR5
lysophosphatidic acid receptor 5
ZNF8
zinc finger protein 8
LRP5L
low density lipoprotein receptor-related protein 5-like
ZNRF3
zinc and ring finger 3
8
Bold letters indicate genes involved in pathways/networks related to lipid metabolism (GO:molecular function, GO:biological process, GO:cellular component, Human phenotype, Mouse phenotype, Pathway as based on the ToppGene Suite).
139
PART THREE CHAPTER 8
We detected several PLN R14del-specific genes with differential chromatin regulation. The KCNE4 gene encodes the potassium channel, voltage gated subfamily E regulatory beta subunit 4.25 Potassium channels are crucial for generation and propagation of action potentials in the heart and related to heart rate and arrhythmias. KCNE4 knockout mice showed significantly dilated left ventricular internal diameters and reduced fractional shortening and ejection fraction.26 COL14A1 encodes the alpha chain of collagen involved in fibrillogenesis27 and RNA-Seq analysis has previously identified differential expression of this gene in tissue samples of patients with ischemic cardiomyopathy and controls pointing to involvement of collagen-related genes in remodelling.28 MMP2 (matrix metalloprotease 2) is also related to collagen turnover and has been proposed as a biomarker to predict prognosis of patients with HF.29 The identified PLN R14del specific pathways related to lipid metabolism could be studied in future for candidate genes related to the characteristic fibrofatty replacement in PLN R14del hearts and arrhythmogenic cardiomyopathy.7, 30 The criteria for annotation of differentially acetylated peaks to potentially relevant genes have consequences on data interpretation. In this study we used a stringent 5 kb window to decrease the chance of false positive gene annotations. Interestingly, a hypoacetylated peak 5,036 bp from a TSS of MIR21 gene, a crucial player upregulated in cardiac fibrosis31, was identified in the PLN group. Given the orientation of the surrounding genes VMP1, TUBD1, and MIR21, the detected peak at chr17:57,922,150-57,925,229 is located downstream of the three genes and unlikely has a promoter function. The peak was detected as 223/958 sorted by significance (p=0.0038) and taking together the unknown role and low expression of the other two genes in cardiac tissue it indicates a possible role of an enhancer region of MIR21. Therefore we enlarged the window used for gene annotation from 5kb to 20kb from TSS and increased the number of resulting annotated genes from 225 to 416. When performing pathway analysis using ToppFun we increase the number of genes in biological processes, such as GO:0030198 (extracellular matrix organization) from 18 to 32 genes, in GO:0007507 (heart development) from 19 to 32 genes, and GO:0015629 (actin cytoskeleton) from 15 to 24 genes. This indicates that a right window size needs to be taken into account for in silico annotation of potentially relevant genes. However, a deeper functional annotation, such as using a chromatin conformation capture method32, is needed to prove the functional relationship between the differentially acetylated region and its regulated gene. Taken together, using an integrative chromatin analysis we identified the major effector genes and pathways involved in PLN R14del cardiomyopathy. Ubiquitously expressed genes with increased H3K27ac occupancy in patients are in line with general fibrotic pathways and developmental programs in end-stage organ failure. The suppression of mitochondrial lipid metabolism needed for muscle function is underlined by cardiac and skeletal muscle-specific genes with increased H3K27ac occupancy. The detection of PLN R14del-specific genes might provide valuable clues to recognize carriers at a high risk of developing cardiomyopathy using a plasma detectable biomarker.Â
140
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Acknowledgments This work was supported by “Stichting Genetische Hartspierziekte PLN” (http://stichtingpln.nl, Middenmeer, the Netherlands). Magdalena Harakalova is supported by NIH Ro1 grant LM010098 and the OZF/14 WKZ-funds. Michal Mokry is supported by OZF/2012 WKZ-fund. Folkert W. Asselbergs is supported by a Dekker scholarship-Junior Staff Member 2014T001 – Netherlands Heart Foundation and UCL Hospitals NIHR Biomedical Research Centre.
8
141
PART THREE CHAPTER 8
REFERENCES 1. 2.
3. 4.
5.
6.
7.
8.
9. 10. 11.
12. 13. 14. 15. 16.
17.
18.
19. 20.
21.
142
Mosterd A, Hoes AW. Clinical epidemiology of heart failure. Heart. 2007;93:1137-46. Arbustini E, Narula N, Dec GW, Reddy KS, Greenberg B, Kushwaha S, et al. The MOGE(S) classification for a phenotype-genotype nomenclature of cardiomyopathy: endorsed by the World Heart Federation. J Am Coll Cardiol. 2013;62:2046-72. Jacoby D, McKenna WJ. Genetics of inherited cardiomyopathy. Eur Heart J. 2012;33:296-304. Haghighi K, Kolokathis F, Gramolini AO, Waggoner JR, Pater L, Lynch RA, et al. A mutation in the human phospholamban gene, deleting arginine 14, results in lethal, hereditary cardiomyopathy. Proc Natl Acad Sci U S A. 2006;103:1388-93. van Rijsingen IA, van der Zwaag PA, Groeneweg JA, Nannenberg EA, Jongbloed JD, Zwinderman AH, et al. Outcome in phospholamban R14del carriers: results of a large multicentre cohort study. Circ Cardiovasc Genet. 2014;7:455-65. van der Zwaag PA, van Rijsingen IA, Asimaki A, Jongbloed JD, van Veldhuisen DJ, Wiesfeld AC, et al. Phospholamban R14del mutation in patients diagnosed with dilated cardiomyopathy or arrhythmogenic right ventricular cardiomyopathy: evidence supporting the concept of arrhythmogenic cardiomyopathy. Eur J Heart Fail. 2012;14:1199-207. Gho JM, van Es R, Stathonikos N, Harakalova M, te Rijdt WP, Suurmeijer AJ, et al. High resolution systematic digital histological quantification of cardiac fibrosis and adipose tissue in phospholamban p.Arg14del mutation associated cardiomyopathy. PLoS ONE. 2014;9:e94820. Karakikes I, Stillitano F, Nonnenmacher M, Tzimas C, Sanoudou D, Termglinchan V, et al. Correction of human phospholamban R14del mutation associated with cardiomyopathy using targeted nucleases and combination therapy. Nat Commun. 2015;6:6955. Levo M, Segal E. In pursuit of design principles of regulatory sequences. Nat Rev Genet. 2014;15:453-68. Kathiriya IS, Nora EP, Bruneau BG. Investigating the transcriptional control of cardiovascular development. Circ Res. 2015;116:700-14. Harakalova M, Mokry M, Hrdlickova B, Renkens I, Duran K, van Roekel H, et al. Multiplexed array-based and in-solution genomic enrichment for flexible and cost-effective targeted nextgeneration sequencing. Nat Protoc. 2011;6:1870-86. Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009;25:1754-60. Ji H, Jiang H, Ma W, Johnson DS, Myers RM, Wong WH. An integrated software system for analyzing ChIP-chip and ChIP-seq data. Nat Biotechnol. 2008;26:1293-300. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15:550. Chen J, Bardes EE, Aronow BJ, Jegga AG. ToppGene Suite for gene list enrichment analysis and candidate gene prioritization. Nucleic Acids Res. 2009;37:W305-11. Krupp M, Marquardt JU, Sahin U, Galle PR, Castle J, Teufel A. RNA-Seq Atlas--a reference database for gene expression profiling in normal tissue by next-generation sequencing. Bioinformatics. 2012;28:1184-5. Warde-Farley D, Donaldson SL, Comes O, Zuberi K, Badrawi R, Chao P, et al. The GeneMANIA prediction server: biological network integration for gene prioritization and predicting gene function. Nucleic Acids Res. 2010;38:W214-20. Harakalova M, Kummeling G, Sammani A, Linschoten M, Baas AF, van der Smagt J, et al. A systematic analysis of genetic dilated cardiomyopathy reveals numerous ubiquitously expressed and muscle-specific genes. Eur J Heart Fail. 2015;17:484-93. Greco CM, Condorelli G. Epigenetic modifications and noncoding RNAs in cardiac hypertrophy and failure. Nat Rev Cardiol. 2015. Yang KC, Yamada KA, Patel AY, Topkara VK, George I, Cheema FH, et al. Deep RNA sequencing reveals dynamic regulation of myocardial noncoding RNAs in failing human heart and remodeling with mechanical circulatory support. Circulation. 2014;129:1009-21. Herrer I, Rosello-Lleti E, Rivera M, Molina-Navarro MM, Tarazon E, Ortega A, et al. RNAsequencing analysis reveals new alterations in cardiomyocyte cytoskeletal genes in patients with heart failure. Lab Invest. 2014;94:645-53.
CHROMATIN REGULATION IN CARDIOMYOPATHIES
22. Morimoto T, Sunagawa Y, Kawamura T, Takaya T, Wada H, Nagasawa A, et al. The dietary compound curcumin inhibits p300 histone acetyltransferase activity and prevents heart failure in rats. J Clin Invest. 2008;118:868-78. 23. Yanazume T, Hasegawa K, Morimoto T, Kawamura T, Wada H, Matsumori A, et al. Cardiac p300 is involved in myocyte growth with decompensated heart failure. Mol Cell Biol. 2003;23:3593606. 24. Chien KR. Stress pathways and heart failure. Cell. 1999;98:555-8. 25. Abbott GW, Sesti F, Splawski I, Buck ME, Lehmann MH, Timothy KW, et al. MiRP1 forms IKr potassium channels with HERG and is associated with cardiac arrhythmia. Cell. 1999;97:17587. 26. Ciampa EJ. Investigating the Function of KCNE4 in Cardiac Physiology [Dissertation]. Nashville, Tennessee: Vanderbilt University; 2011. 27. Ansorge HL, Meng X, Zhang G, Veit G, Sun M, Klement JF, et al. Type XIV Collagen Regulates Fibrillogenesis: PREMATURE COLLAGEN FIBRIL GROWTH AND TISSUE DYSFUNCTION IN NULL MICE. J Biol Chem. 2009;284:8427-38. 28. Gil-Cayuela C, Rivera M, Ortega A, Tarazon E, Trivino JC, Lago F, et al. RNA sequencing analysis identifies new human collagen genes involved in cardiac remodeling. J Am Coll Cardiol. 2015;65:1265-7. 29. Sanchis L, Andrea R, Falces C, Llopis J, Morales-Ruiz M, Lopez-Sobrino T, et al. Prognosis of new-onset heart failure outpatients and collagen biomarkers. Eur J Clin Invest. 2015;45:842-9. 30. Marcus FI, McKenna WJ, Sherrill D, Basso C, Bauce B, Bluemke DA, et al. Diagnosis of arrhythmogenic right ventricular cardiomyopathy/dysplasia: proposed modification of the task force criteria. Circulation. 2010;121:1533-41. 31. Lorenzen JM, Schauerte C, Hubner A, Kolling M, Martino F, Scherf K, et al. Osteopontin is indispensible for AP1-mediated angiotensin II-related miR-21 transcription during cardiac fibrosis. Eur Heart J. 2015. 32. Stadhouders R, Kolovos P, Brouwer R, Zuin J, van den Heuvel A, Kockx C, et al. Multiplexed chromosome conformation capture sequencing for rapid genome-scale high-resolution detection of long-range chromatin interactions. Nat Protoc. 2013;8:509-24.
8
143
PART THREE CHAPTER 8
SUPPLEMENTARY INFORMATION
Supplementary Figure 1. Histological slides derived from the same tissue region as used for ChIP-seq. The slides were stained with Massonâ&#x20AC;&#x2122;s trichrome. A. PLN R14del mutation patients heart sections (n = 6). B. Control heart sections (n = 4). LV = left ventricle; Sept = Septum.
144
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Supplementary Figure 2. Schematic overview of mean fibrosis and adipose tissue in heart slices from hearts used for ChIP-seq. The results of mean percentage of fibrosis or adipose tissue are shown using a color scale. A-B. heart slices of PLN R14del mutation patients (n = 6/6). C-D. heart slices of controls (n = 3/4). LV = left ventricle; RV = right ventricle; Post. = Posterior; Ant. = Anterior.
8
Supplementary Figure 3. Supervised clustering of differentially acetylated regions in PLN patients and healthy controls. The clustering is based on the mean of H3K27ac signal over differentially acetylated regions (p<0.05) in the heart.
145
Supplementary Figure 4. K-mean clustering of PLN R14del cardiomyopathy (1-6), healthy controls (7-10), ischemic cardiomyopathy (11-14) and sarcomeric non-ischemic cardiomyopathy (15-20) based on H3K27ac signal was user to partition the peaks into 12 different clusters. Selected were clusters considered to be specifically upregulated in PLN R14del (A) and downregulated in PLN R14del (B) as compared to other conditions. (n=number of detected peaks)
PART THREE CHAPTER 8
146
road accident victim
DCM
DCM with marks of hypertrophy
F
F
F
M
F
M
M
M
M
F
NA
NA
NA
NA
M
M
F
M
M
F
24
33
26
6
12
17
NA
46
43
NA
NA
NA
NA
NA
20
24
46
55
31
29
505 174* 257 403
35 39
266
34 36
486
37
26
450 NA
52 61
645 315*
64 47
360 NA
NA NA
217 439
NA NA
390 495
43 34
NA 385
NA 51
293 452
37 50
Yes
Yes
No
Yes
No
Yes
No
Yes
No
Yes
NA
NA
NA
NA
Yes
Yes
Yes
Yes
Yes
Yes
Sex Age at Age at Heart ICD diagnosis explant weight (years) (years) (g)
No
No
No
No
No
Yes
Yes
Yes
No
No
NA
NA
NA
NA
No
No
No
Yes
Yes
No
Yes
No
Yes
No
No
No
No
No
Yes
No
NA
NA
NA
NA
Yes
No
No
Yes
Yes
No
Yes
No
No
No
No
No
No
No
No
No
NA
NA
NA
NA
No
No
No
No
No
Yes
No
No
NA
No
No
No
Yes
Yes
Yes
Yes
NA
NA
NA
NA
No
No
No
Yes
No
No
NA
19%
NA
13%
25%
21%
17%
20%
25%
18%
NA
NA
NA
NA
20%
NA
31%
15%
NA
25%
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
PLLV
SEPT
SEPT
PLLV
PLLV
SEPT
RZ
RZ
RZ
RZ
SEPT
SEPT
PLLV
PLLV
PLLV
SEPT
SEPT
SEPT
PLLV
PLLV
CRT-D LVAD Second Significant Last Schematic Location HTx CAD known EF overview1 of sample available before explant
1
Schematic overview of mean percentage of fibrosis or adipose tissue from transversal heart slice; *after removal of heart slice. Abbreviations: LVAD = left ventricular assist device; ICD = implantable cardioverter-defibrillator; CRT-D = cardiac resynchronization therapy defibrillator; HTx = heart transplantation; CAD = coronary artery disease; EF = ejection fraction; PLLV = posterolateral left ventricle; SEPT = septum; RZ = remote zone to myocardiac infarction; ICM = ischemic cardiomyopathy; HCM = hypertrophic cardiomyopathy; HOCM = hypertrophic obstructive cardiomyopathy; DCM = dilated cardiomyopathy; ACM = arrhythmogenic cardiomyopathy; F = female (Sex); M = male (Sex); NA = not available.
Sarcomeric
Sarcomeric
TTN_1
TTN_2
DCM with marks of hypertrophy
HOCM in a dilated phase
Sarcomeric
Sarcomeric
MYH7_2
HCM
MYH7_1
HCM in a dilated phase
ICM
ICM
ICM
ICM
rejected donor
MYBPC3_2 Sarcomeric
Ischemic
Ischemic
ICM_3
ICM_4
rejected donor
rejected donor
MYBPC3_1 Sarcomeric
Ischemic
Ischemic
ICM_1
ICM_2
Control
Control
Control_3
Control_4
Control
Control
Control_1
Control_2
Phospholamban ACM
Phospholamban ACM
PLN_5
PLN_6
Phospholamban DCM
Phospholamban DCM
PLN_3
PLN_4
Phospholamban DCM
Phospholamban DCM
PLN_1
Diagnosis
PLN_2
Group
Type
Supplementary Table 1A. Detailed clinical overview of the included individuals
CHROMATIN REGULATION IN CARDIOMYOPATHIES
8
147
148
Phospholamban
Phospholamban
Phospholamban
Phospholamban
Phospholamban
Phospholamban
Control
Control
Control
Control
Ischemic
Ischemic
Ischemic
Ischemic
Sarcomeric
Sarcomeric
Sarcomeric
Sarcomeric
Sarcomeric
Sarcomeric
PLN_1
PLN_2
PLN_3
PLN_4
PLN_5
PLN_6
Control_1
Control_2
Control_3
Control_4
ICM_1
ICM_2
ICM_3
ICM_4
MYBPC3_1
MYBPC3_2
MYH7_1
MYH7_2
TTN_1
TTN_2 TTN
5xTTN
MYH7
MYH7
MYBPC3
MYBPC3
NA
NA
NA
NA
NA
NA
NA
NA
PLN
PLN
PLN
PLN
PLN/MYH6*
PLN
Mutated gene
c.32041G>A
c.5905G>T, c.82004C>G, c.10733C>T, c.53744G>A, c.87949G>A
NA
c.704C>G
NA
c.2827C>T
NA
NA
NA
NA
NA
NA
NA
NA
c.40_42delAGA
c.40_42delAGA
c.40_42delAGA
c.40_42delAGA
c.40_42delAGA
c.40_42delAGA
CDS change
p.Val10681Ile
p.Asp1969Tyr, p.Thr27335Ser, p.Pro3578Leu, p.Arg17915His, p.Ala29317Thr
NA
p.Thr235Ser
NA
p.Arg943X
NA
NA
NA
NA
NA
NA
NA
NA
p.Arg14del
p.Arg14del
p.Arg14del
p.Arg14del
p.Arg14del
p.Arg14del
Protein change
*variant of uncertain significance (VUS). Abbreviations: CDS = CoDing Sequence; NA = not available.
Group
Type
Supplementary Table 1B. Detailed clinical genetic overview of the included individuals
ACTC1, CRSP3, DES, GLA, LAMP2, LDB3, MYBPC3, MYH7, MYL2, NYL3, PLN, PRKAG2, SVNA, SGTC, STA, TAZ, TCHP, TNNC1, TNNI3, TNNT2, TPM1, TCL
MYH7, LMNA, PKP2, DSP, DSC2, DSG2, PLN
NA
NA
mutation detected in 1981
NA
NA
NA
NA
NA
NA
NA
NA
NA
PKP2
DSC2, DSP DSG2, JUP, TMEM43, PKP2
NA
NA
Cardiome 1.0 - 440 genes - research
NA
Negative DNA assay
PART THREE CHAPTER 8
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Supplementary Table 2. ChIP-seq libraries peak number and selection for Analysis 1-3 Sample ID
Number of detected peaks
Analysis 1
Analysis 2
Analysis 3
PLN_1
40003
+
+
+
PLN_2
30041
+
+
+
PLN_3
34430
+
+
+
PLN_4
27167
+
+
+
PLN_5
25300
+
+
+
PLN_6
11954
+
+
+
Mean±SD
28,149±9,538
Control_1
26032
+
+
+
Control_2
18731
+
+
+
Control_3
39941
+
+
+
Control_4
18178
+
+
+
Mean±SD
25,721±8,460
ICM_1
12613
-
-
+
ICM_2
25387
-
-
+
ICM_3
26081
-
-
+
ICM_4
46978
-
-
+
Mean±SD
27,765±8,139
MYBPC3_1
30736
-
-
+
MYBPC3_2
40574
-
-
+
MYH7_1
30374
-
-
+
MYH7_2
38189
-
-
+
TTN_1
25761
-
-
+
TTN_2
41838
-
-
+
Mean±SD
34,579±6,429
8
SD = standard deviation.
149
PART THREE CHAPTER 8
Supplementary Table 3. Annotation of genes with increased H3K27ac occupancy within 5kb from transcription start site (TSS)
150
Chromosome
Peak start
Peak end
Peak length
Distance to TSS
Distance to Gene TES Symbol
Gene Name
chr9
133708430
133713699
5269
234
-51993
ABL1
ABL proto-oncogene 1, non-receptor tyrosine kinase
chr11
34376480
34378599
2119
1262
-205005
ABTB2
ankyrin repeat and BTB (POZ) domain containing 2
chr5
156997500
156999629
2129
4203
-94252
ADAM19
ADAM metallopeptidase domain 19
chr15
89164200
89166449
2249
798
-10187
AEN
apoptosis enhancing nuclease
chr2
100720930
100723029
2099
65
-558262
AFF3
AF4/FMR2 family, member 3
chr6
151560030
151563649
3619
706
-117852
AKAP12
A kinase (PRKA) anchor protein 12
chr3
105087030
105089199
2169
2402
-207628
ALCAM
activated leukocyte cell adhesion molecule
chr11
94499880
94504049
4169
457
-107952
AMOTL1
angiomotin like 1
chr22
38243750
38246779
3029
-4962
-18402
ANKRD54 ankyrin repeat domain 54
chr12
45608280
45612229
3949
379
-223932
ANO6
anoctamin 6
chr15
60682100
60690599
8499
3835
-46999
ANXA2
annexin A2
chr5
72920350
72923599
3249
-8
-285552
ARHGEF28
Rho guanine nucleotide exchange factor (GEF) 28
chr4
114898700
114901699
2999
678
-78760
ARSJ
arylsulfatase family, member J
chr12
54068250
54072479
4229
-256
-11420
ATP5G2
ATP synthase, H+ transporting, mitochondrial Fo complex, subunit C2 (subunit 9)
chr6
56818050
56820129
2079
-683
-73048
BEND6
BEN domain containing 6
chr20
17537400
17540699
3299
555
-64499
BFSP1
beaded filament structural protein 1, filensin
chr20
11870100
11873499
3399
323
-35442
BTBD3
BTB (POZ) domain containing 3
chr21
18982300
18985429
3129
1403
-17895
BTG3
BTG family, member 3
chr18
47017100
47019279
2179
-4589
-10160
C18orf32
chromosome 18 open reading frame 32
chr1
230993350
230996099
2749
-2943
-21859
C1orf198
chromosome 1 open reading frame 198
chr12
7167130
7169999
2869
585
-9768
C1S
complement component 1, s subcomponent
chr9
89762080
89765099
3019
31
-11051
C9orf170
chromosome 9 open reading frame 170
chr9
117372150
117375429
3279
84
-34906
C9orf91
chromosome 9 open reading frame 91
chr10
115437500
115441529
4029
89
-51147
CASP7
caspase 7, apoptosis-related cysteine peptidase
chr2
202121780
202124479
2699
-2093
-29304
CASP8
caspase 8, apoptosis-related cysteine peptidase
chr3
112358550
112361749
3199
-173
-36739
CCDC80
coiled-coil domain containing 80
chr19
41813200
41817729
4529
-629
-15322
CCDC97
coiled-coil domain containing 97
chr9
21993950
21996999
3049
-985
-27723
CDKN2A
cyclin-dependent kinase inhibitor 2A
chr9
21993950
21996999
3049
685
-125616
CDKN2BAS1
CDKN2B antisense RNA 1
chr10
106111880
106114349
2469
-407
-101724
CFAP58
cilia and flagella associated protein 58
chr7
89873400
89876049
2649
237
-65650
CFAP69
cilia and flagella associated protein 69
chr4
54928400
54934199
5799
-512
-55342
CHIC2
cysteine-rich hydrophobic domain 2
chr3
87275130
87278149
3019
227
-28057
CHMP2B
charged multivesicular body protein 2B
chr11
123063850
123067399
3549
382
-122590
CLMP
CXADR-like membrane protein
chr12
108730630
108733129
2499
1214
-50058
CMKLR1
chemerin chemokine-like receptor 1
chr1
95388530
95396299
7769
320
-29906
CNN3
calponin 3, acidic
chr6
75911500
75917579
6079
1083
-120497
COL12A1
collagen, type XII, alpha 1
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Supplementary Table 3. Continued chr8
121135930
121138849
2919
38
-246876
COL14A1
collagen, type XIV, alpha 1
chr17
48281500
48284299
2799
-3900
-21441
COL1A1
collagen, type I, alpha 1
chr2
238319130
238323499
4369
1535
-88660
COL6A3
collagen, type VI, alpha 3
chr3
99355480
99361399
5919
986
-156718
COL8A1
collagen, type VIII, alpha 1
chr7
137675900
137688499
12599
4646
-122473
CREB3L2
cAMP responsive element binding protein 3-like 2
chr16
84853480
84859029
5549
2668
-86861
CRISPLD2 cysteine-rich secretory protein LCCL domain containing 2
chr11
118757180
118762149
4969
-4436
-7305
CXCR5
chemokine (C-X-C motif) receptor 5
chr2
38299030
38302099
3069
2758
-5819
CYP1B1
cytochrome P450, family 1, subfamily B, polypeptide 1
chr5
39420680
39425729
5049
2130
-51425
DAB2
Dab, mitogen-responsive phosphoprotein, homolog 2 (Drosophila)
chr1
85928750
85932249
3499
389
-146331
DDAH1
dimethylarginine dimethylaminohydrolase 1
chr18
67066750
67069899
3149
34
-447500
DOK6
docking protein 6
chr1
168695080
168697729
2649
2023
-31698
DPT
dermatopontin
chr6
116690650
116693799
3149
115
-67215
DSE
dermatan sulfate epimerase
chr18
65183480
65186129
2649
-838
-10986
DSEL
dermatan sulfate epimerase-like
chr6
56818050
56820129
2079
323
-496303
DST
dystonin
chr2
44000530
44002779
2249
473
-20657
DYNC2LI1 dynein, cytoplasmic 2, light intermediate chain 1
chr20
33102250
33105699
3449
-229
-24786
DYNLRB1 dynein, light chain, roadblock-type 1
chr8
25895730
25903479
7749
2787
-198032
EBF2
early B-cell factor 2
chr6
12291000
12293599
2599
1771
-5126
EDN1
endothelin 1
chr4
148401030
148405129
4099
1011
-63026
EDNRA
endothelin receptor type A
chr22
38243750
38246779
3029
-105
-39372
EIF3L
eukaryotic translation initiation factor 3, subunit L
chr19
18629100
18634699
5599
1037
-78425
ELL
elongation factor RNA polymerase II
chr12
13348000
13354999
6999
1898
-18207
EMP1
epithelial membrane protein 1
chr13
43564880
43568299
3419
-213
-104468
EPSTI1
epithelial stromal interaction 1 (breast)
chr6
152011080
152013599
2519
709
-412066
ESR1
estrogen receptor 1
chr5
76010400
76015449
5049
1057
-18669
F2R
coagulation factor II (thrombin) receptor
chr2
187558150
187560249
2099
411
-69310
FAM171B
family with sequence similarity 171, member B
chr10
120100350
120102399
2049
464
-32803
FAM204A
family with sequence similarity 204, member A
chr2
106011650
106017129
5479
1185
-37105
FHL2
four and a half LIM domains 2
chr11
27013400
27018629
5229
387
-2615
FIBIN
fin bud initiation factor homolog (zebrafish)
chr14
85994900
85997029
2129
-523
-98304
FLRT2
fibronectin leucine rich transmembrane protein 2
chr12
50096950
50100929
3979
2257
-67216
FMNL3
formin-like 3
chr16
86607800
86609849
2049
-3290
-6478
FOXL1
forkhead box L1
chr1
100233150
100235199
2049
-2826
-59916
FRRS1
ferric-chelate reductase 1
chr5
52774380
52779599
5219
395
-4912
FST
follistatin
chr17
9936450
9940679
4229
1499
-124639
GAS7
growth arrest-specific 7
chr15
45668850
45670949
2099
1080
-16576
GATM
glycine amidinotransferase (Larginine:glycine amidinotransferase)
chr5
154319300
154322049
2749
-2899
-53699
GEMIN5
gem (nuclear organelle) associated protein 5
chr12
75872950
75876599
3649
262
-20940
GLIPR1
GLI pathogenesis-related 1
8
151
PART THREE CHAPTER 8
Supplementary Table 3. Continued
152
chr19
47141050
47143379
2329
-4276
-4881
GNG8
guanine nucleotide binding protein (G protein), gamma 8
chr7
50857150
50863399
6249
884
-202515
GRB10
growth factor receptor-bound protein 10
chr19
48900500
48902499
1999
3368
-46687
GRIN2D
glutamate receptor, ionotropic, N-methyl D-aspartate 2D
chr4
54968530
54970579
2049
3307
1433
GSX2
GS homeobox 2
chr1
40096930
40107599
10669
3083
-13161
HEYL
hes-related family bHLH transcription factor with YRPW motif-like
chr6
27838500
27841899
3399
-4841
-5629
HIST1H1B histone cluster 1, H1b
chr6
27112480
27116399
3919
-468
-905
HIST1H2AH
histone cluster 1, H2ah
chr6
27803550
27808499
4949
92
-366
HIST1H2AK
histone cluster 1, H2ak
chr6
27861130
27864299
3169
-1752
-2238
HIST1H2AM
histone cluster 1, H2am
chr6
26271350
26275549
4199
246
-189
HIST1H2BI
histone cluster 1, H2bi
chr6
27112480
27116399
3919
179
-8367
HIST1H2BK
histone cluster 1, H2bk
chr6
27803550
27808499
4949
-415
-862
HIST1H2BN
histone cluster 1, H2bn
chr6
27861130
27864299
3169
1512
1046
HIST1H2BO
histone cluster 1, H2bo
chr6
26271350
26275549
4199
-1838
-2303
HIST1H3G histone cluster 1, H3g
chr6
27838500
27841899
3399
-101
-576
HIST1H3I
chr6
27861130
27864299
3169
-4145
-4621
HIST1H3J histone cluster 1, H3j
chr6
27838500
27841899
3399
1089
727
HIST1H4L histone cluster 1, H4l
chr6
26535750
26540379
4629
-507
-9097
HMGN4
high mobility group nucleosomal binding domain 4
chr1
19280300
19282649
2349
1351
-50701
IFFO2
intermediate filament family orphan 2
chr10
91172850
91176549
3699
375
-6058
IFIT5
interferon-induced protein with tetratricopeptide repeats 5
chr9
27527050
27529899
2849
4163
1981
IFNK
interferon, kappa
chr7
45959050
45961079
2029
806
-8215
IGFBP3
insulin-like growth factor binding protein 3
chr7
41738030
41742429
4399
2476
-11627
INHBA
inhibin, beta A
chr12
66580200
66586079
5879
162
-65252
IRAK3
interleukin-1 receptor-associated kinase 3
chr16
19126650
19132299
5649
4221
-3473
ITPRIPL2
inositol 1,4,5-trisphosphate receptor interacting protein-like 2
chr2
223914350
223923129
8779
1878
-1613
KCNE4
potassium channel, voltage gated subfamily E regulatory beta subunit 4
chr20
36886500
36889449
2949
1199
-49068
KIAA1755 KIAA1755
chr4
38663350
38677349
13999
4560
-32778
KLF3
Kruppel-like factor 3 (basic)
chr1
32572400
32575499
3099
306
-68216
KPNA6
karyopherin alpha 6 (importin alpha 7)
chr6
130339150
130343049
3899
1366
-121484
L3MBTL3
l(3)mbt-like 3 (Drosophila)
chr7
107639600
107645049
5449
1479
-78079
LAMB1
laminin, beta 1
chr6
25279050
25281649
2599
702
-340406
LRRC16A
leucine rich repeat containing 16A
chr17
45907450
45909549
2099
-600
-6190
LRRC46
leucine rich repeat containing 46
chr15
101457850
101460099
2249
-485
-151340
LRRK1
leucine-rich repeat kinase 1
chr12
91501150
91506449
5299
1742
-6567
LUM
lumican
chr8
74902130
74905999
3869
478
-37240
LY96
lymphocyte antigen 96
chr3
184432950
184435649
2699
-4464
-6144
MAGEF1
melanoma antigen family F1
chr11
96070200
96077249
7049
2619
-362285
MAML2
mastermind-like 2 (Drosophila)
histone cluster 1, H3i
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Supplementary Table 3. Continued chr2
210442980
210444999
2019
-413
-154844
MAP2
microtubule-associated protein 2
chr3
187008000
187012429
4429
-463
-74277
MASP1
mannan-binding lectin serine peptidase 1 (C4/C2 activating component of Rareactive factor)
chr1
46267830
46271029
3199
145
-232367
MAST2
microtubule associated serine/threonine kinase 2
chr18
47806350
47808629
2279
654
-12273
MBD1
methyl-CpG binding domain protein 1
chr18
74839080
74844599
5519
2934
-151050
MBP
myelin basic protein
chr12
116713380
116717199
3819
-299
-318907
MED13L
mediator complex subunit 13-like
chr12
31810630
31813549
2919
-519
-9924
METTL20
methyltransferase like 20
chr2
170679980
170682099
2119
313
-12770
METTL5
methyltransferase like 5
chr6
31461050
31466129
5079
-2265
-15308
MICB
MHC class I polypeptide-related sequence B
chr21
17959050
17961729
2679
-2167
-2254
MIR125B2 microRNA 125b-2
chr22
38243750
38246779
3029
-4887
-4986
MIR658
microRNA 658
chr22
38243750
38246779
3029
-1484
-1580
MIR659
microRNA 659
chr16
55508150
55511249
3099
-3381
-30884
MMP2
matrix metallopeptidase 2
chr9
27527050
27529899
2849
1375
-203267
MOB3B
MOB kinase activator 3B
chr17
16943400
16952699
9299
1943
-140820
MPRIP
