Dilated Cardiomyopathy Caused by Truncating Titin Variants

BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) J Med Genet

Dilated cardiomyopathy caused by truncating titin variants

– Long-term outcomes, arrhythmias, response to

treatment and sex differences

SUPPLEMENTARY MATERIAL

Christoffer Rasmus Vissing1*, MD; Torsten Bloch Rasmussen2, MD, PhD; Anne Mette Dybro2, MD; Morten

Salling Olesen3,4, MSc, PhD; Lisbeth Nørum Pedersen5; MSc, PhD; Morten Jensen, MD, PhD2; Henning

Bundgaard1, MD, DMSc; Alex Hørby Christensen1,6 MD, PhD

1The Capital Region’s Unit for Inherited Cardiac Diseases, Department of Cardiology, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark 2Department of Cardiology, Aarhus University Hospital, Aarhus, Denmark 3Laboratory for Molecular Cardiology, University of Copenhagen, Copenhagen, Denmark 4Department of Biomedical Sciences, University of Copenhagen, Copenhagen, 2200 N, Denmark 5Department of Molecular Medicine, Aarhus University Hospital, Denmark. 6Department of Cardiology, Herlev-Gentofte Hospital, Copenhagen University Hospital, Copenhagen, Denmark

*Corresponding author:

Christoffer Rasmus Vissing, MD

The Capital Region’s Unit for Inherited Cardiac Diseases, Department of Cardiology, The Heart Center, Rigshospitalet, Copenhagen University Hospital, Blegdamsvej 9, DK-2100 Copenhagen, Denmark,

E-mail: [email protected]; Phone +45 3545 5045

Vissing CR, et al. J Med Genet 2020;0:1–10. doi: 10.1136/jmedgenet-2020-107178 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) J Med Genet

Supplementary Materials Contents Dilated cardiomyopathy caused by truncating titin variants – Long-term outcomes, arrhythmias, response to treatment and sex differences ...... 1 SUPPLEMENTARY METHODS ...... 3 Definitions ...... 3 Genetic Screening ...... 3 Statistical analyses ...... 4 INFORMATION ON PATIENTS WITH MULTIPLE VARIANTS: ...... 5 FIGURES ...... 6 Supplementary Figure 1: Forest-plots depicting hazard ratios in Cox proportional hazards models for the three studied cardiac outcomes ...... 6 Supplementary Figure 2: Forest-plots depicting univariable hazard ratios of 6 clinical variables in the three studied cardiac outcomes ...... 7 Supplementary Figure 3: Outcomes according to site of truncation in titin ...... 8 Supplementary Figure 4: Outcomes in probands vs relatives ...... 9 SUPPLEMENTARY TABLES ...... 10 Supplementary Table 1. Identified truncating TTN variants. All variants are annotated in relation to the titin-metatranscript (NM_001267550.1) ...... 10 Supplementary Table 2: Association of clinical variables with combined end-point (implantation of LVAD, heart transplantation or death) in univariable and multivariable cox regression analysis ...... 21 Supplementary Table 3: Association of clinical variables with ventricular arrhythmia in univariable and multivariable cox regression analysis...... 23 Supplementary Table 4: Association of clinical variables with atrial fibrillation in univariable cox regression analysis...... 25

SUPPLEMENTARY METHODS

Definitions Severe systemic hypertension

We defined the exclusion criteria severe systemic hypertension as: an untreated systolic blood pressure > 180mmHg, and/or an untreated diastolic blood pressure > 110 mmHg, and/or a systolic blood pressure > 160 mmHg on antihypertensive treatment and/or a diastolic blood pressure > 100 mmHg on antihypertensive treatment.

Diabetes mellitus

Patients were registered as having diabetes if they had a clinical diagnosis of diabetes mellitus or if they were prescribed antidiabetic medications. Severely dysregulated diabetes mellitus was defined as patients with neuropathy, nephropathy or retinopathy.

Metabolic, infectious or inflammatory cardiomyopathies

We defined the exclusion criteria for the above terms to encompass myocarditis caused by viral, bacterial, fungal or parasitic infections. Autoimmune diseases including giant cell myocarditis, non-infectious myocarditis, polymyositis/dermatomyositis, Churg-Strauss syndrome, Wegener’s granulomatosis, systemic lupus erythematosus or sarcoidosis. Nutritional deficiencies or hemochromatosis. Uncontrolled hypothyroidism or hyperthyroidism, Cushing’s disease, Addison disease, pheochromocytoma, acromegaly, fatty acid oxidation disorders, carnitine deficiency, glycogen storage diseases, organic acidurias and disorders of oxidative phosphorylation.