myosin phosphatase Rho interacting protein
chr17
45907450
45909549
2099
403
-7861
MRPL10
mitochondrial ribosomal protein L10
chr5
154319300
154322049
2749
42
-28295
MRPL22
mitochondrial ribosomal protein L22
chr2
17993530
17995749
2219
-3146
-3726
MSGN1
mesogenin 1
chr4
166247200
166250149
2949
-143
-15549
MSMO1
methylsterol monooxygenase 1
chr5
79286750
79288749
1999
-662
-15209
MTX3
metaxin 3
chr8
128746080
128752599
6519
1025
-4338
MYC
v-myc avian myelocytomatosis viral oncogene homolog
chr20
50176580
50181229
4649
263
-171139
NFATC2
nuclear factor of activated T-cells, cytoplasmic, calcineurin-dependent 2
chr11
114161600
114169679
8079
-895
-17598
NNMT
nicotinamide N-methyltransferase
chr7
30506700
30521899
15199
4093
-50156
NOD1
nucleotide-binding oligomerization domain containing 1
chr14
27065800
27068999
3199
-440
-152310
NOVA1
neuro-oncological ventral antigen 1
chr11
89223200
89225999
2799
53
-167076
NOX4
NADPH oxidase 4
chr6
86158280
86161829
3549
753
-45441
NT5E
5â&#x20AC;&#x2122;-nucleotidase, ecto (CD73)
chr1
114520880
114524579
3699
700
-2145
OLFML3
olfactomedin-like 3
chr12
56210030
56213579
3549
-1
-3154
ORMDL2
ORMDL sphingolipid biosynthesis regulator 2
chr2
19562050
19564099
2049
-4703
-11828
OSR1
odd-skipped related transciption factor 1
chr9
112541880
112545699
3819
1201
-390999
PALM2AKAP2
PALM2-AKAP2 readthrough
chr11
93860300
93871049
10749
3581
-49460
PANX1
pannexin 1
chr4
120546050
120550579
4529
127
-132763
PDE5A
phosphodiesterase 5A, cGMP-specific
chr11
104033350
104036349
2999
177
-256935
PDGFD
platelet derived growth factor D
chr4
186452830
186459649
6819
472
-33388
PDLIM3
PDZ and LIM domain 3
chr17
62403900
62407599
3699
-4545
-5886
PECAM1
platelet/endothelial cell adhesion molecule 1
chr12
76421480
76425399
3919
2116
-4212
PHLDA1
pleckstrin homology-like domain, family A, member 1
chr15
68346550
68348779
2229
1093
-132737
PIAS1
protein inhibitor of activated STAT, 1
chr4
88928730
88930749
2019
920
-69188
PKD2
polycystic kidney disease 2 (autosomal dominant)
chr10
95750980
95752979
1999
-1766
-336166
PLCE1
phospholipase C, epsilon 1
8
153
PART THREE CHAPTER 8
Supplementary Table 3. Continued
154
chr8
38756580
38768499
11919
3787
-68888
PLEKHA2
pleckstrin homology domain containing, family A (phosphoinositide binding specific) member 2
chr12
45608280
45612229
3949
-466
-43408
PLEKHA8P1
pleckstrin homology domain containing, family A member 8 pseudogene 1
chr1
145609600
145612399
2799
-116
-18394
POLR3C
polymerase (RNA) III (DNA directed) polypeptide C (62kD)
chr5
54827750
54832679
4929
658
-109530
PPAP2A
phosphatidic acid phosphatase type 2A
chr4
81104650
81107449
2799
-374
-19430
PRDM8
PR domain containing 8
chr3
64205650
64212849
7199
1881
-129723
PRICKLE2 prickle homolog 2 (Drosophila)
chr2
219695200
219697229
2029
297
-9109
PRKAG3
protein kinase, AMP-activated, gamma 3 non-catalytic subunit
chr3
93691200
93693999
2799
334
-100718
PROS1
protein S (alpha)
chr1
214160280
214164029
3749
295
-47605
PROX1
prospero homeobox 1
chr10
129844950
129847429
2479
377
-37972
PTPRE
protein tyrosine phosphatase, receptor type, E
chr1
31536250
31538879
2629
999
-133212
PUM1
pumilio RNA-binding family member 1
chr2
37570400
37572599
2199
-253
-28964
QPCT
glutaminyl-peptide cyclotransferase
chr18
9706200
9710449
4249
97
-154228
RAB31
RAB31, member RAS oncogene family
chr7
4918480
4920479
1999
3855
-80740
RADIL
Ras association and DIL domains
chr15
93615750
93617799
2049
-386
-30138
RGMA
repulsive guidance molecule family member a
chr20
19866100
19872549
6449
-885
-113775
RIN2
Ras and Rab interactor 2
chr1
182556350
182559799
3449
-1967
-15303
RNASEL
ribonuclease L (2â&#x20AC;&#x2122;,5â&#x20AC;&#x2122;-oligoisoadenylate synthetase-dependent)
chr1
145609600
145612399
2799
-36
-77774
RNF115
ring finger protein 115
chr5
54827750
54832679
4929
155
-5545
RNF138P1 ring finger protein 138, E3 ubiquitin protein ligase pseudogene 1
chr18
47017100
47019279
2179
716
-3331
RPL17
ribosomal protein L17
chr4
152020180
152022349
2169
511
-4537
RPS3A
ribosomal protein S3A
chr21
36256180
36261749
5569
2022
-98866
RUNX1
runt-related transcription factor 1
chr1
165412150
165415149
2999
780
-43299
RXRG
retinoid X receptor, gamma
chr12
56210030
56213579
3549
-265
-65558
SARNP
SAP domain containing ribonucleoprotein
chr1
145091450
145093499
2049
-3932
-24447
SEC22B
SEC22 vesicle trafficking protein homolog B (S. cerevisiae) (gene/pseudogene)
chr6
2845930
2848029
2099
-4899
-13244
SERPINB1 serpin peptidase inhibitor, clade B (ovalbumin), member 1
chr18
42260150
42263649
3499
1037
-386573
SETBP1
SET binding protein 1
chr10
10837450
10842499
5049
-3098
-13573
SFTA1P
surfactant associated 1, pseudogene
chr19
16934930
16936999
2069
-4253
-55199
SIN3B
SIN3 transcription regulator family member B
chr20
1873280
1881199
7919
1414
-43299
SIRPA
signal-regulatory protein alpha
chr14
61190150
61192249
2099
-408
-14944
SIX4
SIX homeobox 4
chr1
211752550
211756379
3829
-2366
-6084
SLC30A1
solute carrier family 30 (zinc transporter), member 1
chr3
14442430
14448579
6149
1399
-44186
SLC6A6
solute carrier family 6 (neurotransmitter transporter), member 6
chr15
92393230
92401849
8619
602
-311597
SLCO3A1
solute carrier organic anion transporter family, member 3A1
chr17
33568050
33572649
4599
264
-24404
SLFN5
schlafen family member 5
chr9
72870450
72876579
6129
-363
-96274
SMC5
structural maintenance of chromosomes 5
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Supplementary Table 3. Continued chr17
62656680
62659999
3319
46
-117605
SMURF2
SMAD specific E3 ubiquitin protein ligase 2
chr5
121647450
121649529
2079
670
-151304
SNCAIP
synuclein, alpha interacting protein
chr5
172442450
172446299
3849
-3354
-3557
SNORA74B
small nucleolar RNA, H/ACA box 74B
chr18
47017100
47019279
2179
-473
-537
SNORD58A
small nucleolar RNA, C/D box 58A
chr18
47017100
47019279
2179
-91
-156
SNORD58B
small nucleolar RNA, C/D box 58B
chr18
47017100
47019279
2179
-2496
-2585
SNORD58C
small nucleolar RNA, C/D box 58C
chr4
152020180
152022349
2169
-3714
-3778
SNORD73A
small nucleolar RNA, C/D box 73A
chr8
121820400
121825749
5349
1234
-275089
SNTB1
syntrophin, beta 1 (dystrophin-associated protein A1, 59kDa, basic component 1)
chr12
93964430
93969879
5449
3557
-2823
SOCS2
suppressor of cytokine signaling 2
chr5
141695980
141708449
12469
2405
-12223
SPRY4
sprouty homolog 4 (Drosophila)
chr14
35449630
35453499
3869
-539
-47206
SRP54
signal recognition particle 54kDa
chr12
26348100
26351849
3749
1469
-37731
SSPN
sarcospan
chr8
70403850
70407399
3549
597
-167522
SULF1
sulfatase 1
chr19
39934750
39937949
3199
62
-30956
SUPT5H
suppressor of Ty 5 homolog (S. cerevisiae)
chr9
113341100
113343129
2029
45
-214584
SVEP1
sushi, von Willebrand factor type A, EGF and pentraxin domain containing 1
chr10
30021930
30027449
5519
40
-278413
SVIL
supervillin
chr8
38641650
38646629
4979
-582
-66405
TACC1
transforming, acidic coiled-coil containing protein 1
chr6
134208730
134213179
4449
695
-2436
TCF21
transcription factor 21
chr1
218517250
218523249
5999
859
-97709
TGFB2
transforming growth factor, beta 2
chr20
30696380
30699129
2749
446
-57306
TM9SF4
transmembrane 9 superfamily protein member 4
chr8
116677100
116682699
5599
1328
-259175
TRPS1
trichorhinophalangeal syndrome I
chr15
71059230
71062099
2869
-4815
-113770
UACA
uveal autoantigen with coiled-coil domains and ankyrin repeats
chr2
170679980
170682099
2119
-2978
-259597
UBR3
ubiquitin protein ligase E3 component n-recognin 3 (putative)
chr1
165795450
165799879
4429
775
-79672
UCK2
uridine-cytidine kinase 2
chr12
109535600
109537749
2149
668
-12123
UNG
uracil-DNA glycosylase
chr7
48124050
48131299
7249
-1071
-20654
UPP1
uridine phosphorylase 1
chr5
82769000
82771779
2779
2897
-107732
VCAN
versican
chr10
104531680
104547499
15819
3702
-36424
WBP1L
WW domain binding protein 1-like
chr2
175497100
175500379
3279
567
-74437
WIPF1
WAS/WASL interacting protein family, member 1
chr1
37936150
37946979
10829
1446
-8411
ZC3H12A
zinc finger CCCH-type containing 12A
chr16
73075550
73094829
19279
-2916
-268402
ZFHX3
zinc finger homeobox 3
chr6
28108700
28110749
2049
9
-15511
ZKSCAN8 zinc finger with KRAB and SCAN domains 8
chr20
52201150
52203849
2699
-2864
-18888
ZNF217
zinc finger protein 217
chr16
49858550
49863349
4799
-32
-336428
ZNF423
zinc finger protein 423
chr7
63505180
63507199
2019
369
-32735
ZNF727
zinc finger protein 727
8
TSS = transcription start site; TES = transcription end site.
155
156
ID
GO:0030199
GO:0030198
GO:0043062
GO:0072358
GO:0072359
GO:0048584
GO:0051094
GO:0072075
GO:0060537
GO:0009888
GO:0031497
GO:0072074
GO:0007507
GO:0030335
GO:2000147
GO:0006334
GO:0051272
GO:0071822
GO:0065004
GO:0040017
GO:0006333
GO:0048729
GO:0009967
GO:2000145
GO:0043933
GO:0010647
GO:0006357
GO:0014706
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
GO: Biological Process
striated muscle tissue development
regulation of transcription from RNA polymerase II promoter
positive regulation of cell communication
macromolecular complex subunit organization
regulation of cell motility
positive regulation of signal transduction
tissue morphogenesis
chromatin assembly or disassembly
positive regulation of locomotion
protein-DNA complex assembly
protein complex subunit organization
positive regulation of cellular component movement
nucleosome assembly
positive regulation of cell motility
positive regulation of cell migration
heart development
kidney mesenchyme development
chromatin assembly
tissue development
muscle tissue development
metanephric mesenchyme development
positive regulation of developmental process
positive regulation of response to stimulus
circulatory system development
cardiovascular system development
extracellular structure organization
extracellular matrix organization
collagen fibril organization
Name
1,42E-5
1,35E-5
1,28E-5
1,27E-5
9,84E-6
9,25E-6
8,24E-6
7,35E-6
7,16E-6
6,62E-6
5,81E-6
5,57E-6
5,33E-6
3,98E-6
3,17E-6
2,11E-6
1,80E-6
1,76E-6
1,59E-6
1,30E-6
4,83E-7
4,61E-7
1,48E-7
1,20E-7
1,20E-7
8,79E-8
8,44E-8
1,62E-8
p-value
4,84E-2
4,61E-2
4,37E-2
4,34E-2
3,35E-2
3,15E-2
2,81E-2
2,51E-2
2,44E-2
2,26E-2
1,98E-2
1,90E-2
1,82E-2
1,36E-2
1,08E-2
7,18E-3
6,14E-3
6,00E-3
5,41E-3
4,43E-3
1,65E-3
1,57E-3
5,05E-4
4,08E-4
4,08E-4
3,00E-4
2,88E-4
5,53E-5
q-value Bonferroni
1,73E-3
1,71E-3
1,68E-3
1,68E-3
1,40E-3
1,37E-3
1,28E-3
1,19E-3
1,19E-3
1,19E-3
1,10E-3
1,10E-3
1,10E-3
9,05E-4
7,70E-4
5,52E-4
5,12E-4
5,12E-4
5,12E-4
4,92E-4
2,06E-4
2,06E-4
8,42E-5
8,16E-5
8,16E-5
8,16E-5
8,16E-5
5,53E-5
q-value FDR B&H
1,51E-2
1,49E-2
1,46E-2
1,46E-2
1,22E-2
1,19E-2
1,11E-2
1,04E-2
1,04E-2
1,03E-2
9,58E-3
9,58E-3
9,58E-3
7,88E-3
6,71E-3
4,81E-3
4,46E-3
4,46E-3
4,46E-3
4,28E-3
1,79E-3
1,79E-3
7,33E-4
7,11E-4
7,11E-4
7,11E-4
7,11E-4
4,81E-4
q-value FDR B&Y
15
38
31
40
20
30
20
11
15
11
38
15
10
15
15
19
5
11
43
17
5
29
42
30
30
19
19
8
Genes from Input
357
1629
1196
1752
579
1118
572
182
337
180
1570
330
143
321
315
476
19
157
1794
374
15
911
1584
905
905
386
385
41
Genes in Annotation
development
chromatin assembly
fibrosis
chromatin assembly
fibrosis
fibrosis
development
chromatin assembly
fibrosis
chromatin assembly
chromatin assembly
fibrosis
chromatin assembly
fibrosis
fibrosis
development
development
chromatin assembly
development
development
development
development
fibrosis
development
development
fibrosis
fibrosis
fibrosis
Concluding remark
Supplementary Table 4. Gene enrichment analysis related to genes with increased H3K27ac occupancy within 5kb from transcription start site (TSS)
PART THREE CHAPTER 8
GO:0044420
GO:0005578
GO:0031012
GO:0015629
1
2
3
4
KEGG: 585563
KEGG: 83122
REACTOME: 106552
REACTOME: 106555
REACTOME: 905996
REACTOME: 366238
REACTOME: 205242
REACTOME: 106551
REACTOME: 106550
REACTOME: 905993
REACTOME: 160948
1
2
3
4
5
6
7
8
9
10
11
7,93E-4 2,24E-3 2,75E-2
q-value Bonferroni 6,64E-5 1,80E-4 1,24E-3 8,51E-3 1,01E-2 1,22E-2 1,45E-2 1,72E-2 2,02E-2 3,25E-2
2,16E-6 6,08E-6 7,48E-5
p-value 6,61E-8 1,79E-7 1,24E-6 8,47E-6 1,01E-5 1,21E-5 1,44E-5 1,71E-5 2,01E-5 3,23E-5 4,40E-5
Oxidative Stress Induced Senescence
RNA Polymerase I, RNA Polymerase III, and Mitochondrial Transcription
RNA Polymerase I Transcription
RNA Polymerase I Promoter Clearance
Meiotic Recombination
Amyloids
Senescence-Associated Secretory Phenotype (SASP)
RNA Polymerase I Chain Elongation
RNA Polymerase I Promoter Opening
Systemic lupus erythematosus
Alcoholism
Name
actin cytoskeleton
extracellular matrix
4,42E-2
1,70E-4
4,61E-7
proteinaceous extracellular matrix
extracellular matrix component
q-value Bonferroni
p-value
Name
4,02E-3
3,25E-3
2,25E-3
2,14E-3
2,07E-3
2,03E-3
2,02E-3
2,02E-3
4,14E-4
8,99E-5
6,64E-5
q-value FDR B&H
6,88E-3
7,46E-4
3,97E-4
1,70E-4
q-value FDR B&H
3,01E-2
2,43E-2
1,68E-2
1,61E-2
1,55E-2
1,52E-2
1,51E-2
1,51E-2
3,10E-3
6,73E-4
4,98E-4
q-value FDR B&Y
4,46E-2
4,84E-3
2,57E-3
1,10E-3
q-value FDR B&Y
9
9
8
8
8
8
9
8
8
12
14
Genes from Input
15
17
16
11
Genes from Input
GO = Gene Ontology; FDR B&H = False discovery rate, Benjamini and Hochberg concept; FDR B&Y = False discovery rate, Benjamini and Yekutieli concept.
ID
No.
Pathway
ID
No.
GO: Cellular Component
131
126
91
89
87
85
109
81
63
138
180
Genes in Annotation
415
422
348
138
Genes in Annotation
chromatin assembly
chromatin assembly
chromatin assembly
chromatin assembly
chromatin assembly
chromatin assembly
chromatin assembly
chromatin assembly
chromatin assembly
chromatin assembly
chromatin assembly
Concluding remark
fibrosis
fibrosis
fibrosis
fibrosis
Concluding remark
CHROMATIN REGULATION IN CARDIOMYOPATHIES
8
157
PART THREE CHAPTER 8
Supplementary Table 5. Annotation of genes with decreased H3K27ac occupancy within 5kb from transcription start site (TSS) Chromo足 Peak start some
158
Peak end
Peak length
Distance to TSS
Distance to TES
Gene Symbol
Gene Name
chr17
74439750
74449399
9649
-4858
-21623
AANAT
aralkylamine N-acetyltransferase
chr2
220081700
220083849
2149
897
-8281
ABCB6
ATP-binding cassette, sub-family B (MDR/TAP), member 6 (Langereis blood group)
chr3
183732100
183737229
5129
1062
-96939
ABCC5
ATP-binding cassette, sub-family C (CFTR/MRP), member 5
chr3
183902130
183905599
3469
2
-7930
ABCF3
ATP-binding cassette, sub-family F (GCN20), member 3
chr20
62487700
62491949
4249
-2741
-4516
ABHD16B
abhydrolase domain containing 16B
chr3
58221830
58225429
3599
371
-56829
ABHD6
abhydrolase domain containing 6
chr19
17412600
17418449
5849
-1243
-12584
ABHD8
abhydrolase domain containing 8 actin binding Rho activating protein
chr8
107780580
107784049
3469
157
-10603
ABRA
chr17
43203380
43211299
7919
-2627
-14201
ACBD4
acyl-CoA binding domain containing 4
chr22
41863350
41867199
3849
146
-59717
ACO2
aconitase 2, mitochondrial
chr19
39132800
39136779
3979
-3537
-86380
ACTN4
actinin, alpha 4
chr3
52015700
52019599
3899
87
-5567
ACY1
aminoacylase 1
chr19
50190100
50192799
2699
-492
-2797
ADM5
adrenomedullin 5 (putative)
chr22
24819650
24822699
3049
-2355
-17149
ADORA2A
adenosine A2a receptor
chr14
105192780
105194929
2149
-2373
-19790
ADSSL1
adenylosuccinate synthase like 1
chr12
58119150
58121199
2049
152
-1963
AGAP2-AS1
AGAP2 antisense RNA 1
chr9
139582850
139586629
3779
-2829
-17145
AGPAT2
1-acylglycerol-3-phosphate O-acyltransferase 2
chr21
45342180
45350899
8719
1261
-60934
AGPAT3
1-acylglycerol-3-phosphate O-acyltransferase 3
chr3
46730700
46739949
9249
-154
-24646
ALS2CL
ALS2 C-terminal like
chr9
140081500
140084979
3479
-183
-14004
ANAPC2
anaphase promoting complex subunit 2
chr19
8430330
8434329
3999
3319
-6927
ANGPTL4
angiopoietin-like 4
chr8
41511200
41530879
19679
1764
-10295
ANK1
ankyrin 1, erythrocytic
chr20
3798150
3802979
4829
-638
-5389
AP5S1
adaptor-related protein complex 5, sigma 1 subunit
chr11
57001650
57003999
2349
2102
-1771
APLNR
apelin receptor
chr7
30944450
30966699
22249
4107
-9556
AQP1
aquaporin 1 (Colton blood group)
chr9
33400630
33403099
2469
652
-16916
AQP7
aquaporin 7
chr11
72428900
72435899
6999
1003
-36285
ARAP1
ArfGAP with RhoGAP domain, ankyrin repeat and PH domain 1
chr17
36602800
36617799
14999
-3344
-58327
ARHGAP23
Rho GTPase activating protein 23
chr17
43485880
43490729
4849
-4856
-17036
ARHGAP27
Rho GTPase activating protein 27
chr8
1770700
1774149
3449
276
-134381
ARHGEF10
Rho guanine nucleotide exchange factor (GEF) 10
chr1
3371530
3374849
3319
2043
-24485
ARHGEF16
Rho guanine nucleotide exchange factor (GEF) 16
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Supplementary Table 5. Continued chr12
58001480
58008899
7419
-28
-5835
ARHGEF25
Rho guanine nucleotide exchange factor (GEF) 25
chr1
23804600
23812099
7499
2400
-53294
ASAP3
ArfGAP with SH3 domain, ankyrin repeat and PH domain 3
chr7
150884380
150886499
2119
-962
-12655
ASB10
ankyrin repeat and SOCS box containing 10
chr17
42247200
42250329
3129
691
-7686
ASB16
ankyrin repeat and SOCS box containing 16
chr2
237168450
237173579
5129
1973
-67499
ASB18
ankyrin repeat and SOCS box containing 18
chr17
79933380
79936049
2669
-711
-40565
ASPSCR1
alveolar soft part sarcoma chromosome region, candidate 1
chr9
133319850
133323549
3699
1606
-54960
ASS1
argininosuccinate synthase 1
chr11
64682930
64687299
4369
-393
-23110
ATG2A
autophagy related 2A
chr7
128498250
128501749
3499
-2898
-5902
ATP6V1F
ATPase, H+ transporting, lysosomal 14kDa, V1 subunit F
chr17
42277350
42279399
2049
-2846
-9201
ATXN7L3
ataxin 7-like 3
chr1
1307630
1312749
5119
372
-1079
AURKAIP1
aurora kinase A interacting protein 1
chr6
33243600
33248429
4829
1098
-589
B3GALT4
UDP-Gal:betaGlcNAc beta 1,3-galactosyltransferase, polypeptide 4
chr15
40730530
40733399
2869
-1446
-28476
BAHD1
bromo adjacent homology domain containing 1
chr17
79012250
79014249
1999
4303
-71062
BAIAP2
BAI1-associated protein 2
chr6
33544230
33549349
5119
1280
-6466
BAK1
BCL2-antagonist/killer 1
chr3
52442050
52446799
4749
-416
-9398
BAP1
BRCA1 associated protein-1 (ubiquitin carboxy-terminal hydrolase)
chr19
45310400
45315649
5249
687
-10585
BCAM
basal cell adhesion molecule (Lutheran blood group)
chr19
49312350
49317799
5449
-897
-16755
BCAT2
branched chain amino-acid transaminase 2, mitochondrial
chr16
31122030
31126299
4269
4503
55
BCKDK
branched chain ketoacid dehydrogenase kinase
chr16
30903680
30909349
5669
-1116
-7398
BCL7C
B-cell CLL/lymphoma 7C
chr16
2262050
2266549
4499
-3289
-5046
BRICD5
BRICHOS domain containing 5
chr19
2013430
2017449
4019
262
-29993
BTBD2
BTB (POZ) domain containing 2
chr5
180470430
180473549
3119
4765
-16532
BTNL9
butyrophilin-like 9
chr12
7050980
7056099
5119
337
-1624
C12orf57
chromosome 12 open reading frame 57
chr15
36867750
36870399
2649
-2737
-233373
C15orf41
chromosome 15 open reading frame 41
chr16
685600
694399
8799
-3653
-5571
C16orf13
chromosome 16 open reading frame 13
chr16
67700130
67702549
2419
623
-1319
C16orf86
chromosome 16 open reading frame 86
chr17
76138880
76142799
3919
-1594
-21523
C17orf99
chromosome 17 open reading frame 99
chr1
45138330
45141879
3549
-289
-51158
C1orf228
chromosome 1 open reading frame 228
8
159
PART THREE CHAPTER 8
Supplementary Table 5. Continued
160
chr20
3744750
3749799
5049
1177
-13117
C20orf27
chromosome 20 open reading frame 27
chr15
62457530
62459779
2249
-1173
-2918
C2CD4B
C2 calcium-dependent domain containing 4B
chr2
241836580
241839499
2919
-2467
-12574
C2orf54
chromosome 2 open reading frame 54
chr9
139832150
139847999
15849
377
-1351
C8G
complement component 8, gamma polypeptide
chr8
22453030
22464799
11769
1793
-2745
C8orf58
chromosome 8 open reading frame 58
chr9
139883230
139889449
6219
-530
-2087
C9orf142
chromosome 9 open reading frame 142
chr9
140144030
140149999
5969
1285
-919
C9orf173
chromosome 9 open reading frame 173
chr1
150227300
150233699
6399
282
-6976
CA14
carbonic anhydrase XIV
chr20
60980500
60982679
2179
749
-17904
CABLES2
Cdk5 and Abl enzyme substrate 2
chr17
4888830
4894049
5219
-509
-20147
CAMTA2
calmodulin binding transcription activator 2
chr16
574980
577699
2719
-1516
-28294
CAPN15
calpain 15
chr11
67181830
67188699
6869
2116
-7811
CARNS1
carnosine synthase 1
chr17
73503400
73511679
8279
-1582
-11197
CASKIN2
CASK interacting protein 2
chr7
142981380
142987279
5899
-1072
-20454
CASP2
caspase 2, apoptosis-related cysteine peptidase
chr1
10854430
10856949
2519
1017
-159022
CASZ1
castor zinc finger 1
chr17
77750250
77752249
1999
-743
-5081
CBX2
chromobox homolog 2
chr8
22453030
22464799
11769
-3636
-19060
CCAR2
cell cycle and apoptosis regulator 2
chr19
56159080
56163729
4649
2451
-3119
CCDC106
coiled-coil domain containing 106
chr19
11544680
11547879
3199
-300
-15007
CCDC151
coiled-coil domain containing 151
chr9
139685050
139687199
2149
-4677
-16067
CCDC183
coiled-coil domain containing 183
chr15
74606200
74608699
2499
-3450
-21031
CCDC33
coiled-coil domain containing 33
chr17
80170500
80172549
2049
-836
-112179
CCDC57
coiled-coil domain containing 57
chr3
49201400
49205299
3899
435
-3381
CCDC71
coiled-coil domain containing 71
chr20
30597300
30599829
2529
320
-21419
CCM2L
cerebral cavernous malformation 2-like
chr17
41923180
41927749
4569
918
-15529
CD300LG
CD300 molecule-like family member g
chr19
49841400
49845099
3699
4573
-609
CD37
CD37 molecule
chr6
14117500
14120079
2579
925
-18356
CD83
CD83 molecule
chr11
65077050
65079229
2179
-4191
-11759
CDC42EP2
CDC42 effector protein (Rho GTPase binding) 2
chr1
205471350
205477149
5799
523
-27665
CDK18
cyclin-dependent kinase 18
chr11
67270450
67277679
7229
2037
-96
CDK2AP2
cyclin-dependent kinase 2 associated protein 2
chr11
2910230
2912229
1999
-4235
-6780
CDKN1C
cyclin-dependent kinase inhibitor 1C (p57, Kip2)
chr16
66965730
66971329
5599
183
-10462
CES2
carboxylesterase 2
chr16
56994150
56997349
3199
-85
-22006
CETP
cholesteryl ester transfer protein, plasma
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Supplementary Table 5. Continued chr15
41247600
41251249
3649
3789
708
CHAC1
ChaC glutathione-specific gamma-glutamylcyclotransferase 1
chr22
51014300
51022929
8629
2813
-1228
CHKB
choline kinase beta
chr22
51014300
51022929
8629
2813
-11324
CHKB-CPT1B
CHKB-CPT1B readthrough (NMD candidate)
chr11
45682650
45685499
2849
3097
-13648
CHST1
carbohydrate (keratan sulfate Gal-6) sulfotransferase 1
chr1
16337830
16349649
11819
-4746
-16804
CLCNKA
chloride channel, voltage-sensitive Ka
chr22
19507780
19514699
6919
1620
-693
CLDN5
claudin 5
chr9
139883230
139889449
6219
4684
2720
CLIC3
chloride intracellular channel 3
chr7
73701450
73706329
4879
85
-116382
CLIP2
CAP-GLY domain containing linker protein 2
chr4
17512050
17515299
3249
-3113
-15052
CLRN2
clarin 2
chr17
2609200
2611349
2149
4652
-17595
CLUH
clustered mitochondria (cluA/CLU1) homolog
chr16
81475680
81482179
6499
155
-266435
CMIP
c-Maf inducing protein
chr16
66638980
66641179
2199
1513
-7713
CMTM3
CKLF-like MARVEL transmembrane domain containing 3
chr1
224544200
224546429
2229
720
-18376
CNIH4
cornichon family AMPA receptor auxiliary protein 4
chr19
17666050
17668899
2849
964
-26490
COLGALT1
collagen beta(1-O) galactosyltransferase 1
chr17
27945730
27951749
6019
-299
-6966
CORO6
coronin 6
chr22
51014300
51022929
8629
-2109
-11324
CPT1B
carnitine palmitoyltransferase 1B (muscle)
chr19
50190100
50192799
2699
-2957
-25538
CPT1C
carnitine palmitoyltransferase 1C
chr14
105936030
105952849
16819
3309
-2060
CRIP2
cysteine-rich protein 2
chr6
43273300
43276279
2979
1740
-1579
CRIP3
cysteine-rich protein 3
chr8
19533450
19536879
3429
4929
-273492
CSGALNACT1 chondroitin sulfate N-acetylgalactosaminyltransferase 1
chr4
1242630
1244849
2219
-832
-38512
CTBP1
C-terminal binding protein 1
chr4
1242630
1244849
2219
-437
-2876
CTBP1-AS2
CTBP1 antisense RNA 2 (head to head)
chr16
30903680
30909349
5669
-1413
-8365
CTF1
cardiotrophin 1
chr1
9960150
9973749
13599
3366
-58616
CTNNBIP1
catenin, beta interacting protein 1
chr9
140114550
140116549
1999
-4094
-5212
CYSRT1
cysteine-rich tail protein 1
chr1
100713980
100716899
2919
-31
-62961
DBT
dihydrolipoamide branched chain transacylase E2
chr19
17412600
17418449
5849
-4812
-18579
DDA1
DET1 and DDB1 associated 1
chr1
20985900
20990479
4579
-153
-9930
DDOST
dolichyl-diphosphooligosaccharide-protein glycosyltransferase subunit (non-catalytic)
8
chr20
3183500
3185779
2279
655
-13622
DDRGK1
DDRGK domain containing 1
chr1
153915700
153922549
6849
29
-17148
DENND4B
DENN/MADD domain containing 4B
chr2
220279700
220288849
9149
1176
-7184
DES
desmin
chr8
145546230
145550849
4619
2027
-10293
DGAT1
diacylglycerol O-acyltransferase 1
chr22
18889580
18893529
3949
-2181
-8045
DGCR6
DiGeorge syndrome critical region gene 6
161
PART THREE CHAPTER 8
Supplementary Table 5. Continued chr22
162
20307150
20310579
3429
-1257
-7063
DGCR6L
DiGeorge syndrome critical region gene 6-like
chr11
46359280
46364779
5499
-4957
-40074
DGKZ
diacylglycerol kinase, zeta
chr19
2721400
2725129
3729
-1875
-8700
DIRAS1
DIRAS family, GTP-binding RAS-like 1
chr16
84175700
84179199
3499
-1415
-34073
DNAAF1
dynein, axonemal, assembly factor 1
chr11
6517630
6520149
2519
364
-23206
DNHD1
dynein heavy chain domain 1
chr3
16304800
16310129
5329
-969
-8896
DPH3
diphthamide biosynthesis 3
chr2
162929150
162931229
2079
862
-81433
DPP4
dipeptidyl-peptidase 4
chr18
32072330
32075829
3499
826
-335212
DTNA
dystrobrevin, alpha
chr1
1279200
1283179
3979
3302
-10531
DVL1
dishevelled segment polarity protein 1
chr2
71693380
71698379
4999
2048
-218012
DYSF
dysferlin
chr10
135185580
135188849
3269
-307
-11227
ECHS1
enoyl CoA hydratase, short chain, 1, mitochondrial
chr16
2299730
2303279
3549
98
-11606
ECI1
enoyl-CoA delta isomerase 1
chr20
62127280
62136049
8769
-1160
-12299
EEF1A2
eukaryotic translation elongation factor 1 alpha 2
chr3
127871100
127876849
5749
1662
-253513
EEFSEC
eukaryotic elongation factor, selenocysteine-tRNA-specific
chr11
65637800
65641799
3999
540
-5887
EFEMP2
EGF containing fibulin-like extracellular matrix protein 2
chr9
139549230
139560549
11319
-2492
-12239
EGFL7
EGF-like-domain, multiple 7
chr19
39108050
39111379
3329
-7
-17880
EIF3K
eukaryotic translation initiation factor 3, subunit K
chr3
47553250
47555429
2179
859
-17209
ELP6
elongator acetyltransferase complex subunit 6
chr17
48449130
48451499
2369
-266
-8505
EME1
essential meiotic structure-specific endonuclease 1
chr16
67700130
67702549
2419
-712
-4490
ENKD1
enkurin domain containing 1
chr6
132127250
132129929
2679
-566
-87703
ENPP1
ectonucleotide pyrophosphatase/ phosphodiesterase 1
chr19
45925350
45928979
3629
-345
-10472
ERCC1
excision repair crosscomplementation group 1
chr16
68263750
68268249
4499
4136
-3550
ESRP2
epithelial splicing regulatory protein 2
chr11
64070830
64074629
3799
-314
-11480
ESRRA
estrogen-related receptor alpha
chr1
2123680
2131879
8199
-1566
-6791
FAAP20
Fanconi anemia core complex associated protein 20
chr16
1875050
1878249
3199
-575
-2258
FAHD1
fumarylacetoacetate hydrolase domain containing 1
chr1
26480880
26493149
12269
1504
-2103
FAM110D
family with sequence similarity 110, member D
chr9
140144030
140149999
5969
-4793
-8978
FAM166A
family with sequence similarity 166, member A
chr16
685600
694399
8799
-1849
-8474
FAM195A
family with sequence similarity 195, member A
chr17
79789250
79793849
4599
-383
-11257
FAM195B
family with sequence similarity 195, member B
chr3
49839150
49847149
7999
2463
687
FAM212A
family with sequence similarity 212, member A
chr1
2513830
2518479
4649
-2094
-6747
FAM213B
family with sequence similarity 213, member B
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Supplementary Table 5. Continued chr16
67557730
67568399
10669
311
-17623
FAM65A
family with sequence similarity 65, member A
chr9
134153400
134155549
2149
-2569
-21009
FAM78A
family with sequence similarity 78, member A
chr16
66965730
66971329
5599
-210
-2571
FAM96B
family with sequence similarity 96, member B
chr17
80056450
80060499
4049
-2369
-22260
FASN
fatty acid synthase
chr9
139832150
139847999
15849
-902
-5187
FBXW5
F-box and WD repeat domain containing 5
chr1
27699250
27701399
2149
990
-4722
FCN3
ficolin (collagen/fibrinogen domain containing) 3
chr17
72872300
72875879
3579
-4934
-15470
FDXR
ferredoxin reductase
chr11
64006730
64011049
4319
-698
-2716
FKBP2
FK506 binding protein 2, 13kDa
chr17
43299000
43307899
8899
4158
-21231
FMNL1
formin-like 1
chr9
130563450
130566979
3529
-293
-11341
FPGS
folylpolyglutamate synthase
chr19
50311200
50314199
2999
3768
-2551
FUZ
fuzzy planar cell polarity protein
chr5
170213550
170217499
3949
4802
-25523
GABRP
gamma-aminobutyric acid (GABA) A receptor, pi
chr22
29700700
29709399
8699
2053
-3724
GAS2L1
growth arrest-specific 2 like 1
chr7
56030430
56037579
7149
1709
-33866
GBAS
glioblastoma amplified sequence
chr6
33544230
33549349
5119
-4686
-10012
GGNBP1
gametogenetin binding protein 1 (pseudogene)
chr1
228333250
228335279
2029
-3288
-13260
GJC2
gap junction protein, gamma 2, 47kDa
chr20
62257800
62260899
3099
-969
-40395
GMEB2
glucocorticoid modulatory element binding protein 2
chr19
34854350
34858199
3849
206
-34961
GPI
glucose-6-phosphate isomerase
chr8
144289550
144292349
2799
-4118
-8093
GPIHBP1
glycosylphosphatidylinositol anchored high density lipoprotein binding protein 1
chr14
105530000
105532449
2449
529
-15487
GPR132
G protein-coupled receptor 132
chr7
1087930
1097449
9519
-4451
-6207
GPR146
G protein-coupled receptor 146
chr2
128400730
128402849
2119
-1649
-8422
GPR17
G protein-coupled receptor 17
chr9
139217600
139224349
6749
-957
-15312
GPSM1
G-protein signaling modulator 1
chr8
145725930
145730229
4299
-1385
-4474
GPT
glutamic-pyruvate transaminase (alanine aminotransferase)
chr22
47022480
47025799
3319
1482
-51547
GRAMD4
GRAM domain containing 4
8
chr8
144640180
144642199
2019
713
-4039
GSDMD
gasdermin D
chr4
106628350
106631429
3079
-2080
-137018
GSTCD
glutathione S-transferase, C-terminal domain containing
chr2
27577800
27581299
3499
318
-30829
GTF3C2
general transcription factor IIIC, polypeptide 2, beta 110kDa
chr1
228333250
228335279
2029
1861
-2388
GUK1
guanylate kinase 1
chr2
26463700
26470749
7049
369
-53720
HADHA
hydroxyacyl-CoA dehydrogenase/3ketoacyl-CoA thiolase/enoyl-CoA hydratase (trifunctional protein), alpha subunit
chr2
26463700
26470749
7049
-391
-46107
HADHB
hydroxyacyl-CoA dehydrogenase/3ketoacyl-CoA thiolase/enoyl-CoA hydratase (trifunctional protein), beta subunit
163
PART THREE CHAPTER 8
Supplementary Table 5. Continued
164
chr16
1875050
1878249
3199
545
-17544
HAGH
chr1
937850
941599
3749
-4173
-5383
HES4
hes family bHLH transcription factor 4
chr9
35810700
35815449
4749
1967
-116
HINT2
histidine triad nucleotide binding protein 2
chr19
37824430
37826979
2549
125
-29650
HKR1
HKR1, GLI-Kruppel zinc finger family member
hydroxyacylglutathione hydrolase
chr5
175105900
175116349
10449
2661
-433
HRH2
histamine receptor H2
chr17
40705030
40708349
3319
2706
-534
HSD17B1
hydroxysteroid (17-beta) dehydrogenase 1
chr16
84175700
84179199
3499
1350
-21706
HSDL1
hydroxysteroid dehydrogenase like 1
chr1
16337830
16349649
11819
1545
-3217
HSPB7
heat shock 27kDa protein family, member 7 (cardiovascular)
chr3
50332550
50342399
9849
3529
-154
HYAL1
hyaluronoglucosaminidase 1
chr3
50332550
50342399
9849
-576
-7213
HYAL3
hyaluronoglucosaminidase 3
chr3
50327080
50332249
5169
361
-4500
IFRD2
interferon-related developmental regulator 2
chr11
68669900
68673799
3899
531
-36219
IGHMBP2
immunoglobulin mu binding protein 2
chr19
14141200
14144229
3029
453
-21310
IL27RA
interleukin 27 receptor, alpha
chr19
15232930
15237899
4969
1162
-9628
ILVBL
ilvB (bacterial acetolactate synthase)-like
chr2
86420630
86422849
2219
1153
-50684
IMMT
inner membrane protein, mitochondrial
chr14
105153800
105158049
4249
-33
-15211
INF2
inverted formin, FH2 and WH2 domain containing
chr22
31513550
31520899
7349
-1736
-13457
INPP5J
inositol polyphosphate-5phosphatase J
chr4
106628350
106631429
3079
-9
-26103
INTS12
integrator complex subunit 12
chr6
33712580
33716449
3869
247
-25072
IP6K3
inositol hexakisphosphate kinase 3
chr1
44410650
44415499
4849
597
-20618
IPO13
importin 13
chr9
95430680
95434079
3399
167
-56914
IPPK
inositol 1,3,4,5,6-pentakisphosphate 2-kinase
chr16
19726930
19731429
4499
1402
-139677
IQCK
IQ motif containing K
chr19
49246330
49251199
4869
1401
-4618
IZUMO1
izumo sperm-egg fusion 1
chr19
8405250
8409529
4279
756
-19921
KANK3
KN motif and ankyrin repeat domains 3
chr16
31122030
31126299
4269
-4820
-18548
KAT8
K(lysine) acetyltransferase 8
chr22
38844330
38850849
6519
3613
-25257
KCNJ4
potassium channel, inwardly rectifying subfamily J, member 4
chr11
2478500
2486549
8049
-159
-387814
KCNQ1
potassium channel, voltage gated KQT-like subfamily Q, member 1
chr3
49207300
49215029
7729
2097
-2753
KLHDC8B
kelch domain containing 8B
chr12
27931400
27934829
3429
-72
-22857
KLHL42
kelch-like family member 42
chr16
19726930
19731429
4499
312
-11504
KNOP1
lysine-rich nucleolar protein 1
chr12
52583730
52590249
6519
-1206
-24209
KRT80
keratin 80, type II
chr3
49167180
49174629
7449
-306
-12357
LAMB2
laminin, beta 2 (laminin S)
chr9
139641350
139644029
2679
290
-4221
LCN6
lipocalin 6
chr9
139873150
139879049
5899
-1345
-4110
LCNL1
lipocalin-like 1
chr17
76975600
76977629
2029
-554
-9279
LGALS3BP
lectin, galactoside-binding, soluble, 3 binding protein
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Supplementary Table 5. Continued chr2
128400730
128402849
2119
3963
-5793
LIMS2
LIM and senescent cell antigen-like domains 2
chr1
226492350
226498679
6329
1683
-76654
LIN9
lin-9 DREAM MuvB core complex component
chr16
574980
577699
2719
1067
507
LINC00235
long intergenic non-protein coding RNA 235
chr1
204010500
204015049
4549
-2383
-11200
LINC00303
long intergenic non-protein coding RNA 303
chr21
44894180
44902029
7849
-2
-16131
LINC00313
long intergenic non-protein coding RNA 313
chr17
8128700
8131129
2429
-2554
-5966
LINC00324
long intergenic non-protein coding RNA 324
chr6
33552330
33560199
7869
4850
-2382
LINC00336
long intergenic non-protein coding RNA 336
chr17
79280000
79288399
8399
-1152
-7576
LINC00482
long intergenic non-protein coding RNA 482 lemur tyrosine kinase 2
chr7
97736250
97738279
2029
1068
-101677
LMTK2
chr16
427600
430899
3299
-2991
-13710
LOC100134368 uncharacterized LOC100134368
chr2
132247880
132249929
2049
-1481
-30244
LOC150776
sphingomyelin phosphodiesterase 4, neutral membrane (neutral sphingomyelinase-3) pseudogene
chr1
148927800
148930399
2599
-1305
-23954
LOC645166
lymphocyte-specific protein 1 pseudogene
chr12
6739500
6743499
3999
-685
-13499
LPAR5
lysophosphatidic acid receptor 5
chr22
25759850
25762399
2549
-2601
-13739
LRP5L
low density lipoprotein receptorrelated protein 5-like
chr5
192450
196749
4299
2974
-868
LRRC14B
leucine rich repeat containing 14B
chr18
7232780
7235749
2969
3128
2224
LRRC30
leucine rich repeat containing 30
chr17
79978800
79981299
2499
-1230
-8976
LRRC45
leucine rich repeat containing 45
chr6
43478000
43481749
3749
-1451
-5168
LRRC73
leucine rich repeat containing 73
chr20
60691250
60697399
6149
-3192
-16109
LSM14B
LSM14B, SCD6 homolog B (S. cerevisiae)
chr11
1883600
1887029
3429
-4588
-28177
LSP1
lymphocyte-specific protein 1
8
chr3
46503580
46506349
2769
1430
-27468
LTF
lactotransferrin
chr11
47289380
47293829
4449
-66
-59977
MADD
MAP-kinase activating death domain
chr17
79881300
79885699
4399
-2092
-7353
MAFG
v-maf avian musculoaponeurotic fibrosarcoma oncogene homolog G
chr19
39108050
39111379
3329
-1072
-31434
MAP4K1
mitogen-activated protein kinase kinase kinase kinase 1
chr11
64566650
64572699
6049
1038
-13065
MAP4K2
mitogen-activated protein kinase kinase kinase kinase 2
chr17
19279500
19282399
2899
-824
-5906
MAPK7
mitogen-activated protein kinase 7
chr5
65890530
65892729
2199
-546
-390720
MAST4
microtubule associated serine/ threonine kinase family member 4
chr22
36005130
36020099
14969
769
-9803
MB
myoglobin
chr11
119184500
119193429
8929
-1125
-9730
MCAM
melanoma cell adhesion molecule
chr13
113623380
113625479
2099
1673
-128432
MCF2L
MCF.2 cell line derived transforming sequence-like
chr1
12037930
12047429
9499
2442
-30891
MFN2
mitofusin 2
chr8
145732030
145734849
2819
-1112
-3149
MFSD3
major facilitator superfamily domain containing 3
165
PART THREE CHAPTER 8
Supplementary Table 5. Continued chr4
166
679900
684749
4849
648
-6706
MFSD7
major facilitator superfamily domain containing 7
chr17
37884730
37887429
2699
708
-669
MIEN1
migration and invasion enhancer 1
chr1
12078850
12081379
2529
603
-11991
MIIP
migration and invasion inhibitory protein
chr12
7065200
7074549
9349
-3385
-3479
MIR141
microRNA 141
chr22
20019100
20021679
2579
-272
-352
MIR185
microRNA 185
chr11
64657150
64663099
5949
-1407
-1516
MIR192
microRNA 192
chr11
64657150
64663099
5949
-1214
-1298
MIR194-2
microRNA 194-2
chr12
7065200
7074549
9349
-2987
-3054
MIR200C
microRNA 200c
chr2
219268450
219272979
4529
3346
3270
MIR26B
microRNA 26b
chr8
41511200
41530879
19679
-3014
-3081
MIR486-1
microRNA 486-1
chr2
238393950
238399729
5779
962
-67120
MLPH
melanophilin
chr16
83930830
83940479
9649
2925
-14130
MLYCD
malonyl-CoA decarboxylase
chr10
88712330
88722199
9869
160
-21965
MMRN2
multimerin 2
chr6
29621000
29627249
6249
-684
-16022
MOG
myelin oligodendrocyte glycoprotein
chr16
118550
130449
11899
-3669
-11341
MPG
N-methylpurine-DNA glycosylase
chr17
41907450
41912249
4799
688
-31681
MPP3
membrane protein, palmitoylated 3 (MAGUK p55 subfamily member 3)
chr20
61422600
61429579
6979
-1715
-5853
MRGBP
MRG/MORF4L binding protein
chr11
68669900
68673799
3899
-547
-13103
MRPL21
mitochondrial ribosomal protein L21
chr11
1965980
1972229
6249
603
-8732
MRPL23
mitochondrial ribosomal protein L23
chr11
2007080
2010879
3799
2170
-4540
MRPL23-AS1
MRPL23 antisense RNA 1
chr17
48449130
48451499
2369
247
-5086
MRPL27
mitochondrial ribosomal protein L27
chr19
17412600
17418449
5849
-952
-2127
MRPL34
mitochondrial ribosomal protein L34
chr2
86420630
86422849
2219
-4816
-18736
MRPL35
mitochondrial ribosomal protein L35
chr5
68512880
68516199
3319
967
-11443
MRPS36
mitochondrial ribosomal protein S36
chr14
105884430
105890499
6069
1279
-49583
MTA1
metastasis associated 1
chr9
129087400
129091479
4079
312
-70472
MVB12B
multivesicular body subunit 12B
chr4
2264730
2269349
4619
-3301
-17880
MXD4
MAX dimerization protein 4
chr17
4456380
4460729
4349
126
-16363
MYBBP1A
MYB binding protein (P160) 1a
chr11
69061700
69069049
7349
3753
622
MYEOV
myeloma overexpressed
chr14
23902280
23915729
13449
-4135
-27057
MYH7
myosin, heavy chain 7, cardiac muscle, beta
chr3
46901750
46917629
15879
-4717
-10333
MYL3
myosin, light chain 3, alkali; ventricular, skeletal, slow
chr7
44176550
44180699
4149
2291
-160
MYL7
myosin, light chain 7, regulatory
chr16
46776200
46786679
10479
781
-45244
MYLK3
myosin light chain kinase 3
chr16
30381630
30385179
3549
-2718
-5905
MYLPF
myosin light chain, phosphorylatable, fast skeletal muscle
chr17
73585450
73588399
2949
2786
-36000
MYO15B
myosin XVB pseudogene
chr17
1384900
1400779
15879
-3789
-25358
MYO1C
myosin IC
chr2
132247880
132249929
2049
1159
-7371
MZT2A
mitotic spindle organizing protein 2A
chr11
63705000
63708099
3099
108
-17894
NAA40
N(alpha)-acetyltransferase 40, NatD catalytic subunit
chr8
144659150
144662599
3449
-362
-3919
NAPRT
nicotinate phosphoribosyltransferase
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Supplementary Table 5. Continued chr3
50332550
50342399
9849
-763
-3641
NAT6
N-acetyltransferase 6 (GCN5-related)
chr3
47016580
47030849
14269
2542
-27478
NBEAL2
neurobeachin-like 2
chr3
196667800
196671699
3899
-286
-7477
NCBP2
nuclear cap binding protein subunit 2, 20kDa
chr11
67372880
67376099
3219
167
-5522
NDUFV1
NADH dehydrogenase (ubiquinone) flavoprotein 1, 51kDa
chr9
140144030
140149999
5969
-2744
-20984
NELFB
negative elongation factor complex member B
chr6
31829100
31832549
3449
-116
-3995
NEU1
sialidase 1 (lysosomal sialidase)
chr19
3357000
3359979
2979
-1126
-105112
NFIC
nuclear factor I/C (CCAAT-binding transcription factor)
chr19
3930650
3932849
2199
-1351
-10662
NMRK2
nicotinamide riboside kinase 2
chr9
139440700
139445029
4329
-2627
-53968
NOTCH1
notch 1
chr16
1828450
1830549
2099
-3433
-9691
NUBP2
nucleotide binding protein 2
chr6
34358350
34362149
3799
191
-104248
NUDT3
nudix (nucleoside diphosphate linked moiety X)-type motif 3
chr11
67397380
67400999
3619
-1789
-3780
NUDT8
nudix (nucleoside diphosphate linked moiety X)-type motif 8
chr12
57604630
57608729
4099
-3898
-13551
NXPH4
neurexophilin 4
chr8
145115030
145119379
4349
-1621
-11038
OPLAH
5-oxoprolinase (ATP-hydrolysing)
chr11
4663080
4669349
6269
1059
-10499
OR51E1
olfactory receptor, family 51, subfamily E, member 1
chr3
16304800
16310129
5329
751
-40128
OXNAD1
oxidoreductase NAD-binding domain containing 1
chr11
72926850
72931279
4429
-437
-18328
P2RY2
purinergic receptor P2Y, G-protein coupled, 2
chr17
79820600
79822599
1999
-3056
-20566
P4HB
prolyl 4-hydroxylase, beta polypeptide
chr11
47196880
47211899
15019
3568
-5305
PACSIN3
protein kinase C and casein kinase substrate in neurons 3
chr8
145055750
145062349
6599
1585
-7729
PARP10
poly (ADP-ribose) polymerase family, member 10
chr1
154925950
154931229
5279
-23
-12029
PBXIP1
pre-B-cell leukemia homeobox interacting protein 1
chr3
51995350
52003949
8599
-3998
-8176
PCBP4
poly(rC) binding protein 4
chr4
698200
702999
4799
1027
-63825
PCGF3
polycomb group ring finger 3
chr2
242798300
242803129
4829
343
-8681
PDCD1
programmed cell death 1
chr5
269850
272399
2549
-611
-43964
PDCD6
programmed cell death 6
chr19
10539800
10545299
5499
-561
-37757
PDE4A
phosphodiesterase 4A, cAMPspecific
chr17
48171180
48176399
5219
1094
-14632
PDK2
pyruvate dehydrogenase kinase, isozyme 2
chr7
95223880
95231629
7749
-1830
-14940
PDK4
pyruvate dehydrogenase kinase, isozyme 4
chr7
44101430
44106729
5299
1106
-1753
PGAM2
phosphoglycerate mutase 2 (muscle)
chr16
2262050
2266549
4499
522
-2697
PGP
phosphoglycolate phosphatase
chr22
41863350
41867199
3849
-567
-9553
PHF5A
PHD finger protein 5A
8
chr3
52442050
52446799
4749
-102
-13231
PHF7
PHD finger protein 7
chr11
803650
807549
3899
-1136
-6421
PIDD1
p53-induced death domain protein 1
167
PART THREE CHAPTER 8
Supplementary Table 5. Continued
168
chr20
44043350
44050849
7499
2376
-7784
PIGT
phosphatidylinositol glycan anchor biosynthesis, class T
chr22
50346600
50353799
7199
-3943
-7519
PIM3
Pim-3 proto-oncogene, serine/ threonine kinase
chr11
67270450
67277679
7229
-1222
-14825
PITPNM1
phosphatidylinositol transfer protein, membrane-associated 1
chr12
123575300
123606029
30729
4310
-122638
PITPNM2
phosphatidylinositol transfer protein, membrane-associated 2
chr9
131464700
131469279
4579
2188
-16208
PKN3
protein kinase N3
chr16
68277400
68283829
6429
1368
-14344
PLA2G15
phospholipase A2, group XV
chr17
43203380
43211299
7919
2551
-18331
PLCD3
phospholipase C, delta 3
chr8
145041550
145051029
9479
1407
-56969
PLEC
plectin
chr19
4517900
4520749
2849
-1609
-17133
PLIN4
perilipin 4
chr3
142313680
142316179
2499
-299
-117573
PLS1
plastin 1
chr22
50746300
50752549
6249
-3423
-36017
PLXNB2
plexin B2
chr17
37819630
37837829
18199
4223
2002
PNMT
phenylethanolamine N-methyltransferase
chr11
817800
821749
3949
873
-5441
PNPLA2
patatin-like phospholipase domain containing 2
chr6
43478000
43481749
3749
-4916
-9369
POLR1C
polymerase (RNA) I polypeptide C, 30kDa
chr11
842750
847999
5249
-2846
-5653
POLR2L
polymerase (RNA) II (DNA directed) polypeptide L, 7.6kDa
chr9
134161100
134177199
16099
4069
-15499
PPAPDC3
phosphatidic acid phosphatase type 2 domain containing 3
chr17
79789250
79793849
4599
1376
-182
PPP1R27
protein phosphatase 1, regulatory subunit 27
chr8
9011450
9014329
2879
-4670
-19116
PPP1R3B
protein phosphatase 1, regulatory subunit 3B
chr11
64083730
64089799
6069
1196
-2517
PRDX5
peroxiredoxin 5
chr19
11544680
11547879
3199
11
-15500
PRKCSH
protein kinase C substrate 80K-H
chr19
50091980
50096699
4719
-572
-35355
PRR12
proline rich 12
chr10
694600
697729
3129
148
-14943
PRR26
proline rich 26
chr17
62074250
62078049
3799
439
-5492
PRR29
proline rich 29
chr14
24613880
24617999
4119
-85
-3365
PSME2
proteasome (prosome, macropain) activator subunit 2 (PA28 beta)
chr11
450730
452949
2219
1560
-39546
PTDSS2
phosphatidylserine synthase 2
chr9
139867530
139871599
4069
-2391
-6628
PTGDS
prostaglandin D2 synthase 21kDa (brain)
chr12
7050980
7056099
5119
-2200
-16939
PTPN6
protein tyrosine phosphatase, non-receptor type 6
chr10
134210300
134215799
5499
2348
-18301
PWWP2B
PWWP domain containing 2B
chr10
100173400
100176299
2899
128
-31527
PYROXD2
pyridine nucleotide-disulphide oxidoreductase domain 2
chr3
49140100
49143649
3549
296
-8509
QARS
glutaminyl-tRNA synthetase
chr4
17512050
17515299
3249
182
-25655
QDPR
quinoid dihydropteridine reductase
chr3
49129400
49133149
3749
229
-64131
QRICH1
glutamine-rich 1
chr11
67156500
67160899
4399
-723
-7182
RAD9A
RAD9 homolog A (S. pombe)
chr9
135987050
135999829
12779
3121
-20332
RALGDS
ral guanine nucleotide dissociation stimulator
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Supplementary Table 5. Continued chr7
45195050
45197099
2049
-1292
-27772
RAMP3
receptor (G protein-coupled) activity modifying protein 3
chr22
20103200
20105499
2299
-674
-10354
RANBP1
RAN binding protein 1
chr12
48154830
48160049
5219
-4551
-28985
RAPGEF3
Rap guanine nucleotide exchange factor (GEF) 3
chr11
64510600
64517099
6499
-2220
-19466
RASGRP2
RAS guanyl releasing protein 2 (calcium and DAG-regulated)
chr19
49246330
49251199
4869
-4795
-24922
RASIP1
Ras interacting protein 1
chr11
63684250
63691129
6879
-3374
-8986
RCOR2
REST corepressor 2
chr9
116324450
116338849
14399
4289
-28367
RGS3
regulator of G-protein signaling 3
chr16
118550
130449
11899
-1871
-16441
RHBDF1
rhomboid 5 homolog 1 (Drosophila)
chr19
14141200
14144229
3029
3698
932
RLN3
relaxin 3
chr9
140116700
140119679
2979
-2415
-3483
RNF208
ring finger protein 208
chr14
24613880
24617999
4119
-719
-13930
RNF31
ring finger protein 31
chr11
124765550
124772199
6649
-1044
-14760
ROBO4
roundabout, axon guidance receptor, homolog 4 (Drosophila)
chr1
151803080
151806949
3869
-667
-26467
RORC
RAR-related orphan receptor C
chr8
146015880
146019829
3949
-126
-2700
RPL8
ribosomal protein L8
chr11
803650
807549
3899
-4336
-7276
RPLP2
ribosomal protein, large, P2
chr11
64120350
64128099
7749
-2400
-15461
RPS6KA4
ribosomal protein S6 kinase, 90kDa, polypeptide 4
chr3
9884280
9887099
2819
12
-6156
RPUSD3
RNA pseudouridylate synthase domain containing 3
chr6
42987230
42990699
3469
-420
-8372
RRP36
ribosomal RNA processing 36 homolog (S. cerevisiae)
chr22
42980000
42982049
2049
-3008
-11757
RRP7BP
ribosomal RNA processing 7 homolog B (S. cerevisiae), pseudogene
8
chr1
151339980
151346099
6119
2124
-6259
SELENBP1
selenium binding protein 1
chr2
97535400
97542429
7029
-3180
-13441
SEMA4C
sema domain, immunoglobulin domain (Ig), transmembrane domain (TM) and short cytoplasmic domain, (semaphorin) 4C
chr17
2238380
2242929
4549
-153
-43692
SGSM2
small G protein signaling modulator 2
chr9
130538250
130540529
2279
1658
-38793
SH2D3C
SH2 domain containing 3C
chr4
2787230
2795149
7919
-3560
-51632
SH3BP2
SH3-domain binding protein 2
chr19
457930
462449
4519
806
-43605
SHC2
SHC (Src homology 2 domain containing) transforming protein 2
chr3
48538100
48542099
3999
1561
-30903
SHISA5
shisa family member 5
chr11
65403430
65410199
6769
-777
-11576
SIPA1
signal-induced proliferationassociated 1
chr17
80185000
80189079
4079
-3104
-10328
SLC16A3
solute carrier family 16 (monocarboxylate transporter), member 3
chr17
79676980
79679049
2069
-1356
-10026
SLC25A10
solute carrier family 25 (mitochondrial carrier; dicarboxylate transporter), member 10
chr3
48933350
48937749
4399
852
-41193
SLC25A20
solute carrier family 25 (carnitine/ acylcarnitine translocase), member 20
chr1
16061050
16063329
2279
-619
-5694
SLC25A34
solute carrier family 25, member 34
chr2
220487900
220503449
15549
3383
-11026
SLC4A3
solute carrier family 4 (anion exchanger), member 3
169
PART THREE CHAPTER 8
Supplementary Table 5. Continued chr11
170
74870080
74873249
3169
820
-45779
SLCO2B1
solute carrier organic anion transporter family, member 2B1
chr1
214448850
214457949
9099
-1165
-57077
SMYD2
SET and MYND domain containing 2
chr10
88712330
88722199
9869
-1023
-5752
SNCG
synuclein, gamma (breast cancer-specific protein 1)
chr16
29299600
29304949
5349
-266
-83321
SNX29P2
sorting nexin 29 pseudogene 2
chr12
53494300
53500099
5799
-74
-21122
SOAT2
sterol O-acyltransferase 2
chr20
62683150
62687899
4749
-4546
-6444
SOX18
SRY (sex determining region Y) -box 18
chr17
4868900
4879399
10499
-3018
-11627
SPAG7
sperm associated antigen 7
chr9
35810700
35815449
4749
-816
-5291
SPAG8
sperm associated antigen 8
chr17
48619730
48621949
2219
-3722
-12371
SPATA20
spermatogenesis associated 20
chr16
1828450
1830549
2099
3081
-2786
SPSB3
splA/ryanodine receptor domain and SOCS box containing 3
chr6
43137430
43143849
6419
1720
-8603
SRF
serum response factor (c-fos serum response element-binding transcription factor)
chr9
140081500
140084979
3479
186
-1582
SSNA1
Sjogren syndrome nuclear autoantigen 1
chr7
116588580
116591829
3249
-3176
-273750
ST7
suppression of tumorigenicity 7
chr7
116588580
116591829
3249
4183
2298
ST7-AS1
ST7 antisense RNA 1
chr7
116588580
116591829
3249
-3748
-9662
ST7-OT4
ST7 overlapping transcript 4
chr3
52525050
52530899
5849
-1381
-30534
STAB1
stabilin 1
chr19
1204180
1210999
6819
1792
-20844
STK11
serine/threonine kinase 11
chr17
79978800
79981299
2499
723
-3470
STRA13
stimulated by retinoic acid 13
chr14
25514800
25521199
6399
1095
-236692
STXBP6
syntaxin binding protein 6 (amisyn)
chr14
53196100
53198649
2549
492
-44328
STYX
serine/threonine/tyrosine interacting protein
chr7
854830
858799
3969
563
-57724
SUN1
Sad1 and UNC84 domain containing 1
chr22
24572450
24574579
2129
-3929
-11559
SUSD2
sushi domain containing 2
chr19
15216650
15220329
3679
276
-7299
SYDE1
synapse defective 1, Rho GTPase, homolog 1 (C. elegans)
chr10
75407750
75417129
9379
-1653
-7795
SYNPO2L
synaptopodin 2-like
chr11
61343900
61345999
2099
3348
-62162
SYT7
synaptotagmin VII
chr22
20007100
20014429
7329
2134
-42682
TANGO2
transport and golgi organization 2 homolog (Drosophila)
chr16
30381630
30385179
3549
-1883
-14983
TBC1D10B
TBC1 domain family, member 10B
chr17
45769930
45775129
5199
-100
-16897
TBKBP1
TBK1 binding protein 1
chr16
2020650
2023879
3229
201
-6486
TBL3
transducin (beta)-like 3
chr20
62683150
62687899
4749
-2914
-18174
TCEA2
transcription elongation factor A (SII), 2
chr12
117534350
117536479
2129
1836
-58685
TESC
tescalcin
chr11
64070830
64074629
3799
4867
492
TEX40
testis expressed 40
chr17
76917250
76924629
7379
532
-71877
TIMP2
TIMP metallopeptidase inhibitor 2
chr1
32033050
32056249
23199
2564
-8635
TINAGL1
tubulointerstitial nephritis antigen-like 1
chr16
84536800
84540049
3249
-137
-28457
TLDC1
TBC/LysM-associated domain containing 1
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Supplementary Table 5. Continued chr19
3027000
3031149
4149
90
-31438
TLE2
transducin-like enhancer of split 2
chr17
76122080
76127979
5899
-169
-16031
TMC6
transmembrane channel-like 6
chr17
76122080
76127979
5899
-1829
-14019
TMC8
transmembrane channel-like 8
chr7
75615600
75625479
9879
3452
-4384
TMEM120A
transmembrane protein 120A
chr12
122149000
122153979
4979
832
-68482
TMEM120B
transmembrane protein 120B
chr7
142981380
142987279
5899
2267
-810
TMEM139
transmembrane protein 139
chr9
139685050
139687199
2149
348
-1643
TMEM141
transmembrane protein 141
chr4
153603800
153605849
2049
-3634
-57554
TMEM154
transmembrane protein 154
chr19
47548900
47552299
3399
1282
-1432
TMEM160
transmembrane protein 160
chr1
29448600
29453049
4449
-418
-4885
TMEM200B
transmembrane protein 200B
chr1
9647400
9657229
9829
3338
-12691
TMEM201
transmembrane protein 201
chr14
67979550
67982249
2699
1121
-43916
TMEM229B
transmembrane protein 229B
chr1
45138330
45141879
3549
-6
-20603
TMEM53
transmembrane protein 53
chr1
16064180
16066199
2019
-3727
-9285
TMEM82
transmembrane protein 82
chr1
1361650
1363849
2199
1242
-416
TMEM88B
transmembrane protein 88B
chr16
427600
430899
3299
2700
-8473
TMEM8A
transmembrane protein 8A
chr3
46741050
46744679
3629
42
-9548
TMIE
transmembrane inner ear
chr14
103586900
103589979
3079
-4224
-15336
TNFAIP2
tumor necrosis factor, alpha-induced protein 2
chr19
4638100
4640149
2049
-405
-14827
TNFAIP8L1
tumor necrosis factor, alpha-induced protein 8-like 1
chr1
1149280
1152279
2999
-1268
-4073
TNFRSF4
tumor necrosis factor receptor superfamily, member 4
chr11
1859750
1865449
5699
1168
-310
TNNI2
troponin I type 2 (skeletal, fast)
chr11
1940500
1949349
8849
4126
-15010
TNNT3
troponin T type 3 (skeletal, fast)
chr9
140167930
140173429
5499
-1600
-6412
TOR4A
torsin family 4, member A
chr1
3660900
3663879
2979
1496
-9840
TP73-AS1
TP73 antisense RNA 1
chr11
64083730
64089799
6069
-1732
-2597
TRMT112
tRNA methyltransferase 11-2 homolog (S. cerevisiae)
chr8
125461880
125464499
2619
142
-2074
TRMT12
tRNA methyltransferase 12 homolog (S. cerevisiae)
chr22
20103200
20105499
2299
418
-4952
TRMT2A
tRNA methyltransferase 2 homolog A (S. cerevisiae)
chr14
103992800
103997049
4249
-584
-8483
TRMT61A
tRNA methyltransferase 61A
chr11
842750
847999
5249
929
-21741
TSPAN4
tetraspanin 4
chr17
2238380
2242929
4549
23
-14663
TSR1
TSR1, 20S rRNA accumulation, homolog (S. cerevisiae)
chr7
2665930
2668029
2099
-4623
-37447
TTYH3
tweety family member 3
chr11
8106800
8108899
2099
4941
-19803
TUB
tubby bipartite transcription factor
chr15
41849380
41851899
2519
-592
-20886
TYRO3
TYRO3 protein tyrosine kinase
chr19
56159080
56163729
4649
-4011
-24676
U2AF2
U2 small nuclear RNA auxiliary factor 2
chr17
74439750
74449399
9649
4713
-58961
UBE2O
ubiquitin-conjugating enzyme E2O
chr3
48645280
48647579
2299
668
-9998
UQCRC1
ubiquinol-cytochrome c reductase core protein I
8
chr3
49156450
49159799
3349
88
-12017
USP19
ubiquitin specific peptidase 19
chr8
145652030
145657279
5249
-728
-5655
VPS28
vacuolar protein sorting 28 homolog (S. cerevisiae)
chr11
64862300
64865579
3279
257
-15240
VPS51
vacuolar protein sorting 51 homolog (S. cerevisiae)
171
PART THREE CHAPTER 8
Supplementary Table 5. Continued chr3
184532400
184534899
2499
3719
-236752
VPS8
vacuolar protein sorting 8 homolog (S. cerevisiae)
chr6
31742700
31745649
2949
933
-10803
VWA7
von Willebrand factor A domain containing 7
chr6
43478000
43481749
3749
4827
-310
YIPF3
Yip1 domain family, member 3
chr22
20115300
20119029
3729
-2200
-16793
ZDHHC8
zinc finger, DHHC-type containing 8
chr19
58985900
58989149
3249
-270
-5071
ZNF446
zinc finger protein 446
chr20
62598150
62605699
7549
-707
-13870
ZNF512B
zinc finger protein 512B
chr19
55986350
55992199
5849
1576
-6579
ZNF628
zinc finger protein 628
chr16
30795330
30800349
5019
683
-8070
ZNF629
zinc finger protein 629
chr16
30549450
30552249
2799
-4656
-8070
ZNF747
zinc finger protein 747
chr16
30536350
30540449
4099
-490
-3077
ZNF768
zinc finger protein 768
chr7
150079080
150082479
3399
4374
-14939
ZNF775
zinc finger protein 775
chr19
58788250
58791979
3729
-203
-17139
ZNF8
zinc finger protein 8
chr22
29277680
29279749
2069
-1175
-174760
ZNRF3
zinc and ring finger 3
chr7
76020330
76025699
5369
-3826
-48372
ZP3
zona pellucida glycoprotein 3 (sperm receptor)
TSS = transcription start site; TES = transcription end site.