Genetic Screening Old genetic panel

The first genetic panel used for screening cardiomyopathy patients at our institutions included the following 14 genes: ACTC, CRSP3, LAMP2, LMNA, MYL2, MYL3, MYH7, MYBPC3, PRKAG2, (RBM20), SCN5A, TNNI3, TNNT2 and TPM. RBM20 is in brackets since it was not a part of the panel originally, but all patients have been sequenced in RBM20 after the initial screening. This panel was implemented in 2006 and was in

clinical use until late 2018, although most cardiomyopathy patients were screened with additional genetic panels.

New genetic panel (Copenhagen)

The current genetic panel in Copenhagen consists of the following 62 genes:

ACTC, ACTN2, ANKRD1, BAG3, CALR3, CASQ2, CAV3, CRYAB, CSRP3, CTF1, CTNNA3, DES, DMD, DNAJC19, DSC2, DSG2, DSP, DTNA, EMD, EYA4, FHL1, FHL2, FLNC, GATAD1, GLA, HCN4, ILK, JPH2, JUP, LAMA4, LAMP2, LDB3, LMNA, MIB1, MYBPC3, MYH6, MYH7, MYL2, MYL3, MYLK2, MYOZ2, MYPN, NEBL, NEXN, OBSCN, PKP2, PLN, PRDM16, PRKAG2, PSEN1, PSEN2, RBM20, SCN5A, SGCB, SGCD, SLC22A5, TAZ, TCAP, TMEM43, TMPO, TNNC1, TNNI3, TNNI3K, TNNT2, TPM1, TTN, TTR, VCL

New genetic panel (Aarhus)

The current genetic panel in Aarhus consists of the following 102 genes:

ABCC9, ACTC1, ACTN2, AKAP9, ANK2, ANKRD1, ANO1, APOB, BAG3, BEST3, CACNA1C, CACNA2D1, CACNB2, CALM1, CALM2, CALM3, CASQ2, CAV3, CDH2, CRYAB, CSRP3, CTNNA3, DES, DMD, DNAJC19, DSC2, DSG2, DSP, DTNA, EMD, EYA4, FHL1, FHL2, FKTN, FLNC, FXN, GATA4, GLA, GPD1L, GAA, HCN4, JPH2, JUP, KCND3, KCNE1M KCNE2M KCNE3, KCNE5, KCNH2, KCNJ5, KCNJ8, KCNQ1, LAMA4, LAMP2, LDB3, LDLR, LMNA, MYBPC3, MYH6, MYH7, MYL2, MYL3, MYOZ2, MYPBC3, MYPN, NEBL, NEXN, PCSK9, PKP2, PLN, PRDM16, PRKAG2, PSEN1, PSEN2, PTPN11, RAF1, RANGRF, RBM20, RYR2, SCN10A, SCN1B, SCN2B, SCN3B, SCN4B, SCN5A, SGCD, SLC4AE, SNTA1, TAZ, TCAP, TMEM43, TMPO, TNNC1, TNNI3, TNNT2, TPM1, TRDN, TRPM4, TTN, TTR, VCL and ZBTB17.

Statistical analyses

Cox proportional hazards models were used to examine the relationship between baseline variables and the three studied outcomes which included 1) A composite outcome of occurrence of implantation of left ventricular assist device, heart transplantation or death (the combined outcome), 2) The occurrence of sustained ventricular tachycardia, ventricular fibrillation, sudden cardiac death, aborted sudden cardiac death or appropriate shock by implantable cardioverter defibrillator (the ventricular arrhythmia outcome), and 3) The occurrence of atrial fibrillation or fluttering. The timing to outcomes was defined as follow-up time from time of dilated cardiomyopathy diagnosis. Multivariable Cox proportional-hazards regression analyses were created by including pre-defined variables known to be clinically meaningful: For the combined outcome these predefined variables included age at diagnosis of DCM, LVEF at diagnosis and

whether the patient was a proband. In the two other studied outcomes the predefined variables included age at diagnosis of DCM and whether the patient was a proband or not. Other variables included in the final models were added to the model in a stepwise forward manner. In order for the other variables to be included in the model, they had to fulfill the following criteria: 1) they had to have a p-value of maximum 0.1 in univariable testing, 2) had to adhere to the proportional hazards assumption as evidenced by a Schoenfeld individual test p-value >0.05, 3) obtain a p-value <0.05 in the multivariate model.

INFORMATION ON PATIENTS WITH MULTIPLE VARIANTS:

Patients with variants in other genes associated with cardiomyopathy: Three patients with multiple variants were included.

1. A male patient with a likely pathogenic variant in SCN5A p.(Gly213Asp) and TTN p.(Tyr35141*), presented with atrial fibrillation, LVEF 30%, and LVEDD 33 mm/m2 at the age of 39 years. An ICD was implanted, and he received multiple appropriate therapies during follow-up. At the latest consultation (age 57) he had a LVEF of 50% and an indexed LVEDD of 22 mm/m2. 2. The mother of the patient described above, shared both variants. She was identified through family screening (LVEF of ~45%) at age 71. The patient is currently alive at 82 years of age, has never been admitted to the hospital and did not have any interest in regular follow-up in a hospital setting. 3. A male patient with a likely pathogenic variant in DSG2 p.(Arg46Gln) and TTN p.(Met3833Cysfs*3), presented with aborted sudden cardiac death at age 54 and heart failure (LVEF = 25%, LVEDD = 29 mm/m2). The patient responded well to medical therapy (latest LVEF = 60%). Family screening identified two relatives with DCM who shared the TTNtv but not the DSG2 variant.