172
CHROMATIN REGULATION IN CARDIOMYOPATHIES
Supplementary Table 6. Gene enrichment analysis related to genes with decreased H3K27ac occupancy within 5kb from transcription start site (TSS) GO: Biological Process No. ID
Name
p-value
q-value q-value q-value Genes Bonferroni FDR B&H FDR B&Y from Input
Genes in Concluding Annotation remark
1
GO:0008374
O-acyltransferase activity
1,87E-5
1,96E-2
1,96E-2
0,148
9
63
mitochondrial metabolism
2
GO:0016747
transferase activity, 4,59E-5 transferring acyl groups other than amino-acyl groups
4,83E-2
2,41E-2
0,182
16
208
mitochondrial metabolism
GO: Cellular Component No. ID
Name
p-value
q-value q-value q-value Genes Bonferroni FDR B&H FDR B&Y from Input
Genes in Concluding Annotation remark
1
mitochondrial matrix
6,76E-6
3,38E-3
339
No. ID
Name
p-value
q-value q-value q-value Genes Bonferroni FDR B&H FDR B&Y from Input
Genes in Concluding Annotation remark
1
BIOCYC: 142358
isoleucine degradation I 7,55E-6
1,01E-2
5,05E-3
3,93E-2
5
13
mitochondrial metabolism
2
REACTOME: 477135
Metabolism
1,75E-5
2,35E-2
7,82E-3
6,08E-2
62
1575
mitochondrial metabolism
3
WikiPathways: Fatty Acid Beta 198865 Oxidation
8,80E-7
1,18E-3
1,18E-3
9,17E-3
8
34
mitochondrial metabolism
GO:0005759
3,38E-3
0,023
23
mitochondrial metabolism
Pathway
GO = Gene Ontology; FDR B&H = False discovery rate, Benjamini and Hochberg concept; FDR B&Y = False discovery rate, Benjamini and Yekutieli concept.
8
173
PART THREE
|
Elucidating (Epi)genetic and Translating
Therapeutic Pathways
Chapter
9
Cell Therapy, a Novel Remedy for Dilated Cardiomyopathy? A Systematic Review Published as J Card Fail. 2013;19:494-502
Johannes M.I.H. Gho1, Gijs J.M. Kummeling1,2, Stefan Koudstaal1,2, Sanne J. Jansen of Lorkeers1, Pieter A. Doevendans1,2, Folkert W. Asselbergs1, Steven A.J. Chamuleau1,2 1
Department of Cardiology, Division Heart and Lungs, University Medical Center Utrecht, Utrecht, the Netherlands
2
Interuniversity Cardiology Institute of the Netherlands (ICIN), Utrecht, the Netherlands
PART THREE CHAPTER 9
ABSTRACT Background Dilated cardiomyopathy (DCM) is the most common form of non-ischemic cardiomyopathy worldwide and can lead to sudden cardiac death and heart failure. Despite ongoing advances made in the treatment of DCM, the improvement of outcome remains problematic. Stem cell therapy has been extensively studied in preclinical and clinical models of ischemic heart disease showing potential benefit. Dilated cardiomyopathy is associated with a major health burden and few studies have been performed on cell therapy for DCM. In this systematic review we aim to provide an overview of performed preclinical and clinical studies on cell therapy for DCM. Methods and Results A systematic search, critical appraisal and summarized outcomes are presented. In total 29 preclinical and 15 clinical studies were included. Methodological quality of reported studies in general was low based on the Centre for Evidence Based Medicine (CEBM) criteria, Oxford University. A large heterogeneity in inclusion criteria, procedural characteristics and used outcome measures was noted. The majority of studies showed a significant increase in left ventricular ejection fraction after cell therapy during follow-up. Conclusions Stem cell therapy has shown moderate but significant effects in clinical trials for ischemic heart disease, but it still remains to be determined if we can extrapolate these results to DCM patients. There is a need for methodological sound studies to elucidate underlying mechanisms and translate these into improved therapy for clinical practice. To validate safety and efficacy of cell therapy for DCM, adequate randomized (placebo) controlled trials using different strategies are mandatory.
176
CELL THERAPY FOR DILATED CARDIOMYOPATHY
INTRODUCTION Dilated cardiomyopathy (DCM) is the most common form of non-ischemic cardiomyopathy worldwide.1 Estimated prevalence in adults is 1/2,500 individuals, with an incidence of 7/100,000 per year. It is defined by dilatation and impaired contraction of heart muscle, in the absence of an obvious etiology such as ischemic heart disease or valve disease. Causes are multifactorial including genetics as major contributor. Dilated cardiomyopathy can lead to sudden cardiac death and heart failure, and carries a heavy burden on health care resources because of a high rate of hospital admission and occasional need for ventricular assist device and/or heart transplantation. Over the last years progress has been made in treating DCM, but improving outcome remains a difficult goal to achieve. Stem cell therapy for cardiac repair has been extensively studied in acute myocardial infarction and chronic ischemic heart disease showing potential benefit.2-4 Proposed working mechanisms include for example transdifferentiation of stem cells to cardiomyocytes, activation of intrinsic cardiac stem cells, paracrine effects and angiogenesis. Only a few studies have been reported on cell therapy for DCM, with first-in-man trials since 2006. To the best of our knowledge, no complete overview has been reported so far. In this systematic review, preclinical and clinical studies on cell therapy for DCM are reported including validity, procedural characteristics, safety and outcome measures.
METHODS Data collection and analysis We searched PubMed and Embase for reports of preclinical (all years) and clinical (2006-2012) therapeutic studies on various types of cell therapy for non-ischemic DCM (search syntaxes in Supplementary data, Table S1-2). In order to prevent selection and retrieval bias, outcome was not included in the search syntaxes. Only English and German articles were included, no
9
restrictions were imposed on timing of assessment and follow-up. Excluded were abstracts, reviews, reports from scientific sessions and discussions, case reports, in vitro studies, studies only on ischemic heart disease, pediatric studies and articles describing study design. Retrieved studies were used for cross-reference checking and carefully examined to exclude potentially duplicate or overlapping data. Reviews and case reports were available for background information. A separate search was performed on ClinicalTrials.gov for ongoing trials (Supplementary Table S3). Authors were contacted for missing data. Assessment of quality and risk of bias For assessment of quality, risk of bias (e.g., selection or information bias) and heterogeneity, remaining articles were considered for critical appraisal. Quality of studies were assessed in duplicate (by G.K. and J.G.) using standard forms for relevance and validity according to the Centre for Evidence Based Medicine, Oxford University (CEBM) criteria.5 Study design was determined, adequate randomization (concealed allocation) or comparable groups when not randomized, losses to follow-up and blinding of physician/treatment officer or researcher were reviewed.
177
PART THREE CHAPTER 9
Data extraction and management Study characteristics and used outcome measures were summarized. If necessary, data were estimated graphically or calculated with standard deviations from available data using SPSS 17.0 (SPSS, Chicago, IL, USA). Two-tailed p-values <0.05 were considered significant.
RESULTS Preclinical studies Literature search Our preclinical search yielded 1455 unique articles (Figure 1). After screening of title and abstract using in- and exclusion criteria 29 relevant articles remained, reference checking retrieved no new articles. Study characteristics and quality assessment Quality assessment of included studies is depicted in Supplementary data (Table S4). In summary, around half (15/29) was conducted in a randomized fashion, one study (Jin et al.) reported allocation concealment.6 One study reported blinding of the treatment officer.7 Twelve studies reported blinding of the effect investigators for treatment. All relevant data of the 29 studies are depicted in Table 1. Two studies used a large animal model (Hata et al.: dogs, Psaltis et al.: sheep), all other studies used rodent or rabbit models. Delivery route varied between
Figure 1. Flowchart of search strategy preclinical studies. ENG = English GER = German DCM = dilated cardiomyopathy.
178
CELL THERAPY FOR DILATED CARDIOMYOPATHY
intracoronary or intravenous infusion, intramuscular injection, sheet-graft or surgical injections and in total 10 different stem cell types were used. Fractional shortening, fractional area change and ejection fraction were determined by (2D)-echocardiography in almost all studies, only Psaltis et al. used MRI to determine ejection fraction. Delta pressure/delta time (dP/dt) and peak systolic pressure were measured using hemodynamic measurements. Five studies reported baseline data, showing no significant differences in the study groups. Twenty-one out of 29 studies had sufficient and comparable proportions of follow-up and length of follow-up varied between 2-21 weeks. Lin et al., Sun et al. and Kondoh et al. (2006 and 2007) reported interim results on ejection fraction or fractional shortening. Most studies (24/29) used comparable treatment methods between control and therapy groups. Eighteen out of 29 studies (62%) were published in a journal with impact factor ≥ 3.00. Efficacy and safety in animal models All animal studies demonstrated a positive effect of stem cell therapy on left ventricular function, of which 24 studies showed significant improvement, ranging from 3% to 24% on echocardiographic evaluation (Table 1). In general, no safety issues were found in these studies. Of 18 studies that assessed mortality, 3 studies found improvement in survival.8-10 Ten studies found equal mortality between treatment and control groups, other studies did not mention significance and no cardiac tamponade was reported. Four studies detected no arrhythmias, other studies did not mention any occurrence of arrhythmias.10-13 Clinical studies Literature search Our clinical search strategy yielded 339 unique hits (Figure 2). Screening of title and abstract using in- and exclusion criteria resulted in 16 relevant articles. One relevant peer-reviewed article was found not indexed on PubMed or Embase.14 All principle investigators from ongoing trials were contacted for up-to-date trial results (see ongoing clinical trials section), this resulted in 2 additional complete manuscripts.15, 16 Reference checking retrieved no new articles. Four articles
9
were excluded because of overlapping data. In total 15 articles were considered for critical appraisal (4 RCTs, 2 non-randomized controlled trials, 9 cohort studies). Authors were contacted for articles not full-text available and missing data (8 authors responded). Study characteristics and quality assessment Four out of 15 studies were randomized, 2 non-randomized controlled trials used matched control groups (Table 2). Data in Supplementary material shows further methodological quality of the included studies (Table S5). Allocation concealment was not reported in the RCTs. There was no patient or physician blinding performed in these clinical studies, nor placebo used in the controlled trials. Two of the controlled trials described equally treated groups aside from the allocated treatment.17, 18 Losses to follow-up were reported and adequate (≤20%) in 13 out of 15 studies. All articles were published in peer-reviewed journals, 5 of the 15 studies (33%) in a journal with impact factor ≥ 3.00. Sample size ranged from 3-131 persons and overall follow-up duration from 3-60 months.
179
180 67/20
35/18 120/40
3
3
Lu (2006) 46
41/11 26/14
2
2
Pouly (2004) 52
Ohno (2002)
51
25/16
3 ?/15
30/10
1
2
Nomura (2006) 50
Nakajima (2008)
Nagaya (2005) 48
Mu (2011)
47
40/8
1
45
Lin (2010)
100/24
5
85/57
5
Kondoh (2007) 44
20/10
Kondoh (2006) 9
49
52/18
3
3
38/14
Jin (2010) 6
Ishida (2004)
3
Garbade (2009) 42
43
31/15
3
Dhein (2006) 7
19/6 160/?
3
3
De Angelis (2010) 8
24/8
Chen (2010) 41
3
15/7
3
3
12/5
n= (total/cell treated)
4
Chen (2006) 13
30
Model
Baba (2007) 11
Small animals
Psaltis (2010)
Hata (2006) 29
Large animals
Author
Table 1. Preclinical study characteristics and outcomes
Yes
No
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
No
Yes
No
No
Yes
No
Randomized
4
5
4
4
4
4
4
13
8
12
4
4
4
4
6
4
4
4
8
4
Follow-up (weeks)
MB
MB
MDPC
BMMC
MSC
BM-MSC
BMMC
ADMSC
MB
MB
BMMC
BMMC
BMMC
BM-MSC
CPC
BM-MSC
BMMC
ES
MPC
MB
Cell type
a
FAC
PSP
FS
FS
FS
dP/dT
EF
EF
FS
FS
dP/dt
FS
FS
dP/dT
EF
EF
EF
dP/dT
EF
EF
Outcome
BF, rel
CT, abs
CT, abs
BF, abs
CT, abs
CT, rel
CT, abs
CT, abs
CT, rel
CT, abs
CT, rel
CT, abs
CT, abs
CT, rel
CT, abs
CT, abs
CT, abs
CT, rel
CT, abs
CT, abs
Measuring method
+ 24%
≈ + 68%
≈ + 5%
+ 3.1%
≈ + 11%
b
+ 32%/- 18%
+ 11%
+ 4,1%
+ 31%
≈ + 5%
+ 35%/- 32%
+ 4.1%
+ 3%
+ 62%/- 88%
≈ + 24%
+ 6.4%
+ 11.1%
+ 27%/- 20%
+ 10%
+ 6.5%
Measured effect
0.019
<0.001
<0.05
<0.05
<0.05
<0.05
<0.05
<0.0001
N.S.
<0.05
<0.05
N.S.
<0.01
<0.05
<0.05
<0.05
<0.05
“significant”
<0.05
<0.05
P-value
PART THREE CHAPTER 9
25/12 118/8
2
1
6
12
Werner(2005) 57
Zhou (2007)
42/18 20/10
3
28/10
3
13 4 15 4 21 4 12
Yes No No No Yes Yes Yes
Determined using MRI. bAverage difference in PSP c Singla used 6-8 animals per group, 1 group underwent cell therapy.
a
MSC
BMMC
BMMC
ES
EPC
MB
MB
BMMC
CM or ES
FS
EF
FS
FS
FS
dP/dT
dP/dT
EF
FS
Cell type ADMSC: adipose-derived stem cells BM(M)C: bone marrow (mononuclear) cells CM: conditioned medium CPC: cardiac progenitor cells EPC: endothelial progenitor cells ES: embryonic stem cells MB: myoblasts MDPC: myosphere-derived progenitor cell MPC: mesenchymal precursor cells MSC: mesenchymal stem cells
4 2
No No
Abbreviations: N.S. = non-significant EF = ejection fraction dP/dt = delta pressure/delta time FS = fractional shortening FAC = fractional area change PSP = peak systolic pressure.
Disease model 1 porcine myosin + Freund’s adjuvant 2 genetic δ-sarcoglycan deficiency 3 doxorubicin or adriamycin 4 pacing-induced DCM (230/min) 5 BIO TO-2 hamsters 6 Kir 6.2 knockout + aortic constriction
59
Zhang (2005) 58
Yamada (2008)
10
?/?
3
Suzuki (2001) 56
Tezuka (2008)
16/8
1
Sun (2009) 55
?/? c
3
Singla (2012) 54
33/15
2
Shabbir (2009) 53 CT, abs
≈ + 13%
+ 0.1%
+ 3.8%
+ 12.5%
+ 15%
+ 15%/- 37%
+ 20%/- 20%
+ 5.5%
+ 5%
<0.001
N.S.
N.S.
<0.05
unknown
<0.05
0.01
0.0002
<0.05
≈ italic = estimated effect based on graphics. dP/dt: measured effect (maximum / minimum).
abs = absolute difference rel = relative difference
Measuring method BF = baseline vs follow-up CT = control vs therapy
CT, abs
CT, abs
BF, abs
BF, rel
CT, rel
CT, rel
CT, abs
CT, abs
CELL THERAPY FOR DILATED CARDIOMYOPATHY
9
181
PART THREE CHAPTER 9
Figure 2. Flowchart of search strategy clinical trials. ENG = English GER = German DCM = dilated cardiomyopathy.
Definitions used for DCM were different between studies. Seven studies reported excluding patients with coronary artery disease on coronary angiography.14, 16, 18-22 Only Schannwell et al. performed endomyocardial biopsy in all patients to exclude myocarditis, Seth et al. performed biopsies in 8/85 patients. Three studies included patients with ischemic cardiomyopathy in a separate group, these subgroups were excluded for this review.14, 23, 24 For Fatkhutdinov et al., the groups undergoing concomitant (valve) surgery were also excluded. For cell delivery, the majority used an intracoronary route versus a surgical procedure with intra-myocardial injections (11 vs. 4). Cell type was different between studies, mostly bone marrow stem cells were used, two studies used fetal material. Kirillov et al. had one group with prenatal allogenic skeletal myoblasts and one with prenatal mesenchymal stem cells.25 Efficacy outcomes in clinical trials Main outcomes reported in clinical trials were left ventricular ejection fraction (LVEF), New York Heart Association (NYHA) class, Quality of Life (QoL) and mortality. All but one article mainly used (2D)-echocardiography for determining ejection fraction, Fischer-Rasokat et al. used angiography.26 Cardiac magnetic resonance imaging (MRI) was only used in certain subgroups. Mean difference in LVEF between baseline and final follow-up values is shown in Table 2 (baseline and final LVEF values are shown in Supplementary material, Table S6). LVEF values were calculated for Arg端ero et al. from available patient data, LVEF was estimated in the therapy groups of Fatkhutdinov et al. and the control groups of Vrtovec et al. (2011 and 2013) as these were only graphically depicted. The majority of studies (10/15) showed a significant increase in LVEF in the therapy group (range 3.2% to 11%). Four studies found no significant difference between baseline and follow-up in the therapy groups.17, 20, 24, 25 Kalil et al. showed a significant
182
0.03 unknown
+ 5.9 + 0.4 + 4.6 ≈-2 + 5.7 ≈-3 ≈0 ≈-5 +9 0 + 8.7
12 60
unknown 40 12 12 12 6 6 ± 13b 12
41 28(61)c 27 55(131)c 55 6(14) 8 10(20) 10 3(5)d 10 3 33 9 9
52 27
controls therapy controls therapy controls therapy controls therapy controls
group 1 group 2
Argüero (2006) 23
Benetti (2010) 19
Chin (2011) 24
Fischer-Rasokat (2009) 26
Kalil (2008) 28
Kirillov (2007) 25
Martino (2010) 20
Ruengsakulrach (2011) 14
Suarez de Lezo (2013) 16
BMC-MN BMMC
IC
Cell type ACP = angiogenic cell precursors ASM = allogenic skeletal myoblasts BM(M)C = bone marrow (mononuclear) cells HFDSC = human fetal-derived stem cells MSC = mesenchymal stem cell MN = mononuclear
ACP
Surgical
0.001
0.03
N.S.
N.S.
N.S.
0.023
<0.001
N.S.
0.005
0.0059
N.S.
<0.01
N.S.
N.S.
unknown
0.02
N.S.
<0.05
0.04
Other abbreviations EF = ejection fraction G-CSF = granulocyte colony-stimulating factor IC = intracoronary N.S. = non-significant ≈ italic = estimated difference based on graphics.
+ 11
+ 4.4
+ 1.2
+ 1.6
prenatal MSC IC
7
- 0.5
+ 3.2
+ 15.1
+ 8.2
24
BMC
BMC
BM-MSC
HFDSC
autologous CD34+
BMC-MN
allogenic MSC
autologous CD34+
autologous CD34+
BMC
+2
IC
Surgical
IC
IC
Surgical
Surgical
IC
IC
IC
IC
IC
prenatal ASM
3
12
Design RCT = randomized controlled trial Non-RCT = non-randomized controlled trial Cohort = cohort study a Randomized to G-CSF administration (14) or G-CSF associated to BMSC intracoronary infusion (8). b Bocchi et al. mean 468±374 days (up to 1190 days), Ruengsakulrach et al. mean 409.7±352.4 days. c All patients received G-CSF before randomization, n = 6 (2011) and n = 21 (2013) patients were excluded in this phase. d 3/5 patients included in preliminary report, outcomes calculated.
Cohort
Schannwell (2008) 18
Fatkhutdinov (2010) 17
Non-RCTs
Vrtovec (2013) 15
Vrtovec (2011) 22
36
0.016
44(85)
+ 8.8
therapy
Seth (2010) 21
BMC + 5.6
IC
± 15b
Bocchi (2010) 27
RCTs 8
P-value
14(23)a
EF mean difference (%)
controls
Cell type
therapy
Delivery route
Follow up (months)
Group
n(total)
Author (year)
Table 2. Clinical study characteristics and outcomes
CELL THERAPY FOR DILATED CARDIOMYOPATHY
9
183
PART THREE CHAPTER 9
decrease of 0.5% (p = 0.023) in mean ejection fraction at final follow up (12 months), in this study LVEF did increase in the first two months after surgical cell therapy but decreased afterwards. In the control groups increase of mean LVEF from baseline was less, or LVEF decreased (range ≈ -5% to 5.6%). The highest increase was found in the control group of Bocchi et al., which underwent G-CSF administration. Seth et al. showed a significant difference (p = <0.05) between therapy and controls (LVEF: 5.9% vs. 0.4%) and Vrtovec et al. found a significant change in LVEF mean difference in both studies.15, 21, 22 NYHA class improved in all 9 studies which reported functional class, with significant improvements in more than half of them.14, 17, 19, 20, 27 Quality of life was determined by using questionnaires and was reported to be significantly improved in six studies.14, 19-21, 27, 28 Fatkhutdinov et al. found an improvement in QoL at three months during follow up, but at final follow up a non-significant difference. Safety outcomes in clinical trials Mortality was reported in all studies except for Bocchi et al. Comparing therapy and control groups, Vrtovec et al. (2011) found a significantly lower incidence of heart transplantation or mortality in the therapy group (7% vs. 30%, p = 0.03). At 5-year follow-up, Vrtovec et al. (2013) found a lower total mortality in patients receiving stem cell therapy compared to controls (14% vs. 35%, p = 0.01), occurrence of heart transplantation did not differ between the 2 groups. The other controlled studies found similar mortality rates. Chin et al., Fatkhutdinov et al. and Schannwell et al. found no mortality in both groups and of the 85 patients included by Seth et al. 12 treated patients vs. 14 controls had died after 3 years of follow-up. In the study by Argüero et al., which mainly included patients with ischemic cardiomyopathy, two patients died because of intractable arrhythmias, but there was no information given on to which group these patients belonged. Benetti et al., Kalil et al. and Martino et al. reported some cases with de novo (supra) ventricular arrhythmias, quickly terminated with chemical or electrical cardioversion. The other studies did not show an increase in arrhythmias (only Seth et al. gave no information on occurrence of arrhythmias).
DISCUSSION In total 29 preclinical and 15 clinical studies were included in this systematic review on cell therapy for DCM. Main findings were: (i) a low methodological quality in general; (ii) a large heterogeneity in inclusion criteria, procedural characteristics and used outcome measures; (iii) only 2 preclinical studies involving large animals have been reported; (iv) the majority of clinical studies (from in total 513 enrolled patients) showed a significant increase in LVEF, improvement in NYHA class and QoL at follow-up after cell therapy; (v) several reports on potentially lifethreatening arrhythmias; (vi) there is no increased mortality after cell therapy. From “bench-to-bedside” Study validity in preclinical studies showed a lack in methodological quality, around half was randomized and only one study used concealed allocation. No blinding was performed in the
184
CELL THERAPY FOR DILATED CARDIOMYOPATHY
majority of included articles. We found a large heterogeneity between studies in disease models, therapy and used endpoints. It is therefore difficult to combine the results in order to infer a stronger conclusion on stem cell therapy. However, all studies did show a positive effect on left ventricular function, mostly by attenuation of deterioration, also shown in the studies reporting interim results (Figure 3). Largest improvement was found by De Angelis et al. using CPCâ&#x20AC;&#x2122;s on doxorubicine-induced cardiomyopathy in rats. Contractility was also significantly and positively influenced. Comparing stem cell treated animals with controls, 10 out of 29 studies reported equal mortality or sometimes even improvement of survival was reported. No conclusion can be drawn regarding arrhythmias, while it was reported in only 4 studies. Main proposed mechanisms of stem cell effects on DCM include reduction of fibrosis and angiogenesis by direct (cell engraftment, regenerative transdifferentiation) and paracrine support.29, 30 No additional fibrosis is seen following therapy, sometimes even attenuation is witnessed. It remains possible that publication bias and methods of measurement attributed to an overestimation and positive image of stem cell therapy for DCM. Left ventricular (LV) hemodynamics and MRI are more objective than 2D-echocardiography in measuring LV-function. An optimal animal model for DCM is still lacking. Overall, stem cell therapy seems feasible and effective in preclinical models, however large animal studies are scarce in the transition from bench-to-bedside. Back from â&#x20AC;&#x153;bedside-to-benchâ&#x20AC;? Quality assessment, safety and efficacy in clinical trials In 2006 consensus of the task force of the European Society of Cardiology (ESC) addressed the possible need for autologous cell therapy in DCM.31 In the same year first-in-man clinical trials were reported and in summary these studies were predominantly pioneering trials assessing safety and efficacy. There is a lack in methodological quality, as mainly non-randomized trials were performed. Lack of allocation concealment, placebo effect or procedural characteristics such as delivery techniques and growth factor administration could have influenced the outcome. Especially studies with a short follow-up period should be interpreted with caution whether the positive outcome would be sustained over a longer period. We noticed a large heterogeneity in
9
inclusion criteria, procedural characteristics and used outcome measures.
Figure 3. Preclinical studies reporting interim results on fractional shortening (A) or left ventricular ejection fraction (B). FS = fractional shortening LVEF = left ventricular ejection fraction.
185
PART THREE CHAPTER 9
These studies showed no increase in mortality in the cell therapy groups. There were several reports on potentially life-threatening arrhythmias.19, 20, 23, 28 Furthermore, a case report (n=2) describing a pilot trial on cell therapy in DCM did show severe complications after intracoronary balloon inflation, with hypotension and ventricular arrhythmias in one patient and severe spasm with hypotension requiring bailout stenting in another, therefore this trial was prematurely terminated.32 So these procedures do carry an arrhythmic risk, which should be assessed in future studies. There seems to be a positive effect of cell therapy for DCM on LVEF, NYHA class and QoL, although this should be verified in large, randomized trials. Also more studies are needed for determining optimal delivery route and cell type for cardiac repair. Outcomes should be given as absolute differences, in order to more accurately estimate actual treatment effect. This would result in clear indications for cell therapy in non-ischemic cardiomyopathy, preferable above random patient selection. Using multiple “rising” imaging modalities such as MRI, (3D-) echocardiography and single-photon emission computed tomography (SPECT) would provide more accuracy for determining outcomes.33 Although we would prefer MRI to determine outcome measures, certain heart failure patients would be ineligible because of implanted devices such as implantable cardioverter-defibrillator (ICD) and cardiac resynchronization therapy (CRT-D). Besides clinically relevant LVEF measurements, detailed measuring of local cardiac function can provide information about local changes induced by cardiac regenerative therapy.34 Combined global and regional myocardial function measurements are recommended to assess the effects of cardiac regenerative therapy. Elucidating underlying mechanisms Current evidence shows controversy on stem cell transdifferentiation to cardiomyocytes contributing to cardiac functional gain, which led to investigating paracrine mechanisms for repair.35, 36 In contrast to cell therapy for ischemic cardiomyopathy where angiogenesis plays an important role, in DCM heart muscle disease and myocardial dysfunction poses a different background. However, marked vascular derangements and impaired vasculogenic and angiogenic responses have also been reported in idiopathic DCM.37 Proposed underlying working mechanisms should be differentiated between ischemic cardiomyopathy and DCM, for example by elucidating these issues in preclinical studies. Findings from clinical studies should also be clarified in collaboration with preclinical research. As progress is being made for DCM in unraveling underlying genetic mutations and molecular mechanisms (e.g., calcium metabolism), these could be used in future for tailored cell and/or gene therapy.38, 39 Ongoing clinical trials Several clinical trials on cell therapy for DCM are underway (Table 3). Nine trials were found on ClinicalTrials.gov, see Supplementary data for search syntax (Table S3). All principle investigators were contacted for up-to-date trial results, this resulted in 2 complete manuscripts which have been added to the clinical studies section. Most studies are randomized, only one study is performed double-blind and placebo controlled with another control group receiving G-CSF. Delivery routes are intracoronary, intra-myocardial or transendocardial (mainly using the NOGA® XP/MyoStar™ injection catheter system) and different cell types are used. Primary outcome measures are assessing safety or LVEF and results are expected.
186
RCT
RCT
RCT
RCT
RCT
RCT
RCT
NCT01302171 (REGENERATE-DCM)
NCT01350310 (NOGA-DCM)
NCT00765518 (IMPACT-DCM)
NCT01392625 (POSEIDON-DCM)
NCT01020968
NCT00333827 (MiHeart)
NCT00721045 (Revascor) single-blind
double-blind
open-label
open-label
open-label
single-blind
double-blind
Blinding
Cell type
BM-MNC vs. Placebo/G-CSF CD34+ CRCs autologous vs. allogenic MSC CRCs BMC MPCs
Delivery route IC IC vs. IM IM TE TE IC TE
safety
LVEF
safety
safety
safety
LVEF (echocardiography)
LVEF (MRI)
Primary outcome measure (modality)
Jul 2011
Feb 2009
Feb 2012
Jul 2015
Feb 2012
Mar 2013
Dec 2012
(Estimated) Completion date
unknown
unknown
active, not recruiting
recruiting
completed
recruiting
recruiting
Status
Abbreviations: RCT = randomized controlled trial IC = intracoronary IM = intra-myocardial TE = transendocardial BM-MNC = bone marrow mononuclear cells G-CSF = granulocyte colony-stimulating factor CRCs = cardiac repair cells BMC = bone marrow cells MPCs = mesynchymal precursor cells MSC = mesenchymal stem cells LVEF = left ventricular ejection fraction MRI = magnetic resonance imaging.
Design
Identifier (short name)
Table 3. Ongoing clinical trials on cell therapy for dilated cardiomyopathy
CELL THERAPY FOR DILATED CARDIOMYOPATHY
9
187
PART THREE CHAPTER 9
Limitations The large heterogeneity in study characteristics of included reports could have influenced outcome, therefore we examined reasons for heterogeneity using standardized criteria. Differences between animal DCM models and human disease carries limitations when translating from bench-to-bedside. Publication bias cannot be ruled out as smaller studies are prone to publication bias. Estimates of effect are vulnerable to confounding by lack of randomization. Only reported outcomes were included as we did not have access to complete individual data and, accordingly, mean values are provided; authors were contacted in case of missing data. Future implications Only a few studies on cell therapy for DCM have been performed until now and methodological quality in general was low. The majority of studies show a significant positive effect on LVEF. There is a need for methodological sound (pre)clinical studies to elucidate underlying mechanisms and translate these into improved therapy for clinical practice. Clearly described uniform definitions for inclusion should be used in future studies, for example derived from the ESC classification.40 Adequately randomized (placebo) controlled trials are mandatory to validate safety and efficacy of cell therapy for DCM, preferably with long term follow-up. These studies should focus on occurrence of arrhythmias, mortality, optimal cell delivery techniques and cell type using different modern imaging modalities. Stem cell therapy has shown moderate but significant effects in clinical trials for ischemic heart disease, compromising nearly 2000 patients.4 It still remains to be determined if we can extrapolate these results to DCM patients. Nevertheless, this exciting field of research has shown promising results so far, and warrants continuation of ongoing study programs.
188
CELL THERAPY FOR DILATED CARDIOMYOPATHY
REFERENCES 1. 2.
3.
4.
5. 6. 7.
8.
9.
10. 11.
12. 13.
14. 15.
16.
17.
18.
19.
Jefferies JL, Towbin JA. Dilated cardiomyopathy. Lancet. 2010;375:752-62. van der Spoel TI, Jansen of Lorkeers SJ, Agostoni P, van Belle E, Gyongyosi M, Sluijter JP, et al. Human relevance of pre-clinical studies in stem cell therapy: systematic review and metaanalysis of large animal models of ischaemic heart disease. Cardiovasc Res. 2011;91:649-58. Strauer BE, Steinhoff G. 10 years of intracoronary and intramyocardial bone marrow stem cell therapy of the heart: from the methodological origin to clinical practice. J Am Coll Cardiol. 2011;58:1095-104. Zimmet H, Porapakkham P, Porapakkham P, Sata Y, Haas SJ, Itescu S, et al. Short- and longterm outcomes of intracoronary and endogenously mobilized bone marrow stem cells in the treatment of ST-segment elevation myocardial infarction: a meta-analysis of randomized control trials. Eur J Heart Fail. 2012;14:91-105. Centre for Evidence Based Medicine, Oxford University. EBM Tools. Critical appraisal sheets. [14 Feb 2012]; Available from: http://www.cebm.net/index.aspx?o=1157 (14 February 2012). Jin B, Luo XP, Ni HC, Li Y, Shi HM. Cardiac matrix remodeling following intracoronary cell transplantation in dilated cardiomyopathic rabbits. Mol Biol Rep. 2010;37:3037-42. Dhein S, Garbade J, Rouabah D, Abraham G, Ungemach FR, Schneider K, et al. Effects of autologous bone marrow stem cell transplantation on beta-adrenoceptor density and electrical activation pattern in a rabbit model of non-ischemic heart failure. J Cardiothorac Surg. 2006;1:17. De Angelis A, Piegari E, Cappetta D, Marino L, Filippelli A, Berrino L, et al. Anthracycline cardiomyopathy is mediated by depletion of the cardiac stem cell pool and is rescued by restoration of progenitor cell function. Circulation. 2010;121:276-92. Kondoh H, Sawa Y, Miyagawa S, Sakakida-Kitagawa S, Memon IA, Kawaguchi N, et al. Longer preservation of cardiac performance by sheet-shaped myoblast implantation in dilated cardiomyopathic hamsters. Cardiovasc Res. 2006;69:466-75. Yamada S, Nelson TJ, Crespo-Diaz RJ, Perez-Terzic C, Liu XK, Miki T, et al. Embryonic stem cell therapy of heart failure in genetic cardiomyopathy. Stem Cells. 2008;26:2644-53. Baba S, Heike T, Yoshimoto M, Umeda K, Doi H, Iwasa T, et al. Flk1(+) cardiac stem/progenitor cells derived from embryonic stem cells improve cardiac function in a dilated cardiomyopathy mouse model. Cardiovasc Res. 2007;76:119-31. Tezuka A, Kawada T, Nakazawa M, Masui F, Konno S, Nitta S, et al. Which skeletal myoblasts and how to be transplanted for cardiac repair? Biochem Biophys Res Commun. 2008;369:270-6. Chen M, Fan ZC, Liu XJ, Deng JL, Yang Q, Huang DJ. Cell transplantation with a catheter-based approach: an efficient method for the treatment of heart failure with multiple lesions. Cell Prolif. 2006;39:471-7. Ruengsakulrach P, Visudharom K, Chaothawee L, Belkin M. Peripheral Blood Stem Cell for Cardiomyopathy. Bangkok Med J. 2011;2:19-29. Vrtovec B, Poglajen G, Lezaic L, Sever M, Domanovic D, Cernelc P, et al. Effects of Intracoronary CD34+ Stem Cell Transplantation in Nonischemic Dilated Cardiomyopathy Patients: 5-Year Follow-Up. Circ Res. 2013;112:165-73. Suárez de Lezo J, Herrera C, Romero M, Pan M, Suárez de Lezo J, Carmona MD, et al. Functional improvement in patients with dilated cardiomyopathy after the intracoronary infusion of autologous bone marrow mononuclear cells. Rev Esp Cardiol (Engl Ed). 2013;66:450-7. Fatkhutdinov T, D’Yachkov A V, Koroteyev AV, Goldstein DV, Bochkov NP. Safety and efficiency of transplantation of allogenic multipotent stromal cells in surgical treatment of dilatated cardiomyopathy. Bull Exp Biol Med. 2010;149:119-24. Schannwell CM, Köstering M, Zeus T, Brehm M, Erdmann G, Fleissner T, et al. [Humane autologe intrakoronare stammzelltransplantation zur myokardregeneration bei dilatativer cardiomyopathie (NYHA stadium II-IV)]. The Düsseldorf Autologous Bone Marrow Cells in Dilated Cardiomyopathy Trial. ABCD Trial. J Kardiologie. 2008;15:23-30. Benetti F, Penherrera E, Maldonado T, Vera YD, Subramanian V, Geffner L. Direct myocardial implantation of human fetal stem cells in heart failure patients: long-term results. Heart Surg Forum. 2010;13:E31-5.
9
189
PART THREE CHAPTER 9
20. Martino HF, Oliveira PS, Souza FC, Costa PC, Assuncao ESE, Villela R, et al. A safety and feasibility study of cell therapy in dilated cardiomyopathy. Braz J Med Biol Res. 2010;43:989-95. 21. Seth S, Bhargava B, Narang R, Ray R, Mohanty S, Gulati G, et al. The ABCD (Autologous Bone Marrow Cells in Dilated Cardiomyopathy) trial a long-term follow-up study. J Am Coll Cardiol. 2010;55:1643-4. 22. Vrtovec B, Poglajen G, Sever M, Lezaic L, Domanovic D, Cernelc P, et al. Effects of intracoronary stem cell transplantation in patients with dilated cardiomyopathy. J Card Fail. 2011;17:272-81. 23. Arguero R, Careaga-Reyna G, Castano-Guerra R, Garrido-Garduno MH, Magana-Serrano JA, de Jesus Nambo-Lucio M. Cellular autotransplantation for ischemic and idiopathic dilated cardiomyopathy. Preliminary report. Arch Med Res. 2006;37:1010-4. 24. Chin SP, Poey AC, Wong CY, Chang SK, Tan CS, Ng MT, et al. Intramyocardial and intracoronary autologous bone marrow-derived mesenchymal stromal cell treatment in chronic severe dilated cardiomyopathy. Cytotherapy. 2011;13:814-21. 25. Kirillov AM, Fatkhudinov T, Dyachkov AV, Koroteev AV, Goldshtein DV, Bochkov NP. Transplantation of allogenic cells in the therapy of patients with dilated cardiomyopathy. Bull Exp Biol Med. 2007;144:635-9. 26. Fischer-Rasokat U, Assmus B, Seeger FH, Honold J, Leistner D, Fichtlscherer S, et al. A pilot trial to assess potential effects of selective intracoronary bone marrow-derived progenitor cell infusion in patients with nonischemic dilated cardiomyopathy: final 1-year results of the transplantation of progenitor cells and functional regeneration enhancement pilot trial in patients with nonischemic dilated cardiomyopathy. Circ Heart Fail. 2009;2:417-23. 27. Bocchi EA, Bacal F, Guimaraes G, Mendroni A, Mocelin A, Filho AE, et al. Granulocyte-colony stimulating factor or granulocyte-colony stimulating factor associated to stem cell intracoronary infusion effects in non ischemic refractory heart failure. Int J Cardiol. 2010;138:94-7. 28. Kalil RA, Ott D, Santâ&#x20AC;&#x2122;Anna R, Dias E, Marques-Pereira JP, Delgado-Canedo A, et al. Autologous transplantation of bone marrow mononuclear stem cells by mini-thoracotomy in dilated cardiomyopathy: technique and early results. Sao Paulo Med J. 2008;126:75-81. 29. Hata H, Matsumiya G, Miyagawa S, Kondoh H, Kawaguchi N, Matsuura N, et al. Grafted skeletal myoblast sheets attenuate myocardial remodeling in pacing-induced canine heart failure model. J Thorac Cardiovasc Surg. 2006;132:918-24. 30. Psaltis PJ, Carbone A, Nelson AJ, Lau DH, Jantzen T, Manavis J, et al. Reparative effects of allogeneic mesenchymal precursor cells delivered transendocardially in experimental nonischemic cardiomyopathy. JACC Cardiovasc Interv. 2010;3:974-83. 31. Bartunek J, Dimmeler S, Drexler H, Fernandez-Aviles F, Galinanes M, Janssens S, et al. The consensus of the task force of the European Society of Cardiology concerning the clinical investigation of the use of autologous adult stem cells for repair of the heart. Eur Heart J. 2006;27:1338-40. 32. Widimsky P, Penicka M. Complications after intracoronary stem cell transplantation in idiopathic dilated cardiomyopathy. Int J Cardiol. 2006;111:178-9. 33. Marwick TH, Raman SV, Carrio I, Bax JJ. Recent developments in heart failure imaging. JACC Cardiovasc Imaging. 2010;3:429-39. 34. van Slochteren FJ, Teske AJ, van der Spoel TI, Koudstaal S, Doevendans PA, Sluijter JP, et al. Advanced measurement techniques of regional myocardial function to assess the effects of cardiac regenerative therapy in different models of ischaemic cardiomyopathy. Eur Heart J Cardiovasc Imaging. 2012;13:808-18. 35. Segers VF, Lee RT. Stem-cell therapy for cardiac disease. Nature. 2008;451:937-42. 36. Maxeiner H, Krehbiehl N, Muller A, Woitasky N, Akinturk H, Muller M, et al. New insights into paracrine mechanisms of human cardiac progenitor cells. Eur J Heart Fail. 2010;12:730-7. 37. Roura S, Bayes-Genis A. Vascular dysfunction in idiopathic dilated cardiomyopathy. Nat Rev Cardiol. 2009;6:590-8. 38. Jessup M, Greenberg B, Mancini D, Cappola T, Pauly DF, Jaski B, et al. Calcium Upregulation by Percutaneous Administration of Gene Therapy in Cardiac Disease (CUPID): a phase 2 trial of intracoronary gene therapy of sarcoplasmic reticulum Ca2+-ATPase in patients with advanced heart failure. Circulation. 2011;124:304-13.
190
CELL THERAPY FOR DILATED CARDIOMYOPATHY
39. Sun N, Yazawa M, Liu J, Han L, Sanchez-Freire V, Abilez OJ, et al. Patient-specific induced pluripotent stem cells as a model for familial dilated cardiomyopathy. Sci Transl Med. 2012;4:130ra47. 40. Elliott P, Andersson B, Arbustini E, Bilinska Z, Cecchi F, Charron P, et al. Classification of the cardiomyopathies: a position statement from the European Society Of Cardiology Working Group on Myocardial and Pericardial Diseases. Eur Heart J. 2008;29:270-6. 41. Chen Y, Liu W, Li W, Gao C. Autologous bone marrow mesenchymal cell transplantation improves left ventricular function in a rabbit model of dilated cardiomyopathy. Exp Mol Pathol. 2010;88:311-5. 42. Garbade J, Dhein S, Lipinski C, Aupperle H, Arsalan M, Borger MA, et al. Bone marrow-derived stem cells attenuate impaired contractility and enhance capillary density in a rabbit model of Doxorubicin-induced failing hearts. J Card Surg. 2009;24:591-9. 43. Ishida M, Tomita S, Nakatani T, Fukuhara S, Hamamoto M, Nagaya N, et al. Bone marrow mononuclear cell transplantation had beneficial effects on doxorubicin-induced cardiomyopathy. J Heart Lung Transplant. 2004;23:436-45. 44. Kondoh H, Sawa Y, Fukushima N, Matsumiya G, Miyagawa S, Kitagawa-Sakakida S, et al. Combined strategy using myoblasts and hepatocyte growth factor in dilated cardiomyopathic hamsters. Ann Thorac Surg. 2007;84:134-41. 45. Lin YC, Leu S, Sun CK, Yen CH, Kao YH, Chang LT, et al. Early combined treatment with sildenafil and adipose-derived mesenchymal stem cells preserves heart function in rat dilated cardiomyopathy. J Transl Med. 2010;8:88. 46. Lu C, Arai M, Misao Y, Chen X, Wang N, Onogi H, et al. Autologous bone marrow cell transplantation improves left ventricular function in rabbit hearts with cardiomyopathy via myocardial regeneration-unrelated mechanisms. Heart Vessels. 2006;21:180-7. 47. Mu Y, Cao G, Zeng Q, Li Y. Transplantation of induced bone marrow mesenchymal stem cells improves the cardiac function of rabbits with dilated cardiomyopathy via upregulation of vascular endothelial growth factor and its receptors. Exp Biol Med (Maywood). 2011;236:1100-7. 48. Nagaya N, Kangawa K, Itoh T, Iwase T, Murakami S, Miyahara Y, et al. Transplantation of mesenchymal stem cells improves cardiac function in a rat model of dilated cardiomyopathy. Circulation. 2005;112:1128-35. 49. Nakajima H, Sakakibara Y, Tambara K, Marui A, Yoshimoto M, Premaratne GU, et al. Delivery route in bone marrow cell transplantation should be optimized according to the etiology of heart disease. Circ J. 2008;72:1528-35. 50. Nomura T, Ashihara E, Tateishi K, Asada S, Ueyama T, Takahashi T, et al. Skeletal myospherederived progenitor cell transplantation promotes neovascularization in delta-sarcoglycan knockdown cardiomyopathy. Biochem Biophys Res Commun. 2007;352:668-74. 51. Ohno N, Fedak PW, Weisel RD, Mickle DA, Fujii T, Li RK. Transplantation of cryopreserved muscle cells in dilated cardiomyopathy: effects on left ventricular geometry and function. J Thorac Cardiovasc Surg. 2003;126:1537-48. 52. Pouly J, Hagege AA, Vilquin JT, Bissery A, Rouche A, Bruneval P, et al. Does the functional efficacy of skeletal myoblast transplantation extend to nonischemic cardiomyopathy? Circulation. 2004;110:1626-31. 53. Shabbir A, Zisa D, Suzuki G, Lee T. Heart failure therapy mediated by the trophic activities of bone marrow mesenchymal stem cells: a noninvasive therapeutic regimen. Am J Physiol Heart Circ Physiol. 2009;296:H1888-97. 54. Singla DK, Ahmed A, Singla R, Yan B. Embryonic Stem Cells Improve Cardiac Function in Doxorubicin-Induced Cardiomyopathy Mediated through Multiple Mechanisms. Cell Transplant. 2012;21:1919-30. 55. Sun CK, Chang LT, Sheu JJ, Chiang CH, Lee FY, Wu CJ, et al. Bone marrow-derived mononuclear cell therapy alleviates left ventricular remodeling and improves heart function in rat-dilated cardiomyopathy. Crit Care Med. 2009;37:1197-205. 56. Suzuki K, Murtuza B, Suzuki N, Smolenski RT, Yacoub MH. Intracoronary infusion of skeletal myoblasts improves cardiac function in doxorubicin-induced heart failure. Circulation. 2001;104:I213-7.
9
191
PART THREE CHAPTER 9
57. Werner L, Deutsch V, Barshack I, Miller H, Keren G, George J. Transfer of endothelial progenitor cells improves myocardial performance in rats with dilated cardiomyopathy induced following experimental myocarditis. J Mol Cell Cardiol. 2005;39:691-7. 58. Zhang J, Li GS, Li GC, Zhou Q, Li WQ, Xu HX. Autologous mesenchymal stem cells transplantation in adriamycin-induced cardiomyopathy. Chin Med J (Engl). 2005;118:73-6. 59. Zhou C, Yang C, Xiao S, Fei H. Feasibility of bone marrow stromal cells autologous transplantation for dilated cardiomyopathy. J Huazhong Univ Sci Technolog Med Sci. 2007;27:75-8
192
CELL THERAPY FOR DILATED CARDIOMYOPATHY
SUPPLEMENTAL MATERIAL Table S1. Search syntax preclinical trials (May 1st, 2012) PubMed (DCM OR IDCM OR cardiomyopathy) AND (dog OR canine OR sheep OR ovine OR pig OR pigs OR porcine OR swine OR bovine OR rabbit OR rabbits OR rodents OR rat OR rats OR mouse OR mice OR hamster OR hamsters) AND (“stem cells” OR “stem cell” OR “progenitor cells” OR “progenitor cell” OR “bone marrow” OR transplantation) 897 hits Embase (DCM OR IDCM OR cardiomyopathy) AND (dog OR canine OR sheep OR ovine OR pig OR pigs OR porcine OR swine OR bovine OR rabbit OR rabbits OR rodents OR rat OR rats OR mouse OR mice OR hamster OR hamsters) AND (‘stem cells’ OR ‘stem cell’ OR ‘progenitor cells’ OR ‘progenitor cell’ OR ‘bone marrow’ OR transplantation) 821 hits Total: 1718 hits After removal of duplicates 1455 hits
Table S2. Search syntax clinical trials (February 16th, 2012) PubMed (dilated cardiomyopathy OR DCM) AND (stem cells OR stem cell OR progenitor OR bone marrow) AND (“2006/01/01”[PDat] : “2012/02/16”[PDat]) 158 hits Embase (‘dilated cardiomyopathy’ OR ‘DCM’) AND (‘stem cells’ OR ‘stem cell’ OR ‘progenitor’ OR ‘bone marrow’) AND [2006-2012]/py 291 hits Total: 449 hits After removal of duplicates 339 hits
Table S3. Search ongoing clinical trials (November 16th, 2012) ClinicalTrials.gov Search Terms: ((cardiomyopathy AND (dilated OR idiopathic)) OR DCM) AND (stem OR cells OR cell OR progenitor OR bone marrow) | Adult, Senior
9
56 hits 49 excluded: - 1 terminated (NCT00612911) - 2 pediatric (NCT01219452, NCT01504594) - 4 ischemic cardiomyopathy (NCT01720888, NCT00545610, NCT00690209, NCT01670981) - 5 completed and published (NCT00284713, NCT00615394, NCT00743639, NCT00629096, NCT00629018) - 1 observational study (NCT00962364) - 1 no cell therapy (NCT00585546) - 1 no DCM (NCT00349271) - 1 cardiac resynchronization therapy (NCT00800657) - 2 gene therapy (NCT01643330, NCT00454818) - 31 others (NCT00958087, NCT00815217, NCT01080781, NCT00943059, NCT00193843, NCT00011648, NCT00785525, NCT01152918, NCT00006301, NCT00368225, NCT00353639, NCT00009568, NCT00575211, NCT00078390, NCT00193765, NCT01480479, NCT00923884, NCT00340717, NCT01675102, NCT00342706, NCT00341497, NCT00483379, NCT00867048, NCT00745953, NCT00923962, NCT00342433, NCT01165515, NCT01661283, NCT01247922, NCT01032070, NCT00433862) 7 ongoing clinical trials
193
PART THREE CHAPTER 9
Table S4. Critical appraisal preclinical studies using CEBM criteria Author
Randomized Comparable Comparable Adequate groups treatment follow-up
Blind investigator
Baba
-
-
+
+
-
-
Chen ‘06
-
-
+
+
-
+
Chen ‘10
+/-
-
+
+
-
-
De Angelis
-
-
+
+
-
-
Dhein
+/-
-
+
+
+
+
Garbade
-
-
+
+
-
+
Hata
-
-
-
+
-
-
Ishida
+/-
-
+
+
-
-
Jin
+
-
+
-
-
-
Kondoh ‘06
+/-
-
+
+
-
+
Kondoh ‘07
+/-
+
-
+
-
+
Lin
+/-
+
+
+
-
+
Lu
+/-
-
+
-
-
-
Mu
+/-
-
+
+
-
-
Nagaya
-
-
+
+
-
+
Nakajima
-
-
+
+
-
+
Nomura
-
-
+
-
-
-
Ohno
-
-
-
+
-
+
Pouly
+/-
+
+
+
-
+
Psaltis
+/-
+
+
+
-
-
Shabbir
-
-
+
-
-
+
Singla
-
-
+
-
-
-
Sun
+/-
+
+
+
-
-
Suzuki
-
-
+
+
-
-
Tezuka
-
-
-
-
-
+
Werner
+/-
-
+
+
-
-
Yamada
-
-
-
+
-
-
Zhang
+/-
-
+
-
-
-
Zhou
+/-
-
+
-
-
-
randomized with concealed allocation +
similar groups Y/N
similar treatment groups in controlled trials besides intervention Y/N
<20% lost to follow up Y/N
blind treatment officer Y/N
blind investigator Y/N
randomized, unclear concealed allocation +/nonrandomized Y = yes N = no
194
Blind treatment officer
CELL THERAPY FOR DILATED CARDIOMYOPATHY
Table S5. Critical appraisal clinical studies using CEBM criteria Author
Randomized
Similarity groups
Similar treatment
Adequate follow-up
Blinding
Arguero
-
n/a
n/a
unknown
-
Benetti
-
n/a
n/a
+
-
Bocchi
+/-
unknown
unknown
+
-
Chin
-
n/a
n/a
+
-
Fatkhutdinov
-
+/-
+
+
-
Fischer-Rasokat
-
n/a
n/a
+
-
Kalil
-
n/a
n/a
unknown
-
Kirillov
-
n/a
n/a
+
-
Martino
-
n/a
n/a
+
-
Ruengsakulrach
-
n/a
n/a
+
-
Schannwell
-
+/-
+
+
-
Seth
+/-
+
unknown
+
-
Suarez de Lezo
-
n/a
n/a
+
-
Vrtovec (2011)
+/-
+
unknown
+
-
Vrtovec (2013)
+/-
+
unknown
+
-
randomized with concealed allocation +
similar groups Y/N
similar treatment groups in controlled trials besides intervention Y/N
<20% lost to follow up Y/N
blinding of patient and physician +
randomized, unclear concealed allocation +/-
matched controls +/-
nonrandomized -
blinding of patient or physician +/-
9
no blinding -
Y = yes N = no
195
196 SD
Mean difference
p-value
26.6
21.8
24.5
20.5
30.2
25.9
23.4
24.4
23.8
23.3
17
22.5
27
25.5
24.3
Bocchi
Chin
Fatkhudinov
Fischer-Rasokat
Kalil
Kirillov group 1
Kirillov group 2
Martino
Ruengsakulrach
Schannwell
Seth
Suarez de Lezo
Vrtovec (2011)
Vrtovec (2013) 6.5
7.5
7
8.3
2
7
5.3
10.3
6.7
8.2
10.9
1.6
5.3
3.8
4
4.6
30.0
30.1
38
28.4
26
27.7
25
26
25.4
25.4
33.4
≈ 21
39.6
30.6
34.8
36
5.1
6.7
11
11.8
3
11.3
7.7
7.6
11.1
6.8
11.5
?
7.7
7.3
7.2
5.3
+ 5.7
+ 4.6
+ 11
+ 5.9
+9
+ 4.4
+ 1.2
+ 1.6
+2
- 0.5
+ 3.2
≈0
+ 15.1
+ 8.8
+ 8.2
+ 8.7
0.02
0.03
0.001
<0.05
<0.01
0.03
N.S.
N.S.
N.S.
0.023
<0.001
N.S.
N.S.
0.016
0.005
0.0059
25.7
26.7
20.8
17
26.1
21.3
4.1
3.9
9.3
3
6.0
5.2
SD
≈ 23
≈ 25
21.2
17
≈ 21
26.9
EF final
?
?
9.2
2
?
8.3
SD
≈ - 3b
≈ - 2b
+ 0.4
0
≈-5
+ 5.6b
Mean difference
unknownc
unknownc
N.S.c
N.S.
N.S.
0.04
p-value
a
outcomes calculated b controls with G-CSF administration c significant difference between therapy and controls ≈ italic = estimated effect based on graphics. Abbreviations: EF = ejection fraction SD = standard deviation N.S. = non-significant.
27.3
Benetti
EF baseline
EF final
EF baseline
SD
Controls
Therapy
Argueroa
Author
Table S6. Outcomes of ejection fraction clinical studies
PART THREE CHAPTER 9
197
PART THREE
|
Elucidating (Epi)genetic and Translating
Therapeutic Pathways
Chapter
10
Xenotransplantation of Human Cardiomyocyte Progenitor Cells Does Not Improve Cardiac Function in a Porcine Model of Chronic Ischemic Heart Failure Results from a randomized, blinded, placebo controlled trial
Submitted
Sanne J. Jansen of Lorkeers1, Johannes M.I.H. Gho1, Stefan Koudstaal1,2, Gerardus P.J. van Hout1, Peter-Paul M. Zwetsloot1, Joep W.M. van Oorschot3, Esther C.M. van Eeuwijk1, Tim Leiner3, Imo E. Hoefer1, Marie-JosĂŠ Goumans4, Pieter A. Doevendans1,2, Joost P.G. Sluijter1,2, Steven A.J. Chamuleau1,2
1
Department of Cardiology, Division Heart and Lungs, University Medical Center Utrecht, Utrecht, the Netherlands
2
ICIN - Netherlands Heart Institute, Utrecht, the Netherlands
3
Department of Radiology, University Medical Center Utrecht, the Netherlands
4
Department of Molecular Cell Biology, Leiden University Medical Center, the Netherlands
PART THREE CHAPTER 10
ABSTRACT Background Recently cardiomyocyte progenitor cells (CMPCs) were successfully isolated from fetal and adult human hearts. Direct intramyocardial injection of human CMPCs (hCMPCs) in experimental mouse models of acute myocardial infarction significantly improved cardiac function compared to controls. Aim Here, our aim was to investigate whether xenotransplantation via intracoronary infusion of fetal hCMPCs in a pig model of chronic myocardial infarction is safe and efficacious, in view of translation purposes. Methods & Results We performed a randomized, blinded, placebo controlled trial. Four weeks after ischemia/ reperfusion injury by 90 minutes of percutaneous left anterior descending artery occlusion, pigs (n=16, 68.5 Âą 5.4 kg) received intracoronary infusion of 10 million fetal hCMPCs or placebo. All animals were immunosuppressed by cyclosporin (CsA). Four weeks after infusion, endpoint analysis by MRI displayed no difference in left ventricular ejection fraction, left ventricular end diastolic and left ventricular end systolic volumes between both groups. Serial pressure volume (PV-)loop and echocardiography showed no differences in functional parameters between groups at any timepoint. Infarct size at follow-up, measured by late gadolinium enhancement MRI showed no difference between groups. Intracoronary pressure and flow measurements showed no signs of coronary obstruction 30 minutes after cell infusion. No premature death occurred in cell treated animals. Conclusion Xenotransplantation via intracoronary infusion of hCMPCs is feasible and safe, but not associated with improved left ventricular performance and infarct size compared to placebo in a porcine model of chronic myocardial infarction.
200
CMPCs IN HEART FAILURE
INTRODUCTION The heart has regenerative capacity as it harbours a pool of cardiac stem cells.1 However, this is clearly not sufficient to repair the damage caused by myocardial infarction (MI) to prevent the development of heart failure. The number of stem cells available might just be too little. Ex vivo expansion and reapplication of cardiac stem cells to the injured heart was proposed, however isolation of these cardiac stem cells remains challenging. Our lab succeeded in isolating fetal and adult cardiomyocyte progenitor cells (CMPC) from mouse and human hearts based on the stem cell antigen Sca-1.2 The Sca-1-like positive human CMPCs (hCMPCs) from fetal and adult hearts showed differentiation into spontaneously beating cardiomyocytes after stimulation with 5-azacytidine and TGFβ.3 These hCMPC-derived cardiomyocytes have functional gap junctions, enabling metabolic and electrical coupling of cells.3 Additionally, hCMPCs can also be differentiated into endothelial cells and smooth muscle cells by exposure to vascular endothelial growth factor (VEGF).2–4 Intra-myocardial injection of hCMPCs in a mouse model of acute MI led to engraftment of 3.5% of cells, differentiation towards coupled cardiomyocytes, increased vascular density, and to improved cardiac function.5 These promising results led to the current large animal study, as a next step towards potential clinical application of hCMPCs. Since the greatest burden of ischemic heart disease is based on chronic ischemic heart failure, hCMPCs were tested in a chronic disease model. In the current study, fetal hCMPCs were intracoronary administered in a porcine model of chronic ischemia/reperfusion injury (I/R). We hypothesized that this strategy is safe, improves left ventricular performance and reduces infarct size compared to placebo.
METHODS Experimental design All animal experiments were executed in conformance with the ‘Guide for the Care and Use of Laboratory Animals’. The experiment was evaluated and approved by the Animal Experiments Committee of the Utrecht University, the Netherlands (permit number 2012.II.09.145). Animals were subjected to I/R, randomized to fetal hCMPC or placebo infusion 4 weeks later and endpoint
10
analyses were performed 4 weeks after stem cell injection (see study protocol, Figure 1). All animals were immunosuppressed by Cyclosporin A (CsA) to facilitate xenogeneic cell treatment. Based on a power calculation (estimated effect 7.5%6, standard deviation of 5%, a power of 0.9 and alpha of 0.05) 8 pigs per group were needed. Animals that died before endpoint analysis were supplemented. Animals were randomized after surviving the initial I/R, using a computer based random order generator. Cell and placebo infusion as well as data analysis were performed in a blinded fashion (investigators, technicians and animal caretakers). Deblinding was performed after collecting and analysing all data. The primary endpoint of this study was defined as left ventricular ejection fraction (EF) at the end of follow-up, measured by magnetic resonance imaging (MRI). Secondary endpoints were left ventricular end diastolic volume and left ventricular end systolic volume (EDV and ESV)
201
PART THREE CHAPTER 10
measured by MRI, infarct size measured by ex vivo gross macroscopy after incubation with triphenyltetrazolium chloride (TTC) and late gadolinium enhancement (LGE) MRI, functional parameters serially measured by pressure volume (PV-)loop and echocardiography, coronary microvascular function by intracoronary pressure- and flow measurements and vascular density and fibrosis on histology. Cell isolation Cells were isolated from fetal human heart tissue (derived after elective abortion with informed consent) and cultured as described before.2 Shortly, tissue was minced, incubated with collagenase and grinded through a cell strainer. Cells were incubated with anti-Sca-1 microbeads and separated using a MiniMACS magnet (Miltenyi Biotec, Leiden, the Netherlands). hCMPCs were collected and dissolved in growth medium containing endothelial growth medium (EGM2, Cambrex, CC-4176), FBS, penicillin/streptomycin, non-essential amino acids and bFGF. After attachment of the cells, cells were split after +/- 3 days at 80-90% confluence in a 1:6 fashion. All pigs received hCMPCs of passage 5-7 from the same donor. Animal experiment A comprehensive description of the protocol is also available at http://www.jove.com/ video/51269.7 Female landrace specific pathogen free pigs (n=19) (van Beek SPF Varkensfokkerij B.V., Lelystad, the Netherlands), weighing 68.5 ± 5.4 kg (Upper and lower limit 60.6 – 82.0 kg) at baseline, were pre-treated with amiodaron for 10 days (1200 mg/day for 10 days, 800 mg/ day maintenance), clopidogrel for 3 days (75 mg/day) and acetylsalicylic acid for 1 day (320 mg loading dose, 80 mg/day maintenance) and a fentanyl patch (25µg/h) for 1 day. One day before cell or placebo delivery, cyclosporin (CsA) was started. Based on clinical organ transplantation protocol, pigs received a loading dose of 800 mg, then 400 mg b.i.d. for 1 week and 200 mg b.i.d for the remaining 3 weeks by oral administration (Neoral drink, 100mg/ml, Novartis Pharma bv). At the day of surgery, one dosage was i.v. infused as 200 mg in 100mL over 2 hours (Sandimmune 50mg/ml, Novartis Pharma bv). All medication, except for the fentanyl patch, was continued until the end of follow-up. Pigs received fiber rich pellets (Abdiets animal nutrition, product 2755, Woerden, The Netherlands) twice a day and water was available ad libitum. Animals were kept fasted the day of surgery (except for medication). Anaesthesia was obtained by intramuscular injection of 10 mg/kg ketamine, 0.4 mg/kg midazolam and 0.5 mg/kg atropine in the cage. Pigs were intubated and transported to the operating theatre. Maintenance anaesthesia consisted of continuous infusion of 0.5 mg/kg/h midazolam, 2.5 µg/ kg/h sufentanyl and 0.1 mg/kg/h pancuronium. Other perioperative medication consisted of 300 mg amiodaron, amoxicillin + clavulanic acid 750/75 mg and heparin (100 IE/kg after positioning the sheaths and 50 IE/kg every 2 hours). Pigs were mechanically ventilated with a positive pressure ventilator with FiO2 0.5, 10ml/kg tidal volume and a frequency of 12/minute under continuous capnography. Arterial access was achieved by cannulating the internal carotid artery with an 8F sheath. Venous access was achieved by cannulating the jugular vein with a 9F sheath. An additional arterial line was inserted in a small peripheral artery in the hind limb for continuous stable arterial pressure registration.
202
CMPCs IN HEART FAILURE
Figure 1. Study protocol
Echocardiography Pigs were positioned in the right lateral position. Parasternal short axis images were obtained at 3 levels (mitral valve, papillary muscle and apex) during 5 beats per level (iE33 ultrasound device Philips, Eindhoven, The Netherlands). Short axis images were analysed offline using Xcelera R2.L1 (Philips Healthcare, Best, The Netherlands) and fractional area shortening (FAS), fractional shortening (FS) and septal wall thickening (WTsept) at the level of the mitral valve, papillary muscle and apex (mitral, pap, apex respectively) were calculated. Because of apical dilatation of the left ventricle after infarct, the modified Simpsons rule was not applicable and no reliable volumes could be calculated. Pressure volume loop measurement Admittance based PV-loop measurements were performed as recently described.8 Cardiac output (CO) was measured three times by thermodilution and stroke volume was calculated out of three measurements. The 7F tetra-polar admittance catheter (7.0 VSL Pigtail/no lumen, Transonic Scisense, London, Canada) was inserted into the left ventricle through the arterial sheath under fluoroscopic guidance. Inferior caval vein occlusion was performed using an 8F Fogarty occlusion catheter (62080814F, Edwards Lifesciences). All measurements were performed during apnea. Data were offline analysed using Iworx analysis software (Labscribe V2.0).
10
Intracoronary pressure and flow velocity assessment Intracoronary pressure and flow were assessed by positioning a Combowire in the left anterior descending artery (LAD) and the left circumflex coronary artery (LCX) acting as a control subsequently. Intracoronary pressure and flow measurements, together with arterial pressure and ECG, were recorded using the ComboMap system (Volcano Corporation). An intracoronary bolus of 200 Âľg nitroglycerin was administered to prevent coronary spasms. Three successive measurements were recorded in rest and during hyperaemia, achieved by intracoronary administration of 60 Âľg adenosine. Data were analysed offline, using AMC Volcano Studymanager (versus 6.0, Borland Software Corporation and Delphi versus 2010, Embarcadero, San Francisco, CA, USA).
203
PART THREE CHAPTER 10
Ischemia/reperfusion injury After baseline measurements, the diameter of the LAD, distal to the second diagonal branch (D2) was measured in anterior-posterior and left anterior oblique 30° view. Before positioning the balloon, an intracardiac defibrillation catheter was placed in the right ventricle. A PCI-balloon catheter with a suitable diameter was positioned distal from the D2 and inflated for 90 minutes. In case of ventricular arrhythmias, chest compressions were started immediately, amiodaron 300mg was infused intravenously and pigs were defibrillated intracardiac with 50J. In case of multiple unsuccessful shocks, defibrillation was switched to transthoracic defibrillation with 150-200J. Cell infusion Ten million cells were resuspended in 10mL of phosphate buffered saline (PBS), one hour before intracoronary infusion. Cell suspension or placebo (10mL of PBS) was handed to the animal technician in a covered sterile syringe so that the personnel at the operating theatre remained blinded for the treatment allocation. An over-the-wire PCI balloon catheter (Apex, Boston Scientific), was positioned distal to the D2, according to the position of the PCI catheter during infarct creation. The balloon was inflated until pressure matched the coronary diameter. 10mL of cell suspension or placebo was infused in three sessions consisting of 30 seconds infusion of 3.3mL during 2 minutes balloon occlusion followed by 3 minutes of reperfusion. Intracoronary pressure and flow measurements were repeated, 30 minutes after the last reperfusion period, to detect possible flow restrictions caused by cell or placebo infusion. MRI At the end of follow-up, cardiac MRI and all other functional measurements were performed. All MR studies were performed on a 3T MRI scanner (Achieva TX, software release 3.2; Philips Healthcare, Best, The Netherlands). Both cine images and LGE images in short axis and twochamber long axis orientations were obtained under continuous anesthesia and mechanical ventilation. LGE images were obtained at least 15 minutes after injection of 0.2 ml/kg gadobutrol (Gadovist, Bayer Healthcare). Offline imaging analysis for functional measurements was performed in Qmass MR 7.4 enterprise solutions (Medis medical imaging systems BV, Leiden) and by Segment version v1.9R 3293 (Medviso AB, Lund, Sweden) for infarct size. Infarct size at the end of follow-up was analysed based on the 2SD method. Ex vivo Infarct size Animals were sacrificed by exsanguination under general anesthesia. The hearts were excised, washed and cooled with running tap water and the LV was then cut into 5 equal slices from apex to base. Slices were incubated in 1% TTC (Sigma-Aldrich Chemicals, Zwijndrecht, the Netherlands) in 37°C 0.9%NaCl for 15 min to discriminate infarcted tissue from viable myocardium. The infarcted area was calculated as percentage of the left ventricle.
204
CMPCs IN HEART FAILURE
Histology Sections of infarct zone, border zone and remote area of all pigs were isolated and fixated in 4% formalin for at least 1 week and then embedded in paraffin. Sections of 5 µm were stained for endothelial cells by lectin (Lectin fom Bandeiraea simplicifolia, Sigma Aldrich L5391) and for fibrosis by picrosirius red (Sirius red F3B, BDH and picrin acid, Boom. Cat. 12388). For vascular density, 5 random pictures per slide with 40x magnification were taken from the border zone area. The number of vessels was counted manually by two independent observers (S.J. and J.G.) and the absolute number of lectin positive vessels per field were averaged. For assessment of fibrosis, 3 random pictures per slide (slides for infarct zone, border zone and remote area) were taken using polarized light during one session. Pictures were analysed in Cell^p (version 5.0 Olympus) for mean grey value and percentage of fibrotic area by using the same settings for all slides. Cyclosporin assay The effect of CsA on the cells was tested by a migration assay, sprouting matrigel assay and by testing cellular growth factor release. Based on the pig serum levels of CsA, in vitro assays were executed to analyze performance of CMPCs in presence of CsA. CMPCs were maintained in 0.1% gelatin coated plates with standard CMPC culture medium. For all in vitro experiments low passages were used (passage 8-15). For all assays, standard culture medium was used as a negative control. Three different concentrations of CsA were used as experimental conditions (50 ng/mL, 150 ng/mL and 300 ng/mL). See supporting information for detailed description of the methods. Statistical analysis Data are reported as mean ± standard deviation. Comparisons of serial measurements over time within one group or for all animals combined were performed by paired t-tests. Comparisons between both groups were made by independent t-tests. Differences between conditions in cyclosporin assays were tested by one-way ANOVA. The statistical software used is IBM SPSS statistics, version 20.0 (IBM Corporation, Armonk, NY, USA).
RESULTS
10
A total of 19 pigs were used in this study. In 9 pigs VF occurred during occlusion, of which 8 converted successfully by defibrillation and amiodaron infusion. A total of three pigs died; one pig died of VF during LAD occlusion, one pig died by periprocedural complications before cell/ placebo infusion and one animal died one day after placebo infusion by unknown cause. All other 16 animals were included in this study. Safety Two hours after cell infusion during short coronary occlusion, high sensitive troponin I (TnI) levels were 545 ± 460 ng/L n=8 after cell infusion and 623 ± 764 ng/L n=8 after placebo infusion.
205
PART THREE CHAPTER 10
In comparison, mean TnI levels 3.5 hours after the start of 90 minutes balloon occlusion, were 1147822 ± 584559 ng/L (n=16), confirming the induction of a large MI. Intracoronary pressure and flow measurement (see below) showed no evidence for intracoronary embolization or obstruction in CMPC treated animals 30 minutes after infusion. None of the CMPC treated animals died by arrhythmias, or any other cause. Left ventricular performance MRI At the end of follow-up EF, EDV and ESV did not differ between groups. EF in CMPC treated animals was 40.6 ± 4.2 % versus 38.3 ± 8.6% in placebo animals (p=0.52). EDV was 172.7 ± 19.0 ml in CMPC treated animals and 150.0 ± 27.1 ml in placebo (p=0.07). ESV was 102.6 ± 13.3 ml in CMPC treated and 91.5 ± 14.9 ml in placebo animals (p=0.14). (Figure 2) PV-loop Serial functional measurements were performed by PV-loop. Because of technical failure, we could not complete data analysis of all animals at all time points. For EF, ESV and EDV we could include 7 out of 8 cell treated animals and all 8 placebo animals. For end systolic pressures volume relationship (ESPVR), end diastolic pressure volume relationship (EDPVR) and V0, we could include 6 animals per group. Induction of MI caused a significant decrease in EF over 4 weeks (-8.7 ± 11.6% from 54.8 ± 8.7 to 46.1 ± 9.5 p=0.01) and an increase in ESV (+40.6 ± 62.3 ml from 85.3 ± 27.1 to 125.9 ± 49.1
Figure 2. Functional measurements by MRI of cell- and placebo treated animals at follow-up. Individual data and mean with SD, n=8 per group. EF Left ventricular ejection fraction, EDV End diastolic volume, ESV End systolic volume, NS = non-significant.
206
CMPCs IN HEART FAILURE
Figure 3. Functional outcome measurements by PV-loop. Mean and SD for PV-loop parameters. EDV End diastolic volume, ESV End systolic volume, EF Left ventricular ejection fraction, ESPVR End systolic pressures volume relationship, EDPVR end diastolic pressure volume relationship. Placebo n=8, cell n=7 for ESV, EDV and EF, n=6 per group for ESPVR, EDPVR and V0.
p=0.02), and trend towards an increased EDV (+38.1 ± 73.6 ml, from 186.7 ± 37.6 ml to 224.7 ± 50.9 p=0.07), confirming successful creation of a chronic MI model. No significant differences between both groups were observed at pre-infusion time point (EF p=0.34, EDV p=0.40, ESV p=0.27). (Figure 3) For all parameters (EDV, ESV, EF, ESPVR, EDPVR and V0), no differences between both groups were found at any time point except for EDPVR at follow-up. EDPVR was 0.004 ± 0.002 for cell treated animals and 0.008 ± 0.004 for placebo animals (p=0.04). Visual appearing trends in functional outcomes are observed in Figure 3 (and Supporting table 2). However, EF did not increase after cell or placebo infusion (p=0.85 and p=0.17 respectively) EDV and ESV did not change after cell treatment (p=0.25 and p=0.58 respectively), but both significantly increased after placebo infusion (p=0.01 and p=0.004), suggesting further deterioration after placebo infusion compared to cell treatment. No statistical significant differences in ΔEDV and ΔESV were found between cell treated animals and placebo animals.
10
Echocardiography WTsept and FAS at the apical level significantly decreased after MI (WTsept -0.34 ± 0.23 p < 0.001, FAS -0.08 ± 0.11 p = 0.02), confirming successful apical infarct creation. A trend towards decrease of FS at the apical level was also observed (-0.50 ± 0.11 p=0.08) in both groups. For FAS, FS and WTsept, no statistical significant differences were identified between groups at any time point. A trend was observed for a higher mean FAS (mean of three levels) at follow -up in cell treated animals compared to placebo animals (cell treated 0.49 ± 0.06 placebo 0.44 ± 0.04 p=0.08). Next, no differences in decrease or increase of any other echocardiographic parameter after hCMPC or placebo infusion was observed. (Figure 4 and Supporting table 3)
207
PART THREE CHAPTER 10
Figure 4. Functional measurements by echocardiography. Mean and SD, n=8 per group. FAS Fractional area shortening, FS Fractional shortening, sWT Septal wall thickening at the level of the apex.
Intracoronary pressure and flow Hyperemic microvascular resistance is a parameter for assessing the microvascular bed and is calculated as the ratio of intracoronary pressure (Pd) and hyperemic average peak flow velocity (pAPV). It is known from previous studies that HMR increases after I/R, accompanied by decreased pAPV and a constant Pd.9 At the time point of cell infusion, measurements in one animal failed because of technical problems. HMR in the LAD region significantly increased after I/R as expected (p=0.03), Pd in the LAD significantly increased (p=0.03) and pAPV in the LAD tended towards a decrease (p=0.10). All parameters in the control region remained stable over time. Thirty minutes after cell or placebo infusion, HMR was significantly decreased in all animals (2.7 ± 1.0 mmHg/cm/s pre-infusion, 2.2 ± 0.6 mmHg/cm/s post-infusion, p = <0.001), without any differences between hCMPC and placebo treated animals (ΔHMR -0.50 ± 0.48 mmHg/cm/s for hCMPC treated, -0.56 ± 0.66 mmHg/cm/s for placebo treated animals, p=0.21). Decrease in HMR was accompanied by an increase in pAPV (41.5 ± 11.0 cm/s pre-infusion, 48.9 ± 11.7 post-infusion, p=0.01), without differences between groups (p=0.7), both advocating against intracoronary (micro-)embolization or obstruction by cell infusion. Decreased HMR and increased pAPV can be explained as a hyperemic response after short ischemia during infusion. Neither cell-, nor placebo infusion caused a change in microvascular parameters over time. Infarct size For LGE, infarct size was 17.5 ± 3.8% and 18.0 ± 4.5% of the left ventricular mass for cell treated and placebo animals respectively at follow-up. For TTC, infarct size in the cell treated group was 14.6 ± 2.5% of the left ventricle, and 13.8 ± 3.9 in the control group. There was no statistical difference in infarct size between both groups at termination of follow-up, for both modalities. (Figure 5) Histology For fibrosis, there were no differences between both groups. A significant higher vascular density was observed in the placebo group compared to the hCMPC animals. (Table 1, Figure 6)
208
CMPCs IN HEART FAILURE
Cyclosporin CsA did not affect functioning of CMPCs in vitro with regard to migration capacity, angiogenesis and growth factor secretion. See supporting information for results and figures.
Figure 5. Infarct size Representative pictures of TTC stained heart slice and LGE image of the same animal.
Table 1. Fibrosis % Fibrosis
Grey value
CMPC (n=8)
Placebo (n=8)
Infarct
39.1 ± 12.2
40.0 ± 16.0
Border
1.6 ± 1.2
2.9 ± 2.3
Remote
6.5 ± 7.8
2.8 ± 2.3
Infarct
27.0 ± 8.9
28.5 ± 12.7
Border
1.2 ± 0.7
2.0 ± 1.4
Remote
4.3 ± 4.3
2.6 ± 2.2
33.9 ± 10.1
48.3 ± 10.1 *
Vascular density * p=0.01 between CMPC and placebo.
10
209
PART THREE CHAPTER 10
Figure 6. Histology Representative pictures of picrosirius red staining of infarct area (A), borderzone (B) and remote area (C) and lectin staining of the borderzone (D). Table presents mean Âą SD percentage fibrosis and mean grey value of infarct area, border zone and remote area and vascular density of the borderzone of cell and placebo treated animals.
DISCUSSION In this preclinical model of chronic MI, we show that xenotransplantation via intracoronary infusion of fetal hCMPCs is feasible and safe, but we did not find a benefit in the current model on cardiac function, infarct size, fibrosis and vascular density. The non-significant difference in EF between both groups of 2.3% measured by the gold standard MRI is lower than the effect size of 7.5% as documented in a recent meta-analysis and used for power calculation. We believe that these data in the field of cardiac regeneration are of great importance. It complements the preclinical evidence needed for successful translation of cardiac regeneration by cardiac cell therapy from bench to bedside. We present a solid and sound piece of evidence since this study is conducted by clinical standards in terms of blinding, randomization and outcome assessment. Preclinical studies of cardiac cell therapy The neutral results presented in the current study, are partly in line with other large animal studies with cardiac stem cells for MI (Table 2). Over the last decade, five different cardiac stem
210
CMPCs IN HEART FAILURE
Source
Dose
Timing
Follow- up
IC
Autologous NA
1,0E+07
4 wk
8 wk
No EF or IS improvement
29
IM
Autologous NA
1,0E+07
4 wk
8 wk
No functional improvement on echo. EF improved on LV angiography
10
IC
Allogeneic
No
1,25E+07 2-3 wk
8 wk
EF preservation, IS reduction
14
IC
Allogeneic
No
8,5-9E+06 30 mins 2 days
No difference in EF, IS reduction
22
TE
Allogeneic
No
1,5E+08
8 wk
8 wk
No functional improvement, IS reduction
10
IM
Human
Cyclosporin
1,0E+06
14 days
1 mnth Restored EF, IS reduction
21
IC
Autologous
NA
5,0E+05
3-4 mnth
1 mnth
Result
Route
17
Immuno
N
suppression
Table 2. Overview of studies of cardiac derived stem cells in pig models
CDC Johnston et al. 31 Circulation 2009
Lee et al. 32 JACC 2011
Malliaras et al. 33 Circulation 2013
Kanazawa et al.34 Circ. Heart Fail. 2015
Yee et al. *35 PLoS ONE 2014
CSC Williams et al. 22 Circulation 2013
Bolli et al. 23 Circulation 2013
Improved EF, IS reduction
IM intramyocardial injections, IC intracoronary infusion, TE transendocardial injections, EF ejection fraction, IS infarct size. * used cardiospheres instead of CDCs.
cell sources have been identified10 and two of them have been extensively tested in large animal studies. Clonogenic, multipotent and self-renewing cardiac stem cells, based on the C-kit epitope (CSCs)11,12 and cluster-forming cells, positive for C-kit and Sca-1, called cardiosphere derived cells (CDCs)13 were used in these studies. Both cell types have gone through the process of preclinical investigation and progressed from bench to bedside testing. In a recent study, expanded cells upon different cardiac progenitor cell isolation procedures were compared
10
(including, c-kit+, cardiospheres, and sca1+ cells), and it was suggested that these cells have highly similar transcriptional profiles.14 Therefore, it is important to closely look at particular study design aspects of the study to look for possible explanations. Study design considerations A multitude of parameters in the research field of cell therapy for cardiac repair must be considered when conducting preclinical or clinical studies, like timing of therapy, administration route, cell source, cell number, etc. In a meta-analysis including all large animal studies on cell therapy for MI, none of these parameters showed significant impact on effect size.15 Additionally, no ruling about these issues was provided based on previous CSC studies (Table 2). Therefore, we made upfront decisions about all these variables based on experience and clinical relevance.
211
PART THREE CHAPTER 10
Treatment of acute myocardial infarction is greatly improved over the last decades, but patients surviving the initial episode often suffer from chronic ischemic heart failure. The current animal model was designed to represent this group. A direct comparison of efficacy of stem cells in acute versus chronic ischemic heart disease has never been performed, but safety is at least similar in acute and chronic patients.16 Table 2 shows negative and positive results of cardiac cell therapy in both acute and chronic settings. We administered cells via intracoronary infusion, which is considered as efficient as other routes with regard to cell retention17–19 and functional improvement20,21 but is less invasive then intramyocardial injections and less time consuming compared to transendocardial delivery. We investigated efficacy of infusion of 10 million cells. Table 2 show both positive and neutral results from CDCs in comparable doses. For CSCs, promising results are found by infusion of 0.5 and 1 million cells in smaller pigs (1.5 ± 0.8kg and 35-40 kg respectively, compared to 68.5 ± 5.4 kg in the current study).22,23 Roughly, we applied 4 times higher indexed doses (37,000/kg versus 140,000/kg) of cells. These data show that relative under-dosing was certainly not anticipated for in the present study. Xenogenicity The impact of xenogenicity is important since human cells were administered in a porcine model. Alloreactivity (and xenoreactivity) depends on foreign peptide presentation by the major histocompatibility complex (MHC) on antigen presenting cells and detection by T cells.24 In clinical practice, alloreactivity is suppressed by T cell suppressors like CsA, and this strategy was applied accordingly in the present study. Next to immunosuppression, the effect of CsA includes prevention of apoptosis25 and protection of the myocardium in the setting of acute myocardial infarction.26 Since CsA was administered at 4 weeks after MI until the end of follow-up, we believe it did not affect the outcome by that cause. Adequate dosing by oral administration of CsA is safe in pigs27, but not much is known about the pharmacodynamics. We did not observe differences in serum levels of leukocytes, nor on renal function between hCMPC and placebo treated animals 30 minutes and 4 weeks after treatment, all receiving CsA. Also, we did not find a correlation between serum CsA levels at follow-up and effect size. There is no consensus about the effect of CsA on stem cells in vitro and in vivo.27 In this respect, we tested the effect of different levels of CsA on migration, angiogenesis and growth factor expression of hCMPCs in vitro (See supporting information). No significant effect of CsA was noted. Taken together, we assume that our results are not confounded by the chosen CsA regimen. Internal and external validity Internal validity has major impact on effect size, as it is known that blinded and randomized studies produce smaller effect sizes compared to non-blinded and non-randomized reports.28 Based on clinical standards, our preclinical study has been performed in a randomized, blinded (both for intervention and outcome assessment) and placebo-controlled manner. The animal model is a surrogate for human disease. We included large pigs (68.5 ± 5.4 kg), as the heart corresponds with human (coronary) anatomy and size, to increase external validity (the
212
CMPCs IN HEART FAILURE
translatability towards other models or populations). However, age and co-morbidity are not accounted for in this model. Our standardized way of infarct creation caused a decrease in EF of 8.7% based on PV-loop measurements, despite high TnI levels and a mean infarct size of 17.8%. It is known from clinical studies that patients with a low baseline EF benefit most from cell therapy.29,30 In retrospect, the anticipated effect size of 7.5% (difference in EF at follow-up between cell and placebo treated animals on MRI) might have been too high. Power calculation was based on difference in EF at follow-up, measured by MRI and not for the other functional measurement modalities (echocardiography and PV-loop), which might thereby be underpowered to report significant differences.
CONCLUSION In this randomized, blinded placebo controlled preclinical study, xenotransplantation via intracoronary infusion of fetal human CMPCs in a pig model of chronic I/R injury appeared to be feasible and safe, but showed no significant improvement with regard to cardiac function or infarct size. Acknowledgments We kindly thank Corina Metz, Lizanne Bosma, Marlijn Jansen, Joyce Visser, Martijn van Nieuwburg, Grace Croft, Gerard Marchal, Merel Schurink and Evelyn Velema for their excellent technical assistance.
10
213
PART THREE CHAPTER 10
REFERENCES 1. Goumans MJ, Maring JA, Smits AM. A straightforward guide to the basic science behind cardiovascular cell-based therapies. Heart. 2014;100:1153–7. 2. Smits AM, van Vliet P, Metz CH, Korfage T, Sluijter JP, Doevendans PA, et al. Human cardiomyocyte progenitor cells differentiate into functional mature cardiomyocytes: an in vitro model for studying human cardiac physiology and pathophysiology. Nat Protoc. 2009;4:232–43. 3. Goumans MJ, de Boer TP, Smits AM, van Laake LW, van Vliet P, Metz CH, et al. TGF-beta1 induces efficient differentiation of human cardiomyocyte progenitor cells into functional cardiomyocytes in vitro. Stem Cell Res. 2007;1:138–49. 4. Van Vliet P, Smits AM, de Boer TP, Korfage TH, Metz CH, Roccio M, et al. Foetal and adult cardiomyocyte progenitor cells have different developmental potential. J Cell Mol Med. 2010;14:861–70. 5. Smits AM, van Laake LW, den Ouden K, Schreurs C, Szuhai K, van Echteld CJ, et al. Human cardiomyocyte progenitor cell transplantation preserves long-term function of the infarcted mouse myocardium. Cardiovasc Res. 2009;83:527–35. 6. van der Spoel TI, Jansen of Lorkeers SJ, Agostoni P, van Belle E, Gyöngyösi M, Sluijter JPG, et al. Human relevance of pre-clinical studies in stem cell therapy: systematic review and metaanalysis of large animal models of ischaemic heart disease. Cardiovasc Res. 2011;91:649–58. 7. Koudstaal S, Jansen of Lorkeers S, Gho JM, van Hout GP, Jansen MS, Gründeman PF, et al. Myocardial infarction and functional outcome assessment in pigs. J Vis Exp. 2014. 8. Van Hout GP, de Jong R, Vrijenhoek JE, Timmers L, Duckers HJ, Hoefer IE. Admittance based pressure volume loop measurements in a porcine model of chronic myocardial infarction. Exp Physiol. 2013;98:1565–75. 9. Koudstaal S, Jansen Of Lorkeers SJ, van Slochteren FJ, van der Spoel TI, van de Hoef TP, Sluijter JP, et al. Assessment of coronary microvascular resistance in the chronic infarcted pig heart. J Cell Mol Med. 2013;17:1128–35. 10. Koudstaal S, Jansen Of Lorkeers SJ, Gaetani R, Gho JM, van Slochteren FJ, Sluijter JP, et al. Concise review: heart regeneration and the role of cardiac stem cells. Stem Cells Transl Med. 2013;2:434-43. 11. Beltrami AP, Barlucchi L, Torella D, Baker M, Limana F, Chimenti S, et al. Adult cardiac stem cells are multipotent and support myocardial regeneration. Cell. 2003;114:763–76. 12. Bearzi C, Leri A, Lo Monaco F, Rota M, Gonzalez A, Hosoda T, et al. Identification of a coronary vascular progenitor cell in the human heart. Proc Natl Acad Sci U S A. 2009;106:15885–90. 13. Messina E, De Angelis L, Frati G, Morrone S, Chimenti S, Fiordaliso F, et al. Isolation and expansion of adult cardiac stem cells from human and murine heart. Circ Res. 2004;95:911–21. 14. Gaetani R, Feyen DA, Doevendans PA, Gremmels H, Forte E, Fledderus JO, et al. Different types of cultured human adult cardiac progenitor cells have a high degree of transcriptome similarity. J Cell Mol Med. 2014;18:2147–51. 15. Jansen Of Lorkeers SJ, Eding JE, Vesterinen HM, van der Spoel TI, Sena ES, Duckers HJ, et al. Similar effect of autologous and allogeneic cell therapy for ischemic heart disease: Systematic review and meta-analysis of large animal studies. Circ Res. 2015;116:80–6. 16. De Rosa S, Seeger FH, Honold J, Fischer-Rasokat U, Lehmann R, Fichtlscherer S, et al. Procedural safety and predictors of acute outcome of intracoronary administration of progenitor cells in 775 consecutive procedures performed for acute myocardial infarction or chronic heart failure. Circ Cardiovasc Interv. 2013;6:44–51. 17. Freyman T, Polin G, Osman H, Crary J, Lu M, Cheng L, et al. A quantitative, randomized study evaluating three methods of mesenchymal stem cell delivery following myocardial infarction. Eur Heart J. 2006;27:1114–22. 18. Perin EC, Silva G V, Assad JA, Vela D, Buja LM, Sousa AL, et al. Comparison of intracoronary and transendocardial delivery of allogeneic mesenchymal cells in a canine model of acute myocardial infarction. J Mol Cell Cardiol. 2008;44:486–95. 19. van der Spoel TI, Vrijsen KR, Koudstaal S, Sluijter JPG, Nijsen JF, de Jong HW, et al. Transendocardial cell injection is not superior to intracoronary infusion in a porcine model of ischaemic cardiomyopathy: a study on delivery efficiency. J Cell Mol Med. 2012;16:2768–76.
214
CMPCs IN HEART FAILURE
20. Fukushima S, Coppen SR, Lee J, Yamahara K, Felkin LE, Terracciano CM, et al. Choice of celldelivery route for skeletal myoblast transplantation for treating post-infarction chronic heart failure in rat. PLoS One. 2008;3:e3071. 21. Ang KL, Chin D, Leyva F, Foley P, Kubal C, Chalil S, et al. Randomized, controlled trial of intramuscular or intracoronary injection of autologous bone marrow cells into scarred myocardium during CABG versus CABG alone. Nat Clin Pract Cardiovasc Med. 2008;5:663–70. 22. Williams AR, Hatzistergos KE, Addicott B, McCall F, Carvalho D, Suncion V, et al. Enhanced Effect of Human Cardiac Stem Cells and Bone Marrow Mesenchymal Stem Cells to Reduce Infarct Size and Restore Cardiac Function after Myocardial Infarction. Circulation. 2012;15:213– 23. 23. Bolli R, Tang XL, Sanganalmath SK, Rimoldi O, Mosna F, Abdel-Latif A, et al. Intracoronary delivery of autologous cardiac stem cells improves cardiac function in a porcine model of chronic ischemic cardiomyopathy. Circulation. 2013;128:122–31. 24. Felix NJ, Allen PM. Specificity of T-cell alloreactivity. Nat Rev Immunol. 2007;7:942–53. 25. Matsuda S, Koyasu S. Mechanisms of action of cyclosporine. Immunopharmacology. 2000;47:119–25. 26. Piot C, Croisille P, Staat P, Thibault H, Rioufol G, Mewton N, et al. Effect of cyclosporine on reperfusion injury in acute myocardial infarction. N Engl J Med. 2008;359:473–81. 27. Jansen Of Lorkeers SJ, Hart E, Tang XL, Chamuleau ME, Doevendans PA, Bolli R, et al. Cyclosporin in cell therapy for cardiac regeneration. J Cardiovasc Transl Res. 2014;7:475–82. 28. Sena E, van der Worp HB, Howells D, Macleod M. How can we improve the pre-clinical development of drugs for stroke? Trends Neurosci. 2007;30:433–9. 29. Schächinger V, Erbs S, Elsässer A, Haberbosch W, Hambrecht R, Hölschermann H, et al. Improved clinical outcome after intracoronary administration of bone-marrow-derived progenitor cells in acute myocardial infarction: final 1-year results of the REPAIR-AMI trial. Eur Heart J. 2006;27:2775–83. 30. Delewi R, Hirsch A, Tijssen JG, Schächinger V, Wojakowski W, Roncalli J, et al. Impact of intracoronary bone marrow cell therapy on left ventricular function in the setting of ST-segment elevation myocardial infarction: a collaborative meta-analysis. Eur Heart J. 2014;35:989–98. 31. Johnston PV, Sasano T, Mills K, Evers R, Lee ST, Smith RR, et al. Engraftment, differentiation, and functional benefits of autologous cardiosphere-derived cells in porcine ischemic cardiomyopathy. Circulation. 2009;120:1075–83. 32. Lee ST, White AJ, Matsushita S, Malliaras K, Steenbergen C, Zhang Y, et al. Intramyocardial injection of autologous cardiospheres or cardiosphere-derived cells preserves function and minimizes adverse ventricular remodeling in pigs with heart failure post-myocardial infarction. J Am Coll Cardiol. 2011;57:455–65. 33. Malliaras K, Li TS, Luthringer D, Terrovitis J, Cheng K, Chakravarty T, et al. Safety and efficacy of allogeneic cell therapy in infarcted rats transplanted with mismatched cardiosphere-derived cells. Circulation. 2012;125:100–12. 34. Kanazawa H, Tseliou E, Malliaras K, Yee K, Dawkins JF, Couto G De, et al. Cellular PostConditioning : Allogeneic Cardiosphere-Derived Cells Reduce Infarct Size and Attenuate Microvascular Obstruction When Administered After Reperfusion in Pigs With Acute Myocardial Infarction. Circ Heart Fail. 2015;8:322–32. 35. Yee K, Malliaras K, Kanazawa H, Tseliou E, Cheng K, Luthringer DJ, et al. Allogeneic Cardiospheres Delivered via Percutaneous Transendocardial Injection Increase Viable Myocardium, Decrease Scar Size, and Attenuate Cardiac Dilatation in Porcine Ischemic Cardiomyopathy. PLoS One. 2014;9:e113805.
10
215
PART THREE CHAPTER 10
SUPPORTING INFORMATION Supporting methods Histology For staining, slides were deparafinated by 2x10 minutes ultraclear, 2x5 minutes 99% EtOH, 2x5 minutes 96% EtOH and 5 minutes 7% EtOH. Slides were washed three times with dH2O for 5 minutes. For picrosirius staining, all slides were stained at once for 30 minutes in filtered picrosirus red. Slides were washed twice 0.2N HCl, then in dH2O for 5 minutes and dehydrated (quick 70% EtOH, 2x quick in 96% EtOH, 2x5 minutes 99% EtOH, 2x10 minutes Ultraclear), mounted with entallan and covered with coverslips. For lectin, deparaffinated slides were incubated with 3% H2O2 for 20 minutes and with 1% BSA in PBS for 15 minutes. Slides were incubated over night with Lectin (1:50 in PBS). After washing twice with PBS, slides were stained with DAB+ chromogen in DAB+ substrate buffer (Dako K3467) for a maximum of 4 minutes. Slides were washed with running tap water, dehydrated as above, mounted with entellan and covered with coverslips. Cyclosporin assays Migration assay A migration scratch assay was performed to compare migratory capacity. CMPCs were plated in 0.1% gelatin coated 6 well plates in medium with additional CsA in three concentrations. At 90% confluency a scratch was placed vertically in the middle of the well using a pipet tip, after which medium was changed and replaced with either fresh medium or medium containing CsA. Pictures of the initial scratches were immediately taken. After 3 and 5 hours new pictures were taken of the scratched area. Analysis of migration capacity was performed using the ImageJ software. Sprouting matrigel assay Angiogenesis μ-slides (Ibidi) were used to quantify matrigel tube formation. Slide chambers were coated with 10 μl ECMatrixTM (Millipore). CMPCs were seeded on the matrigel in 50 μl, with or without CsA. Tube formation was imaged after 13 hours and quantified using the Angioquant software in Matlab. Growth factor antibody array Conditioned medium was collected from the CMPCs after 9 days of culture in low serum medium without any additional growth factors, with or without CsA. Human Angiogenesis Array G1 (RayBiotech) was performed on the conditioned medium according to the provided instructions. Three conditions were tested; normal medium, 150 ng/mL and 300 ng/mL. Next, we tested culture medium from a different CMPC cell line as a positive control.
216
CMPCs IN HEART FAILURE
RESULTS Migration of CMPC was the same for all conditions (Supporting Figure 1). The migration was 51.0 ± 9.8 % for the normal medium and 42.8 ± 10.7%, 50.2 ± 10.1% and 57.5 ± 8.5% in the presence of CsA (50ng/mL, 150 ng/mL and 300 ng/mL respectively) (p=0.95) For angiogenesis, the number of segments per area, total length per area and length per segment were calculated. For all 3 measures, no difference exists between conditions (Supporting table 3, Figure 2). Conditioned medium of CMPCs cultured for 9 days contained Angiogenin (ANG), CXCL 1, 2&3 (GRO), Il-6, Il-8, CCL2 (MCP1), CCL5 (RANTES) and metallopeptidase inhibitor 1 (TIMP1). Presence of CsA did not affect growth factor secretion by CMPCs (Supporting Figure 3).
Supporting Figure 1. Quantification of cell migration.
10
Supporting Figure 2. Sprouting matrigel assay. A. bar plot of angiogenesis parameters. B. representative pictures of matrigel assays for all 4 conditions.
217
PART THREE CHAPTER 10
Supporting Figure 3. Array map and pictures of the array of 4 conditions at three different intensities. Positive control meaning different cell line without Cyclosporin.
Supporting table 1. Angiogenesis
218
medium
50 µg/L
150 µg/L
300 µg/L
P-value
Total number/area
1.26 ± 0.23
1.10 ± 0.09
1.05 ± 0.25
1.00 ± 0.25
0.49
Total length/area
86.0 ± 15.0
70.6 ± 12.0
63.1 ± 15.9
64.5 ± 12.8
0.25
Length/number
67.9 ± 1.8
64.1 ± 5.6
60.1 ± 1.8
64.9 ± 4.6
0.18
CMPCs IN HEART FAILURE
Supporting table 2. Functional outcome measured by PV-loop Cell Baseline
Placebo
Pre-infusion
Follow-up
Baseline
Pre-infusion
Follow-up
EDV (ml) 185.1 ± 43.7
237.1 ± 51.4
275.5 ± 64.1
188.1 ± 34.3
213.9 ± 51.3
274.9 ± 51.0
ESV (ml)
85.6 ± 33.3
141.2 ± 57.0
157.0 ± 38.3
84.9 ± 22.9
112.5 ± 39.9
154.2 ± 34.3
EF (%)
54.6 ± 9.8
43.5 ± 9.8
42.7 ± 6.0
55.0 ± 8.3
48.4 ± 9.3
43.7 ± 9.7
SW
1.3 ± 0.4
0.9 ± 0.8
1.6 ± 0.8
1.3 ± 0.4
0.9 ±0.7
1.7 ± 0.4
ESPVR
1.3 ± 0.6
1.2 ± 0.7
1.9 ± 1.8
1.3 ± 0.6
1.5 ± 0.9
1.2 ± 0.7
EDPVR
0.022 ± 0.020
0.008 ± 0.007
0.007 ± 0.004
0.017 ± 0.013
0.006 ± 0.004
0.004 ± 0.002*
dPdT+
1559.3 ± 196.3 1483.9 ± 414.0 1667.1 ± 324.9
1544 ± 210.4
1575.9 ± 182.3 1316.9 ± 97.9
dPdT-
-1231.3 ± 195.9 -1213.1 ± 379.4 -1305.5 ± 274.7
1388.4 ± 224.7 -1223.0 ± 248.8 -1077.7 ± 269.2
V0
-49.9 ± 71.8
-14.4 ± 71.5
-6.9 ± 139.0
-34.7 ± 49.0
-14.5 ± 57.5
-26.8 ± 54.3
Tau
45.8 ± 9.6
55.2 ± 22.4
53.4 ± 10.2
40.2 ± 8.4
48.4 ± 7.5
60.6 ± 14.8
EDV End diastolic volume, ESV End systolic volume, EF Left ventricular ejection fraction, SW Stroke work, ESPRV End systolic pressure volume relationship, EDPVR End diastolic pressure volume relationship, dPdT+ maximum pressure rise, dPdT- Minimum pressure rise, V0 Theoretical volume at zero pressure. * p=0.03 Cell treated animals compared to placebo. For EDV, ESV and EF n=7 for cell treated animals, n=8 for placebo treated animals. For SW, ESPRV, EDPVR, dPdT+, dPdT-, V0 and Tau n=6 per group.
Supporting table 3. Functional outcomes measured by echocardiography Cell Baseline t=0 (n=8)
Placebo
Pre-infusion (n=8)
Follow-up (n=8)
Baseline t=0 (n=8)
Pre-infusion (n=8)
Follow-up (n=8)
sWT mitral 0.18 ± 0.12
0.23 ± 0.12
0.27 ± 0.20
0.29 ± 0.20
0.28 ± 0.11
0.14 ± 0.16
sWT pap
0.36 ± 0.13
0.37 ± 0.15
0.35 ± 0.24
0.39 ± 0.15
0.35 ± 0.32
0.42 ± 0.41
sWT apex
0.58 ± 0.26
0.17 ±0.24
0.22 ± 0.27
0.47 ± 0.14
0.18 ± 0.21
0.23 ± 0.32
sWT mean
0.37 ± 0.13
0.26 ± 0.10
0.28 ± 0.13
0.38 ± 0.14
0.27 ± 0.16
0.26 ± 0.21
FS mitral
0.23 ± 0.06
0.29 ± 0.05
0.29 ± 0.05
0.28 ± 0.05
0.30 ± 0.04
0.27 ± 0.06
FS pap
0.23 ± 0.05
0.28 ± 0.05
0.27 ± 0.04
0.24 ± 0.05
0.30 ± 0.04
0.25 ± 0.06
FS apex
0.24 ± 0.08
0.20 ± 0.06
0.19 ± 0.08
0.24 ± 0.05
0.19 ± 0.06
0.18 ± 0.05
FS mean
0.23 ± 0.03
0.25 ± 0.04
0.25 ± 0.04
0.26 ± 0.05
0.24 ± 0.05
0.23 ± 0.04
FAS mitral
0.49 ± 0.09
0.55 ± 0.05
0.57 ± 0.05
0.54 ± 0.08
0.55 ± 0.07
0.53 ± 0.07
FAS pap
0.46 ± 0.08
0.47 ± 0.06
0.49 ± 0.08
0.48 ± 0.09
0.478 ± 0.12
0.44 ± 0.07
FAS apex
0.45 ± 0.07
0.40 ± 0.07
0.42 ± 0.08
0.47 ± 0.12
0.37 ± 0.08
0.36 ± 0.05
FAS mean
0.47 ± 0.03
0.42 ± 0.17
0.49 ± 0.06*
0.49 ± 0.08
0.47 ± 0.07
0.44 ± 0.04*
10
sWT Septal wall thickening, FS Fractional shortening, FAS Fractional area shortening at levels of the mitral valve (mitral) papillary muscle (pap), apex and mean of three levels (mean). N=8 per group. *p=0.08 for cell treated animals compared to placebo.
219
PART THREE
|
Elucidating (Epi)genetic and Translating
Therapeutic Pathways
Chapter
11
General Discussion
CHAPTER 11
GENERAL DISCUSSION The studies in this thesis tried to unravel the pathways relevant for cardiac regeneration and the basic mechanisms leading to heart failure (HF). The first part portrays the clinical course regarding the incidence of and prognostic factors associated with HF following myocardial infarction. The second part aimed to improve fibrosis detection in HF by optimizing cardiovascular magnetic resonance imaging (CMR) and histological analyses. The third and final part aimed to identify underlying (epi)genetic pathways leading to HF and focused on translation of cell therapy to the clinic. The goal of this thesis was to provide opportunities from bench-to-bedside to elucidate pathways associated with the failing heart. Opportunities in the failing heart following myocardial infarction The Global Burden of Disease 2010 Study estimated a global increased prevalence of ischaemic heart disease between 1990 and 2010, predominantly in males.1 The incidence of HF after MI could be changing due to variation in baseline population characteristics and improved treatment of MI.2 Currently, contemporary data of large cohorts regarding the incidence of HF following myocardial infarction (MI) are lacking. Information has become more widely available for clinical research with the electronic health record (EHR). While the EHR is not a direct reflection of the patients and physiology, but a reflection of the recording process inherent in healthcare with noise and feedback loops, it has been proposed to study the EHR as an object of interest in itself.3 Particularly in chronic disease, patients data is collected over time in different sources and linkage of these sources must be performed to obtain a more complete picture.4 Using linked EHR sources from primary and secondary care to identify MI patients (n = 24,745) in Chapter 2, we observed that in the current era 1) a substantial amount of patients develop HF after a first MI (24.3%, median follow-up 3.7 years) 2) increasing age, higher socioeconomic deprivation, a history of hypertension, diabetes, atrial fibrillation, peripheral arterial disease or COPD and a STEMI at presentation were important risk factors for HF following MI. The identified amendable prognostic factors can be used in future to decrease the incidence of HF following MI. The high incidence is supported by a different Swedish study (1993-2004) showing nearly one-third of MI patients aged 65-84 developed HF within 3 years. 5 On the other hand, ascertainment of cardiovascular outcomes based on clinical criteria creates a more direct reflection of the patients and physiology. In Chapter 3 we used a meticulously phenotyped cohort of patients (n = 1360) with a first ST-elevation myocardial infarction (STEMI) where followup was performed for incident HF. In this population, we found a relatively low HF incidence of 6.3% after STEMI during a median follow-up of 6.7 years. Higher peak creatine kinase (CK-MB) levels and an LAD (left anterior descending) artery culprit lesion at index STEMI were important risk factors for the outcome of HF. At the time of this research there was no EHR linkage with primary care data available in the Netherlands. With the advent of EHRs it would be important to create an infrastructure for linkage of EHR sources (primary care, secondary care and registers) for a more accurate estimation of health outcomes in the Netherlands.
222
GENERAL DISCUSSION
Improved cardiac fibrosis detection Cardiovascular Magnetic Resonance Imaging The aim of Chapter 4 was to systematically compare the relation between multiple CMR techniques and myocardial fibrosis in transverse cardiac slices of a chronic porcine infarct model. Hereby we tried to identify the most optimal technique to assess local myocardial fibrosis for translation into clinical practice. This systematic comparison can be readily adapted for use with other imaging modalities (e.g. SPECT, CT, PET and echocardiography). Myocardial strain and wall thickening showed modest explained variance for fibrosis. This might be explained by the resulting deformation caused by the contraction of cardiomyocytes in the vicinity of the infarct, as strain is not a strictly local phenomenon. For late gadolinium enhancement (LGE) imaging, we found that the full width at half maximum (FWHM) technique is preferred for the detection of myocardial fibrosis. This is in line with the patient study published by Flett et al., who studied patients with acute MI, chronic MI and hypertrophic cardiomyopathy and found that the FWHM technique for LGE quantification gave mean LGE volume results similar to manual quantification and was the most reproducible, regardless of underlying etiology.6 Myocardial fibrosis can lead to HF and act as a substrate for cardiac arrhythmias. Assessment of cardiac fibrosis is important for diagnosis, predicting prognosis, therapy guidance and monitoring. While LGE CMR provides an accurate qualitative measure of fibrosis, it has potential drawbacks as an indirect method requiring contrast agents without quantitative measurement of cardiac collagen. Assessment of extracellular volume (ECV) could be of key importance to predict prognosis in diffuse myocardial fibrosis.7, 8 Contrast-enhanced T1 mapping seems a promising method to quantify myocardial ECV in diffuse myocardial fibrosis.9 To overcome the limitations associated with contrast agent use, several endogenous contrast mechanisms for fibrosis detection are reviewed in Chapter 5. In this Chapter we also showed examples of T1rho and T2* mapping from a pilot study in a porcine chronic infarct model. Endogenous CMR fibrosis detection methods could obviate the need for exogenous contrast agents, favourable for patients with renal dysfunction, could improve reproducibility and decrease scan duration. Although T2* mapping might be a promising method for endogenous contrast, it is unclear whether postreperfusion intramyocardial hemorrhage with iron deposits can be distinguished from fibrosis.10 We recently showed the feasibility of native T1rho mapping translated from animals to patients.11 In 21 patients with chronic myocardial infarction, we found a significantly higher T1rho relaxation time in the infarct region compared to remote myocardium. Recently, a study using T1rho CMR to detect fibrosis in HCM patients has been reported.12 They observed a significant correlation between T1rho at 3SD and 4SD with visual LGE measurement. Although endogenous contrast
11
methods for fibrosis detection seem promising, more research is required before a large-scale application for clinical decision-making can be recommended. These study results warrant further investigation of endogenous contrast CMR in patients with replacement or diffuse fibrosis. Histological fibrosis quantification Phospholamban (PLN) plays a key role in calcium homeostasis and calcium cycling is crucial for cardiac excitation and contraction (Figure 1).13-15 While several causal PLN mutations have been described in humans, the PLN R14del mutation has been associated with dilated and
223
CHAPTER 11
arrhythmogenic cardiomyopathy.16 Previously, histological analysis of the myocardium of patients with the PLN R14del mutation has not been extensively performed. To the best of our knowledge, Chapter 6 is the first study investigating fibrosis and fatty tissue patterns in transverse heart slices of PLN R14del patients. Myocardial fibrosis was mainly observed in the posterolateral wall of the left ventricle, whilst adipose tissue was more pronounced in the outer myocardium of the right ventricle. Previous histopathological studies (not specifically PLN mutation associated) have shown similar areas of predilection in arrhythmogenic cardiomyopathy.
Figure 1. Calcium cycling in cardiomyocytes. In response to depolarization due to sodium (Na+) influx, calcium (Ca2+) enters the cytosol through the L-type calcium channels (LTCC) in the plasma membrane. This Ca2+ influx leads to calcium-induced calcium release from the sarcoplasmic reticulum mediated by the Ca2+ release channels (ryanodine receptors; RyRs). Calcium binds to troponin in the thin filaments of myofibrils to activate muscle contraction and Ca2+ is removed from the cytosol by sarcoplasmic reticulum Ca2+ ATPase (SERCA2a) and the sodium-calcium exchanger (NCX) on the plasma membrane. Phospholamban (PLN) is a reversible inhibitor of SERCA2a and in its dephosphorylated state PLN inhibits SERCA2a activity. Upon phosphorylation by protein kinase A (PKA), through the β-adrenergic receptor pathway, or Ca2+/calmodulindependent protein kinase (CaMKII), PLN dissociates from SERCA2a, thereby relieving Ca2+ pump inhibition and enhancing cardiac relaxation and contractility. Phospholamban is dephosphorylated by protein phosphatase (PP1), which ends the stimulation phase and PP1 is regulated by inhibitor-1 (I-1).
224
GENERAL DISCUSSION
According to our findings the sensitivity of endomyocardial biopsies from the right ventricular septum in PLN mutation associated cardiomyopathies might be low. The systematic fibrosis quantification method presented in this chapter could be useful to assess fibrosis and fatty tissue in a broad range of human cardiomyopathies and in preclinical studies as an outcome measure. Using the systematic histological quantification method developed in Chapter 6, in Chapter 7 we found a characteristic distribution of fibrosis in genetic cardiomyopathies related to the mutation group. The posterolateral wall of the left ventricle was found to be highly discriminating for the identification of the different groups of mutations. These histological patterns may provide a roadmap for cardiac imaging in patients and could be related to genetics and molecular biology of the corresponding tissue to elucidate underlying pathophysiology. In future nomenclature of cardiomyopathy types, it might be more suitable to use the underlying genetic defect.17 Identifying genes and pathways associated with the failing heart Despite advances in genetic research, limited genetic loci associated with risk of incident HF have been identified. A genome-wide association study (GWAS) examines many common variants in individuals to determine associations of these variants with a trait.18 Typically, associations between single nucleotide polymorphisms (SNPs) and disease are studied. Previously, 1 GWAS has been reported in unrelated individuals with HF.19 This GWAS only identified a single locus in each ethnic cohort that reached genome-wide significance. This might be explained by a large heterogeneity due to inclusion of HF cases with different underlying etiologies. Furthermore, only common genetic variants with a minor allele frequency of >5% were investigated. In future, clustering of weakly associated interacting genes might be identified using a pathway-based analysis approach.20 The zebrafish embryo can be studied to investigate the role of genes associated with cardiac disease.21 Molecular and cellular techniques to characterize gene function in vivo can be used to fully elucidate the role of those found genes. While the PLN R14del mutation has been associated with both dilated cardiomyopathy and arrhythmogenic cardiomyopathy, genetic cardiomyopathies have a variable disease penetrance and phenotype severity, even in the same family, thus other regulatory factors might play a role.22, 23 Next-generation sequencing allows for profiling of DNA regulatory elements (e.g., histone marks related to active promoters and enhancers) that control the timing, localization and level of gene transcription. Changes in these epigenetic profiles could serve as an indicator of disease pathogenesis. In Chapter 8, we aimed to identify differentially regulated genes and pathways in HF by chromatin immunoprecipitation and sequencing (ChIP-seq) of human
11
cardiac tissue from PLN R14del cardiomyopathy patients and healthy controls. We found 208 PLN R14del mutation specific regulated genes which could be used as a biomarker for risk stratification. These genes were mostly involved with general fibrotic pathways and developmental programs. Of these genes, 55 involved in lipid metabolism can be further investigated as candidate genes associated with fibrofatty replacement. The identified differentially regulated gene regions in HF could ultimately lead to a better understanding of underlying disease mechanisms, risk stratification and novel therapeutic strategies. Previously, p300, a transcriptional activator, has been associated with hypertrophy and HF through
225
CHAPTER 11
transcription by myocyte enhancer factor-2 and GATA-4 and it has been suggested that inhibition of p300 histone acetyltransferase activity by curcumin (a polyphenol responsible for the yellow color of the spice turmeric) may provide a novel therapeutic strategy for HF in humans.24, 25 Therapeutic opportunities Using the genetic knowledge, molecular defects associated with HF can be factored in therapeutic decision making and targeted with gene therapy.23 In HF sarcoplasmic reticulum Ca2+ ATPase (SERCA2a) expression is decreased26, thereby impairing cardiac contractility and relaxation. Gene transfer of SERCA2a in preclinical models, in rodents27, pigs28 and sheep29, showed improved cardiac function and energy potential in failing hearts. This led to the initiation of human clinical trials of SERCA2a gene transfer.30 Unfortunately, the SERCA2a CUPID phase 2b, double-blind, placebo-controlled, randomised trial (ClinicalTrials.gov Identifier: NCT01643330) did not meet the primary endpoint of time to recurrent event (heart failure-related hospitalisations in the presence of terminal events), but the procedures (e.g., dosing and delivery methods) could be optimised and this might pave the way for other gene therapy trials.31 MicroRNAs implicated in HF could also be potentially important targets for HF therapy.32, 33 Using the hypothesis of enhancement of the sensitivity of cardiac myosin to calcium, another approach is directed at the activation of cardiac myosin in patients with HF and reduced ejection fraction. Omecamtiv mecarbil, a cardiac myosin activator, does not seem to increase oxygen consumption or intracellular calcium and clinical trials in chronic HF are currently underway.34 Furthermore, major advances have been made with (assist) device therapy and the described therapeutic approaches, alone or in combination, could become important weapons in the ‘war’ against HF.35 Cell therapy could be a promising technique for cardiac repair, by regeneration of heart muscle in failing hearts and thereby improving cardiac function. Currently, the ideal cell type has not yet emerged. The advantages of cell therapy in chronic cardiomyopathy include a “stable” environment and a relatively large target patient population. In Chapter 9, the results of cell therapy for dilated cardiomyopathy seem promising, but this article has recommendations for future studies, such as improved methodological quality, elucidating underlying mechanisms and validating safety and efficacy before implementation in clinical practice. This includes recommendations for clearly described uniform definitions for inclusion and use of CMR to determine outcome measures. There is a need of consensus for methodological sound studies. It might be more appropriate to use differences between study arms measured in clinical endpoints such as (recurrent) hospitalisation for HF or mortality.35 With the aim of translation to the clinic, in Chapter 10, we used the previously reported human cardiomyocyte progenitor cells (CMPCs) as cell therapy in a preclinical study.36 We performed a randomised, placebo controlled study with intracoronary infusion of CMPCs in large animal model of ischaemic HF. While this therapeutic method seems feasible and safe, we did not find significant positive effects on left ventricular performance or infarct size. Therefore translation of the current CMPCs towards a first-in-man trial is presently not justified.
226
GENERAL DISCUSSION
With the advance of induced pluripotent stem cells (iPSCs)37 functional cardiomyocytes can be generated for patient-specific disease modelling or regenerative approaches. In a recent collaboration iPSCs were generated from a patient harbouring the PLN R14del mutation and differentiated into cardiomyocytes.38 These iPSC derived cardiomyocytes recapitulated Ca2+ handling abnormalities, electrical instability, an abnormal distribution of PLN protein in the cytoplasm and increased expression of cardiac hypertrophy markers. Gene correction of the causative R14del mutation using transcription activator-like effector nucleases (TALENs) restored the phenotype. These findings may pave the way for tailored therapy of genetic mutations associated with cardiomyopathies. Conclusion The failing heart presents challenges, ranging from basic mechanisms through the translational axis to the clinical course and vice versa. In this thesis we have shed light on opportunities to elucidate underlying mechanisms, visualize pathophysiology, identify the clinical picture and pave the way for future therapeutic strategies in the failing heart.
11
227
CHAPTER 11
REFERENCES 1.
2. 3. 4. 5.
6.
7.
8.
9.
10.
11.
12.
13. 14. 15. 16.
17.
18.
19.
228
Moran AE, Forouzanfar MH, Roth GA, Mensah GA, Ezzati M, Flaxman A, et al. The global burden of ischemic heart disease in 1990 and 2010: the Global Burden of Disease 2010 study. Circulation. 2014;129:1493-501. Mosterd A, Hoes AW. Clinical epidemiology of heart failure. Heart. 2007;93:1137-46. Hripcsak G, Albers DJ. Next-generation phenotyping of electronic health records. J Am Med Inform Assoc. 2013;20:117-21. Stevens RJ, McManus R. Unlinked data sources underestimate risk of cardiovascular disease. BMJ. 2013;346:f3737. Shafazand M, Rosengren A, Lappas G, Swedberg K, Schaufelberger M. Decreasing trends in the incidence of heart failure after acute myocardial infarction from 1993-2004: a study of 175,216 patients with a first acute myocardial infarction in Sweden. Eur J Heart Fail. 2011;13:135-41. Flett AS, Hasleton J, Cook C, Hausenloy D, Quarta G, Ariti C, et al. Evaluation of techniques for the quantification of myocardial scar of differing etiology using cardiac magnetic resonance. JACC Cardiovasc Imaging. 2011;4:150-6. Wong TC, Piehler K, Meier CG, Testa SM, Klock AM, Aneizi AA, et al. Association between extracellular matrix expansion quantified by cardiovascular magnetic resonance and short-term mortality. Circulation. 2012;126:1206-16. Wong TC, Piehler KM, Kang IA, Kadakkal A, Kellman P, Schwartzman DS, et al. Myocardial extracellular volume fraction quantified by cardiovascular magnetic resonance is increased in diabetes and associated with mortality and incident heart failure admission. Eur Heart J. 2014;35:657-64. Moon JC, Messroghli DR, Kellman P, Piechnik SK, Robson MD, Ugander M, et al. Myocardial T1 mapping and extracellular volume quantification: a Society for Cardiovascular Magnetic Resonance (SCMR) and CMR Working Group of the European Society of Cardiology consensus statement. J Cardiovasc Magn Reson. 2013;15:92. Gho JM, van Oorschot JW, Jansen of Lorkeers S, Froeling M, Luijten PR, Doevendans PA, et al. Endogenous contrast MRI of cardiac fibrosis: beyond late gadolinium enhancement. J Am Coll Cardiol. 2014;63(12_S):. doi:10.1016/S0735-1097(14)61008-1. van Oorschot JW, El Aidi H, Jansen of Lorkeers SJ, Gho JM, Froeling M, Visser F, et al. Endogenous assessment of chronic myocardial infarction with T(1rho)-mapping in patients. J Cardiovasc Magn Reson. 2014;16:104. Wang C, Zheng J, Sun J, Wang Y, Xia R, Yin Q, et al. Endogenous contrast T1rho cardiac magnetic resonance for myocardial fibrosis in hypertrophic cardiomyopathy patients. J Cardiol. 2015. doi: 10.1016/j.jjcc.2015.03.005. [Epub ahead of print] MacLennan DH, Kranias EG. Phospholamban: a crucial regulator of cardiac contractility. Nat Rev Mol Cell Biol. 2003;4:566-77. Bers DM. Cardiac excitation-contraction coupling. Nature. 2002;415:198-205. Kranias EG, Hajjar RJ. Modulation of cardiac contractility by the phospholamban/SERCA2a regulatome. Circ Res. 2012;110:1646-60. van der Zwaag PA, van Rijsingen IA, Asimaki A, Jongbloed JD, van Veldhuisen DJ, Wiesfeld AC, et al. Phospholamban R14del mutation in patients diagnosed with dilated cardiomyopathy or arrhythmogenic right ventricular cardiomyopathy: evidence supporting the concept of arrhythmogenic cardiomyopathy. Eur J Heart Fail. 2012;14:1199-207. Arbustini E, Narula N, Dec GW, Reddy KS, Greenberg B, Kushwaha S, et al. The MOGE(S) classification for a phenotype-genotype nomenclature of cardiomyopathy: endorsed by the World Heart Federation. J Am Coll Cardiol. 2013;62:2046-72. McCarthy MI, Abecasis GR, Cardon LR, Goldstein DB, Little J, Ioannidis JP, et al. Genome-wide association studies for complex traits: consensus, uncertainty and challenges. Nat Rev Genet. 2008;9:356-69. Smith NL, Felix JF, Morrison AC, Demissie S, Glazer NL, Loehr LR, et al. Association of genome-wide variation with the risk of incident heart failure in adults of European and African ancestry: a prospective meta-analysis from the cohorts for heart and aging research in genomic epidemiology (CHARGE) consortium. Circ Cardiovasc Genet. 2010;3:256-66.
GENERAL DISCUSSION
20. Wang K, Li M, Bucan M. Pathway-based approaches for analysis of genomewide association studies. Am J Hum Genet. 2007;81:1278-83. 21. Bakkers J. Zebrafish as a model to study cardiac development and human cardiac disease. Cardiovasc Res. 2011;91:279-88. 22. Cahill TJ, Ashrafian H, Watkins H. Genetic cardiomyopathies causing heart failure. Circ Res. 2013;113:660-75. 23. Fontaine GH, Zhang L. Is the phenotype-genotype relationship necessary to understand cardiomyopathies? Circ Cardiovasc Genet. 2014;7:405-6. 24. Morimoto T, Sunagawa Y, Kawamura T, Takaya T, Wada H, Nagasawa A, et al. The dietary compound curcumin inhibits p300 histone acetyltransferase activity and prevents heart failure in rats. J Clin Invest. 2008;118:868-78. 25. Yanazume T, Hasegawa K, Morimoto T, Kawamura T, Wada H, Matsumori A, et al. Cardiac p300 is involved in myocyte growth with decompensated heart failure. Mol Cell Biol. 2003;23:3593606. 26. Hasenfuss G, Reinecke H, Studer R, Meyer M, Pieske B, Holtz J, et al. Relation between myocardial function and expression of sarcoplasmic reticulum Ca(2+)-ATPase in failing and nonfailing human myocardium. Circ Res. 1994;75:434-42. 27. Sakata S, Lebeche D, Sakata N, Sakata Y, Chemaly ER, Liang LF, et al. Restoration of mechanical and energetic function in failing aortic-banded rat hearts by gene transfer of calcium cycling proteins. J Mol Cell Cardiol. 2007;42:852-61. 28. Kawase Y, Ly HQ, Prunier F, Lebeche D, Shi Y, Jin H, et al. Reversal of cardiac dysfunction after long-term expression of SERCA2a by gene transfer in a pre-clinical model of heart failure. J Am Coll Cardiol. 2008;51:1112-9. 29. Byrne MJ, Power JM, Preovolos A, Mariani JA, Hajjar RJ, Kaye DM. Recirculating cardiac delivery of AAV2/1SERCA2a improves myocardial function in an experimental model of heart failure in large animals. Gene Ther. 2008;15:1550-7. 30. Jaski BE, Jessup ML, Mancini DM, Cappola TP, Pauly DF, Greenberg B, et al. Calcium upregulation by percutaneous administration of gene therapy in cardiac disease (CUPID Trial), a first-in-human phase 1/2 clinical trial. J Card Fail. 2009;15:171-81. 31. Ratner M. Heart failure gene therapy disappoints but experts keep the faith. Nat Biotechnol. 2015;33:573-4. 32. Thum T, Gross C, Fiedler J, Fischer T, Kissler S, Bussen M, et al. MicroRNA-21 contributes to myocardial disease by stimulating MAP kinase signalling in fibroblasts. Nature. 2008;456:980-4. 33. Wahlquist C, Jeong D, Rojas-Munoz A, Kho C, Lee A, Mitsuyama S, et al. Inhibition of miR-25 improves cardiac contractility in the failing heart. Nature. 2014;508:531-5. 34. Cleland JG, Teerlink JR, Senior R, Nifontov EM, Mc Murray JJ, Lang CC, et al. The effects of the cardiac myosin activator, omecamtiv mecarbil, on cardiac function in systolic heart failure: a double-blind, placebo-controlled, crossover, dose-ranging phase 2 trial. Lancet. 2011;378:676-83. 35. Braunwald E. The war against heart failure: the Lancet lecture. Lancet. 2015;385:812-24. 36. Goumans MJ, de Boer TP, Smits AM, van Laake LW, van Vliet P, Metz CH, et al. TGF-beta1 induces efficient differentiation of human cardiomyocyte progenitor cells into functional cardiomyocytes in vitro. Stem Cell Res. 2007;1:138-49. 37. Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell. 2006;126:663-76. 38. Karakikes I, Stillitano F, Nonnenmacher M, Tzimas C, Sanoudou D, Termglinchan V, et al. Correction of human phospholamban R14del mutation associated with cardiomyopathy using targeted nucleases and combination therapy. Nat Commun. 2015;6:6955.
11
229
Samenvatting in het Nederlands List of publications Acknowledgements / Dankwoord Curriculum vitae
SAMENVATTING De studies in dit proefschrift probeerden de wegen te ontrafelen die relevant zijn voor cardiale regeneratieve geneeskunde en de basale mechanismen die leiden tot hartfalen. Het eerste deel portretteert het klinische beloop betreft de incidentie van en voorspellende factoren voor hartfalen na een hartinfarct. Het tweede deel was erop gericht om littekenweefsel in het hart beter te detecteren door optimalisatie van MRI technieken en histologische analyses. Het derde deel had ten doel om onderliggende erfelijke oorzaken te identificeren die leiden tot hartfalen en om de vertaalslag van stamceltherapie naar de kliniek te onderzoeken. Het uiteindelijke doel van dit proefschrift was mogelijkheden te creëren om resultaten uit het laboratorium naar de praktijk te vertalen en de wegen op te helderen die in verband zijn gebracht met het falende hart. Mogelijkheden in het falende hart na een hartinfarct Mogelijk verandert het vóórkomen van hartfalen na een hartinfarct vanwege verschillen in populatiekarakteristieken en een betere behandeling van het hartinfarct. Echter zijn er weinig hedendaagse studies op grote schaal verricht betreft het aantal nieuwe ziektegevallen (incidentie) van hartfalen na een hartinfarct. De ontwikkeling van elektronische patiëntendossiers (EPD’s) creëert de mogelijkheid om op grote schaal onderzoek te doen naar de epidemiologie van aandoeningen. Hoewel het EPD geen directe weerspiegeling is van de patiënt en fysiologie, met name bij chronische ziekten wordt data van patiënten verzameld gedurende de tijd in verschillende bronnen, kan door koppeling van meerdere bronnen een completer beeld worden gevormd. Door middel van een onderzoeksplatform welke meerdere EPD’s (uit Engelse huisartspraktijken, ziekenhuizen en registers) aan elkaar koppelt om patiënten met een hartinfarct te identificeren (n = 24.745) in hoofdstuk 2, zagen we dat in het huidige tijdperk 1) een substantieel deel van de patiënten hartfalen ontwikkelt na een eerste hartinfarct (24,3%, mediane follow-up 3,7 jaren) 2) een hogere leeftijd, toenemende mate van sociale deprivatie, een voorgeschiedenis met hoge bloeddruk, suikerziekte, boezemfibrilleren, perifeer vaatlijden of COPD en een ST-elevatie myocardinfarct (STEMI) bij presentatie belangrijke risicofactoren waren voor hartfalen volgend na een hartinfarct. De geïdentificeerde beïnvloedbare prognostische factoren kunnen in de toekomst gebruikt worden om de incidentie van hartfalen volgend na een hartinfarct te verminderen. Anderzijds, het vaststellen van uitkomsten betreft hart- en vaatziekten gebaseerd op klinische maatstaven creëert een meer directe weerspiegeling van de patiënt en fysiologie. In hoofdstuk 3 hebben we gebruik gemaakt van een zorgvuldig in kaart gebracht patiënten cohort (n = 1360) met een eerste STEMI waar vervolgonderzoek was verricht voor het optreden van hartfalen. In deze studiepopulatie vonden we een relatief lage hartfalen incidentie van 6,3% na STEMI gedurende een mediane follow-up van 6,7 jaren. Hogere piek creatine kinase (CK-MB) niveaus en een afgesloten LAD (left anterior descending) kransslagader bij index STEMI waren belangrijke risicofactoren voor de uitkomst hartfalen. Op het moment van dit onderzoek was er geen EPD koppeling met de eerste lijn beschikbaar in Nederland. Met de komst van de EPD’s is het van belang om een infrastructuur te creëren voor koppeling van verscheidene EPD bronnen (bijv. eerste lijn, tweede lijn en registers) voor een meer nauwkeurige schatting van gezondheidsuitkomsten in Nederland.
232
SAMENVATTING IN HET NEDERLANDS
Verbeterde herkenning van littekenweefsel in het hart MRI van het hart Het doel van hoofdstuk 4 was om een systematische vergelijking te maken van de relatie tussen verscheidene MRI technieken en littekenweefsel in het hart in transversale hartplakken bij een chronisch varkens infarct model. Hierbij probeerden wij de optimale techniek te identificeren om lokaal littekenweefsel in het hart vast te stellen voor translatie naar de kliniek. Deze systematische vergelijking kan eenvoudig aangepast worden voor gebruik met andere beeldvormingstechnieken (bijv. SPECT, CT, PET en echocardiografie). Lokale vervorming van de hartspier en wandverdikking op MRI lieten weinig verklaarde variantie voor littekenweefsel zien. Dit kan mogelijk verklaard worden doordat de lokale vervorming veroorzaakt wordt door samentrekking van hartspiercellen in de directe nabijheid van het infarct, daar lokale vervorming geen strikt lokaal fenomeen is. Op MRI na contrast toediening met late aankleuring (late gadolinium enhancement, LGE), bleek de halfwaardebreedte (full width at half maximum, FWHM) techniek bij voorkeur geschikt voor het herkennen van littekenweefsel in het hart. Littekenweefsel in het hart kan leiden tot hartfalen en hartritmestoornissen. Een inschatting maken van littekenweefsel in het hart is belangrijk voor het stellen van een diagnose, voorspellen van de prognose, het sturen van de behandeling en ter controle. Alhoewel de MRI met late aankleuring een nauwkeurige kwalitatieve maat van littekenweefsel in het hart verzorgt, heeft het potentiele nadelen als indirecte methode met de noodzaak om contrastmiddelen te gebruiken zonder te resulteren in een kwantitatieve maat voor littekenweefsel. Om de beperkingen gepaard gaande met het gebruik van contrastmiddelen te ondervangen, worden er een aantal MRI technieken zonder contrastmiddel (endogeen contrast) voor het herkennen van littekenweefsel besproken in hoofdstuk 5. In dit hoofdstuk worden ook voorbeelden van endogeen contrast MRI in de vorm van T1rho en T2* mapping bij een pilot studie in een chronisch varkens infarct model getoond. Endogeen contrast MRI van het hart voor herkenning van littekenweefsel kan de noodzaak van contrastmiddelen gebruik overbodig maken, wat gunstig is voor patiĂŤnten met nierfunctiestoornissen, kan de reproduceerbaarheid verbeteren en de scan tijd verkorten. Alhoewel T2* mapping een veelbelovende methode voor endogeen contrast kan zijn, is het onduidelijk of na herstel van de bloedstroom in de kransslagader bloedingen in de hartspier met ijzerafzettingen onderscheiden kunnen worden van littekenweefsel. Alhoewel endogene contrast methoden voor herkenning van littekenweefsel veelbelovend lijken, is er meer onderzoek nodig voordat het op grote schaal voor klinische besluitvorming kan worden aanbevolen. Deze studie resultaten rechtvaardigen verder onderzoek naar endogeen contrast MRI van het hart in patiĂŤnten met lokaal of diffuus littekenweefsel. Histologie Phospholamban (PLN) speelt een sleutelrol in de calciumhuishouding en calciumionen zijn cruciaal voor de elektrische activatie en het laten samentrekken van de hartspier. Alhoewel een aantal oorzakelijke mutaties in het PLN gen zijn beschreven in mensen, is de PLN R14del mutatie welke relatief vaak in Nederland voorkomt in verband gebracht met gedilateerde en aritmogene hartspierziekte. In het verleden zijn er slechts beperkte histologische (histologie = leer van de weefsels) analyses verricht van het hartweefsel van patiĂŤnten met de PLN R14del mutatie.
233
Voor zover wij weten is hoofdstuk 6 de eerste studie die onderzoek doet naar patronen van litteken- en vetweefsel in dwarsdoorsnedes van PLN R14del patiënten harten. Het littekenweefsel in het hart werd voornamelijk gezien in de achter- en zijwand van de linkerkamer, terwijl vetweefsel meer uitgesproken was in de buitenste laag hartweefsel van de rechterkamer. Naar aanleiding van onze bevindingen kan de gevoeligheid van hartbiopten uit de rechterkamer van het tussenschot in hartspierziekten met de PLN mutatie laag zijn. De gepresenteerde systematische methode om littekenweefsel te kwantificeren in dit hoofdstuk kan nuttig zijn om litteken- en vetweefsel vast te stellen in een breed scala aan hartspierziekten en als uitkomstmaat in preklinisch onderzoek. Door gebruik te maken van de methode uit hoofdstuk 6, hebben we in hoofdstuk 7 een karakteristiek verdelingspatroon van littekenweefsel in genetische hartspierziekten gevonden gerelateerd aan de mutatiegroep. De achter- en zijwand van de linkerkamer had een groot onderscheidend vermogen betreft het identificeren van de verschillende groepen mutaties. Deze patronen op histologie kunnen de weg wijzen voor beeldvorming van het hart in patiënten en kunnen gerelateerd worden aan de erfelijke informatie en processen op molecuulniveau in de cellen van het bijbehorende weefsel om onderliggende ziektemechanismen op te helderen. Bij de toekomstige naamgeving van soorten hartspierziekten is het wellicht meer gepast om het onderliggende gendefect te gebruiken. Identificeren van genen en paden die verband houden met het falende hart Ondanks vooruitgang in erfelijkheidsonderzoek, zijn er tot nu toe weinig plaatsen (loci) van genen in verband gebracht met het optreden van hartfalen. Hoewel de PLN R14del mutatie in verband gebracht is met zowel gedilateerde als aritmogene hartspierziekte, hebben genetische hartspierziekten een wisselende uiting en ernst van de ziekte, zelfs in dezelfde familie, dus mogelijk spelen andere regulerende factoren een rol. Next-generation sequencing maakt het mogelijk om regulerende elementen op DNA niveau te karakteriseren die het moment, de lokalisatie en het niveau van gen transcriptie beïnvloeden. Veranderingen in deze epigenetische karakteristieken kunnen dienen als een indicator voor het ontwikkelen van een ziekte. In hoofdstuk 8 hadden we tot doel om verschillen te identificeren in genregulatie en paden bij hartfalen tussen humaan hartweefsel van patiënten met een hartspierziekte op basis van de PLN R14del mutatie en gezonde controles met chromatine-immunoprecipitatie sequencing (ChIP-seq). We vonden 208 PLN R14del mutatie specifiek gereguleerde genen die gebruikt kunnen worden als biomarker om een risico-inschatting te maken. Deze genen waren voornamelijk betrokken bij paden die leiden tot littekenweefsel en ontwikkeling van het hart. Van deze genen zijn er 55 betrokken bij vetstofwisseling die verder bestudeerd kunnen worden als kandidaat genen voor ziekteprocessen met vervanging door litteken- en vetweefsel. De geïdentificeerde verschillend gereguleerde erfelijke gebieden in hartfalen kunnen uiteindelijk leiden tot een beter begrip van onderliggende ziektemechanismen, een betere risico-inschatting en nieuwe behandelopties. Mogelijkheden voor behandeling Door kennis van de erfelijke oorzaken te gebruiken kunnen gebreken op molecuulniveau die in verband zijn gebracht met hartfalen mogelijkheden bieden voor behandelopties, bijvoorbeeld met gentherapie. Stamceltherapie kan een veelbelovende techniek zijn voor herstel van de
234
SAMENVATTING IN HET NEDERLANDS
hartspier, door regeneratie van de hartspier bij hartfalen en daardoor een verbetering van de hartspierfunctie. Op dit moment is het ideale celtype nog niet naar voren gekomen. De voordelen van stamceltherapie bij chronische hartspierziekten behelzen een “stabiel” milieu en een relatief grote geschikte patiëntenpopulatie. In hoofdstuk 9 lijken de resultaten van stameltherapie voor gedilateerde hartspierziekte veelbelovend, maar dit hoofdstuk heeft aanbevelingen voor toekomstige studies, zoals het verbeteren van de methodologische kwaliteit, ophelderen van onderliggende mechanismen en valideren van veiligheid en effectiviteit voorafgaand aan implementatie in de klinische praktijk. Dit omvat aanbevelingen voor helder beschreven uniforme definities qua inclusiecriteria en gebruik van cardiale MRI om uitkomstmaten te bepalen. Er is een noodzaak voor overeenstemming binnen het onderzoeksveld betreft methodologisch deugdelijke studies. Met als doel om een vertaalslag naar de kliniek te maken in hoofdstuk 10, hebben we hartspier voorlopercellen (CMPCs) uit het menselijke hart als stamceltherapie gebruikt in een preklinische studie. We hebben een gerandomiseerde, placebo gecontroleerde studie met infusie van CMPCs in de kransslagader in een groot diermodel van ischemisch hartfalen uitgevoerd. Hoewel deze behandel methode uitvoerbaar en veilig lijkt, vonden we geen significante positieve effecten op de linkerkamerfunctie of grootte van het infarct. Daarom is de vertaling van de huidige CMPCs naar een first-in-man trial op dit moment nog niet gerechtvaardigd. Conclusie Het falende hart biedt uitdagingen die variëren van basale mechanismen via de translationele as naar de kliniek en omgekeerd met bevindingen uit de praktijk als aanleiding voor (laboratorium) onderzoek. In dit proefschrift hebben we mogelijkheden belicht om onderliggende mechanismen op te helderen, ziektemechanismen in beeld te brengen, het klinisch beeld te identificeren en de weg te banen voor toekomstige behandelstrategieën in het falende hart.
235
LIST OF PUBLICATIONS Ganesh SK*, Tragante V*, Guo W*, Guo Y, Lanktree MB, Smith EN, Johnson T, Castillo BA, Barnard J, Baumert J, Chang YP, Elbers CC, Farrall M, Fischer ME, Franceschini N, Gaunt TR, Gho JM, …, Doevendans PA, …, de Bakker PI*, Zhu X*, Levy D*, Keating BJ*, Asselbergs FW*. Loci influencing blood pressure identified using a cardiovascular gene-centric array. Hum Mol Genet. 2013;22:1663-78. *These authors contributed equally. Koudstaal S, Jansen of Lorkeers SJ, Gaetani R, Gho JM, van Slochteren FJ, Sluijter JP, Doevendans PA, Ellison GM, Chamuleau SA. Concise review: heart regeneration and the role of cardiac stem cells. Stem Cells Transl Med. 2013;2:434-43. Gho JM, Kummeling GJ, Koudstaal S, Jansen of Lorkeers SJ, Doevendans PA, Asselbergs FW, Chamuleau SA. Cell therapy, a novel remedy for dilated cardiomyopathy? A systematic review. J Card Fail. 2013;19:494-502. Gho JM, van Es R, Stathonikos N, Harakalova M, te Rijdt WP, Suurmeijer AJ, van der Heijden JF, de Jonge N, Chamuleau SA, de Weger RA, Asselbergs FW, Vink A. High resolution systematic digital histological quantification of cardiac fibrosis and adipose tissue in phospholamban p.Arg14del mutation associated cardiomyopathy. PloS One. 2014;9:e94820. Tragante V*, Barnes MR*, Ganesh SK*, Lanktree MB, Guo W, Franceschini N, Smith EN, Johnson T, Holmes MV, Karczewski KJ, Almoguera B, Barnard J, Baumert J, Chang YP, Elbers CC, Farrall M, Fischer ME, Gaunt TR, Gho JM, ..., Doevendans PA, ..., Levy D*, Asselbergs FW*, Munroe PB*, Keating BJ*. Gene-centric meta-analysis in 87,736 individuals of European ancestry identifies multiple blood pressure related loci. Am J Hum Genet. 2014;94:349-60. *These authors contributed equally. van Hout GP*, Jansen of Lorkeers SJ*, Gho JM, Doevendans PA, van Solinge WW, Pasterkamp GP, Chamuleau SA, Hoefer IE. Admittance based pressure volume loops versus gold standard cardiac magnetic resonance imaging in a porcine model of myocardial infarction. Physiol Rep. 2014;2:e00287. *These authors contributed equally. Koudstaal S, Jansen of Lorkeers SJ, Gho JM, van Hout GP, Jansen MS, Grundeman PF, Pasterkamp G, Doevendans PA, Hoefer IE, Chamuleau SA. Myocardial infarction and functional outcome assessment in pigs. J Vis Exp. 2014;(86). van Oorschot JW, El Aidi H, Jansen of Lorkeers SJ, Gho JM, Froeling M, Visser F, Chamuleau SA, Doevendans PA, Luijten PR, Leiner T, Zwanenburg JJ. Endogenous assessment of chronic myocardial infarction with T1ρ-mapping in patients. J Cardiovasc Magn Reson. 2014;16:104.
236
LIST OF PUBLICATIONS
Karakikes I*, Stillitano F*, Nonnenmacher M*, Tzimas C, Sanoudou D, Termglinchan V, Kong CW, Rushing S, Hansen J, Ceholski D, Kolokathis F, Kremastinos D, Katoulis A, Ren L, Cohen N, Gho JM, Tsiapras D, Vink A, Wu JC, Asselbergs FW, Li RA, Hulot JS, Kranias EG, Hajjar RJ. Correction of the human phospholamban R14del mutation associated with cardiomyopathy using targeted nucleases and combination therapy. Nat Commun. 2015;6:6955. *These authors contributed equally. van Oorschot JW*, Gho JM*, van Hout GP, Froeling M, Jansen of Lorkeers SJ, Hoefer IE, Doevendans PA, Luijten PR, Chamuleau SA, Zwanenburg JJ. Endogenous contrast MRI of cardiac fibrosis: beyond late gadolinium enhancement. J Magn Reson Imaging 2015;41:1181-9. *These authors contributed equally. Gho JM*, van Es R*, van Slochteren FJ, Jansen of Lorkeers SJ, Hauer AJ, van Oorschot JW, Doevendans PA, Leiner T, Vink A, Asselbergs FW, Chamuleau SA. A systematic comparison of cardiovascular magnetic resonance and high resolution histological fibrosis quantification in a chronic porcine infarct model. Submitted. *These authors contributed equally. Jansen of Lorkeers SJ, Gho JM, Koudstaal S, van Hout GP, Zwetsloot PP, van Oorschot JW, van Eeuwijk EC, Leiner T, Hoefer IE, Goumans MJ, Doevendans PA, Sluijter JP, Chamuleau SA. Xenotransplantation of human cardiomyocyte progenitor cells does not improve cardiac function in a porcine model of chronic ischemic heart failure. Submitted.
237
ACKNOWLEDGEMENTS / DANKWOORD In de afgelopen jaren is dit proefschrift tot stand gekomen door de samenwerking tussen uiteenlopende disciplines, in binnen- en buitenland. Ik heb volop genoten van deze leerzame en gezellige promotietijd. Hierbij wil ik iedereen bedanken die hier een bijdrage aan hebben geleverd. Geachte prof.dr. Doevendans, beste Pieter, bij ons eerste gesprek betreft een mogelijk promotietraject had je meteen een aantal mogelijkheden in gedachten betreft onderzoek en het advies was om eens met de andere begeleiders te gaan praten. Ik wist nog niet waar dit naartoe zou leiden toen je vertelde dat het met deze begeleiders een “mooi feestje” zou worden. Uiteindelijk had ik me geen ander promotietraject kunnen wensen. Dankzij je sturende rol zijn er onder andere mooie samenwerkingsverbanden ontstaan binnen SMARTCARE en met Stichting Genetische Hartspierziekte PLN die binnen dit proefschrift samenkomen en hopelijk vervolg krijgen. Beste Folkert, inmiddels prof.dr. Asselbergs, je streven om zaken bij voorkeur gisteren gedaan te krijgen en respons op e-mails met de snelheid van next-generation sequencing houdt de vaart erin. Daarnaast viel mij meteen je directheid op welke goed van pas komt om je heldere visie, die je goed weet over te brengen, waar te maken. Ik vond het leuk om relatief ongebaande paden te verkennen waarbij nieuwe onderzoekslijnen werden gecreëerd. Dank voor de mogelijkheid om aan verschillende studies in binnen- en buitenland mee te werken, als wereldreiziger schoot ook de begeleiding in Londen nooit tekort. Ik zie ernaar uit om binnenkort te gaan tennissen als je een keer in Utrecht bent. Geachte dr. Chamuleau, beste Steven, ik zal nooit de wijze les vergeten tijdens een van de eerste weekenddiensten waar ik als semi-arts meeliep in het UMC Utrecht. Zoals je van je eigen opleider had geleerd, adviseerde je me om na een paar dagen nog even contact op te nemen met de patiënt die zich op de spoedeisende hulp had gepresenteerd en weer huiswaarts was gegaan. Gelukkig voelde de patiënt zich alweer een stuk beter. Die onderwijzende en motiverende rol is je op het lijf geschreven. Een van de assistenten toentertijd raadde me aan om contact met je op te nemen als ik interesse had in wetenschappelijk onderzoek. Uiteindelijk ben ik zeer dankbaar voor dit traject in de breedste zin van het woord, met teveel opportunities om op te noemen zoals de leuke een leerzame congresbezoeken in Washington en Madrid, vooruitstrevende projecten en de nuttige besteding van het persoonlijk budget. Ik waardeer je openheid en eerlijkheid en hoop dat ik de komende tijd nog veel van je kan leren. Nu ik zo in mijn mailbox terugkijk was het de bedoeling om onderzoek te doen naar DCM (op het snijvlak imaging, cardiogenetica en stamcellen) en als ik door dit boekje heen blader is dat to the best of my knowledge aardig gelukt. Maar we zijn er nog niet, er is in de toekomst nog genoeg te onderzoeken en ik hoop dat we kunnen blijven samenwerken from bench-to-bedside.
238
ACKNOWLEDGEMENTS / DANKWOORD
Beste Aryan, dr. Vink, het is een eer voor mij dat je op het laatste moment als 2e copromotor op kan treden. Datgene wat volgens mij als zijproject is begonnen vormt nu een integraal deel van dit proefschrift. Ik zal niet vergeten dat je mij bij het eerste telefoongesprek adviseerde om in Robbins te kijken. Dank voor je intensieve begeleiding, enthousiasme en bevlogenheid. Beste leden van de leescommissie, prof.dr. R. Goldschmeding, prof.dr. M.A.A.J. van den Bosch, prof.dr. M.R. Vriens, prof.dr. A.C. van Rossum en prof.dr. E.D. de Muinck, bedankt voor de bereidheid om zitting te nemen in de beoordelingscommissie van dit proefschrift. De stafleden van de afdeling Cardiologie in het UMC Utrecht, dank voor de bijdragen aan mijn onderzoek in de afgelopen jaren en de leerzame supervisie in de kliniek. Beste dr. Kirkels, dank voor de mogelijkheid om in de kliniek aan de slag te gaan met vele onderwijs momenten. Maarten-Jan, dank voor je nimmer aflatende enthousiasme en goede raad betreft de kliniek en onderzoek. Het stafsecretariaat, Jantine, Sylvia en Tamara, dank voor al jullie hulp in de afgelopen jaren. Jonne en Ingrid, dank voor jullie gezelligheid, hulp en kunde om afspraken te maken met mijn dagelijkse begeleiders en velerlei vragen tot een goed einde te brengen. Dank aan alle mede-onderzoekers van de afgelopen jaren, Anneline (voor de discussies over PLN, tot snel in de kliniek!), Cas, Cheyenne (sightseeing Madrid), Crystel, Dirk (zie je bij de volgende BBQ), Fatih G, Freek (eindelijk gebruik maken van de Museumkaart met Mariam), Geert van H (naast onderzoek ook voor de gezamenlijke tijd in de kliniek), Gijs (voor je onderzoeksen muzikale input), Hamza, Iris, Jetske (voor de vermakelijke borrels en filmpjes), Judith (brainstormen over PLN), Manon (hopelijk tot snel in het Diak!), Mariam, Marloes, Martine, Mieke (fietsen en basketbal kijken in Washington), Mira (op zoek naar het Villa pleintje in Madrid), Peter-Paul (gecellige borrels en leerzame discussies), Remco (als kamergenoot in Oirschot tot in de Villa en organisator van menig uitje), RenĂŠ (voor het tot in de late uurtjes optimaliseren van de manuscripten, borrelen en de vele sportactiviteiten), Rosemarijn (van buur bij de IBB tot de Rotaractse activiteiten), Sanne (voor de vele en gezellige uren, Oink! momenten en hopelijk tot ziens in de kliniek!), Sjoukje, Sofieke, Stefan (van bunker tot Villa en een paar keer de Noordzee over), Thijs (nice dat je ook begonnen bent), Thomas, Tycho (van tips in de bunker, op de tennisbaan tot aan B4West), Vivan, Willemien (vanaf A0 in Amersfoort tot het ACC in Washington), Wouter G en Wouter van E, voor al jullie hulp en gezelligheid. Alle arts-assistenten Cardiologie in het UMC Utrecht, dank voor jullie collegialiteit, geduld en de leerzame en gezellige sfeer. De R&D van de Cardiologie, dank voor jullie gezelligheid, inzet en hulp bij de verschillende studies.
239
Alle betrokkenen van de Experimentele Cardiologie. Dank voor jullie bijdrage aan de projecten, met name op basaal gebied. Alle biotechnici, GDL medewerkers en dierverzorgers, dank voor jullie hulp, expertise en geduld bij de soms langdurige experimenten. Alle betrokkenen van de afdeling Radiologie betreft de vele MRI uren, in het bijzonder Joep en Jaco onder begeleiding van prof.dr. Luijten dank voor de fysieke inzet en de fysische input en prof.dr. Leiner dank voor de waardevolle scherpe klinische blik. Mooi om de vertaling van concept naar klinische studie te zien. Een aantal van de afdeling Pathologie zijn al eerder genoemd, maar hierbij wil ik ook in het bijzonder Shahrzad, Roel en de betrokken analisten bedanken voor de onmisbare hulp, mogelijkheden en expertise bij het onderzoek naar de erfelijke hartspierziekten. Nikolas, thank you for your extensive help with creating the fibrosis quantifier (FibroQuant). Magdalena, thank you for all your help and precise efforts, especially regarding epigenetics, also together with Michal. Vinicius, thanks for all your bioinformatic input and output over the last years. Alle studenten die aan de projecten bij dit proefschrift hebben meegewerkt. In het bijzonder Maartje voor de hulp bij de AGNES studie. En Einar, succes met je onderzoek. Uit het AMC Amsterdam alle betrokkenen bij de AGNES studie. Ik hoop voor de toekomst op een blijvende samenwerking, bijvoorbeeld in het GENIUS-CHD consortium. I would like to thank all the people at the Farr Institute. Thank you for all your help with epidemiological and statistical input, R-coding skills, but also the wonderful time with (International cuisine) lunches and post-work drinks in London. I am still not tired of London and hope to visit you again soon. Dr. Hajjar and Francesca, thank you for your collaboration regarding PLN and visit to your lab in New York. Hope your research will pave the way for novel therapies in genetic cardiomyopathies. Pieter Glijnis, namens Stichting Genetische Hartspierziekte PLN. Dank voor alle ondersteuning en de interesse gedurende het onderzoek, ik hoop dat er in de toekomst betere behandelmogelijkheden komen voor deze hartspierziekte. Van het Meander MC de maatschap Cardiologie en verpleegkundigen, dank voor de leerzame perioden onderzoek en in de kliniek, maar vooral ook gezellige tijd in Amersfoort. In het bijzonder dr. Senden, beste Jeff, bedankt voor de leerzame tijd in de kliniek en het advies om onderzoek te gaan doen met â&#x20AC;&#x153;stamcellenâ&#x20AC;?. Hoop jullie weer te zien bij het volgende zeiluitje!
240
ACKNOWLEDGEMENTS / DANKWOORD
William Kramer dank voor de begeleiding tijdens de studie in de congrescommissie en bij het duikgeneeskunde artikel. Wendy, dank voor je inzet en betrokkenheid bij de lay-out. Robert, dank voor je sprekende ontwerp voor de omslag. Mijn paranimfen, Ing Lok en Frebus. Beste Ing Lok, lief broertje, ik vind het mooi dat we zoveel interesses delen, we hebben de afgelopen jaren al menig uitstapje gemaakt van wintersport tot Wimbledon. Ik bewonder je sociale kant en hulpvaardigheid en vind het leuk dat je nu ook in het Meander MC aan de slag bent gegaan. Beste Frebus, mien jong, Slo is inmiddels een begrip geworden, leuk om de afgelopen jaren met je samengewerkt te hebben waarbij je unieke technische blik echt een eyeopener is geweest. Daarnaast herinner ik me ook vele mooie uitstapjes en feestjes en heb ik je zien groeien als â&#x20AC;&#x153;CEslOâ&#x20AC;? van je eigen bedrijf. Leuk dat jullie mijn paranimfen zijn. Dank aan al mijn familie en vrienden die mij de afgelopen jaren hebben gesteund en interesse hebben getoond voor wat ik ook alweer aan het doen was. Dank voor jullie gastvrijheid en gezelligheid. Mama en papa, dank voor jullie opvoeding, steun en goede raad in de afgelopen jaren waar ik altijd op kan rekenen. Mama, ik ben nog altijd dankbaar voor je advies om Biologie toch als keuzevak erbij te kiezen op de middelbare school wat ik toen toch maar heb opgevolgd.
241
CURRICULUM VITAE Johannes Michael Ing Han Gho was born on 29 September 1986 in Leusden, the Netherlands. After attending the Meridiaan College, Het Nieuwe Eemland in Amersfoort where he obtained his gymnasium diploma in 2004, he studied medicine at Utrecht University. During his medical study, he passed the propaedeutic exam cum laude (with honor), spent time abroad in Malaysia, Australia and Malawi for internships and became interested in Cardiology and scientific research. After graduation from medical school in 2010, he started working as a resident (ANIOS) at the Cardiology department of the Meander Medical Center in Amersfoort under supervision of dr. P.J. Senden. In January 2012 he started working as a PhD student at the Cardiology department of the University Medical Center Utrecht under supervision of prof.dr. P.A.F.M. Doevendans, prof.dr. F.W. Asselbergs, dr. S.A.J. Chamuleau and dr. A. Vink. During this period he worked on several research projects with a main focus on heart failure, including genetic, regenerative medicine, imaging, histological and epidemiological studies. He also conducted 10 weeks of research at the Farr Institute of Health Informatics Research, University College London (UCL), London, United Kingdom, which resulted in a manuscript included in this thesis (Chapter 2). In July 2015 he started working as a resident (ANIOS) at the Cardiology department of the University Medical Center Utrecht and he will start his formal Cardiology training under supervision of dr. J.H. Kirkels in January 2016.
242