FIGURES

Supplementary Figure 1: Forest-plots depicting hazard ratios in Cox proportional hazards models for the three studied cardiac outcomes

Supplementary Figure 2: Forest-plots depicting univariable hazard ratios of 6 clinical variables in the three studied cardiac outcomes

Supplementary Figure 3: Outcomes according to site of truncation in titin

Legend: Cumulative incidences of four studied cardiac outcomes according to location of truncating variants in the different regions of titin. Patients with truncating variants in the Z-disc or I-band were younger at diagnosis of dilated cardiomyopathy, when compared with patients with truncating variants in tge A-band of titin (Panel A). No difference was observed in other studied outcomes (Panel B-D) according to the location of the genetic variants. Censoring is indicated by vertical lines. P-values are calculated from log-rank tests. If performing competing risk analysis in panel C and D the likelihood ratio p-value would be 0.795 for panel C and 0.889 for panel D.

Supplementary Figure 4: Outcomes in probands vs relatives

Legend: Cumulative incidences of four studied cardiac outcomes according to whether patients were probands (blue) or relatives (red). Probands tended to be younger at diagnosis of dilated cardiomyopathy (Panel A). In probands, the combined end-stage heart failure outcome and the ventricular arrhythmia outcome occurred at a significantly younger age than in relatives (Panel B-C). No difference was observed between probands and relatives in the timing of the atrial fibrillation outcome. Censoring is indicated by vertical lines. P-values are calculated from log-rank tests. If perfoming competing risk analysis in panel C and D the likelihood ratio p-value would be 0.021 for panel C and 0.577 for panel D.

SUPPLEMENTARY TABLES Supplementary Table 1. Identified truncating TTN variants. All variants are annotated in relation to the titin- metatranscript (NM_001267550.1)

Genetic variant Predicted protein change/ Variant Variant Protein Number of TTN ACMG MAF in Clin Var accession

genomic description of splice- position type domain DCM Exon criteria and gnomAD number and call

variants (GRCh37) subjects (n call or ExAC (checked March

of men) 2020)

c.1245+3A>G Broken Donor site 2:179659646 Splice Z-disk 1 (0) 7 PVS1, PP3; 3.3E-5 VCV000223251.1

VUS VUS

c.1594delG p.(Ala532Profs*19) 2:179656867 Frameshift Z-disk 2 (2) 10 PVS1, PM2, Absent Not previously

PP1; reported

Pathogenic

c.2137C>T p.(Arg713*) 2:179650808 Nonsense Z-disk 1 (0) 14 PVS1, PM2 Absent VCV000180003.2

; Likely VUS

pathogenic

c.5981C>A p.(Ser1994*) 2:179640610 Nonsense Near Z-disk 1 (1) 28 PVS1, PM2; Absent VCV000642471.1

Likely VUS

pathogenic

c.6527delT p.(Phe2176Serfs*30) 2:179639911 Frameshift I-band 1 (0) 29 PVS1, PM2; Absent Not previously

Likely reported

pathogenic

c.6679A>T p.(Lys2227*) 2:179639759 Nonsense I-band 1 (1) 29 PVS1, PM2; Absent Not previously

Likely reported

pathogenic

c.9220C>T p.(Arg3074*) 2:179632826 Nonsense I-band 2 (2) 39 PVS1, PP1; 5.8E-5 VCV000284118.1

VUS VUS

c.11497_11515delATGG p.(Met3833Cysfs*3) 2:179606463 Frameshift I-band 3 (3) 48 PVS1, PM2, Absent Not previously

GGGATGTGGCTACAC PP1; reported

Pathogenic

c.11952C>A p.(Tyr3984*) 2:179606008 Nonsense I-band 5 (2) 48 PVS1, PM2, Absent Not previously

PP1; reported

Pathogenic

c.12271C>T p.(Gln4091*) 2:179605689 Nonsense I-band 6 (3) 48 PVS1, PM2, Absent Not previously

PP1; reported

Pathogenic

c.31763-1G>A† Broken acceptor site 2:179554624 Splice I-band 2 (1) 122 PVS1, PP1, 0.0003 VCV000046855.4

PP3; VUS Conflicting: LP (1),

LB (1), VUS (6)

c.40626dupA p.(Pro13543Thrfs*13) 2:179505975 Frameshift I-band 1 (0) 220 PVS1, PM2, Absent VCV000446420.1

PP4; Likely pathogenic

Pathogenic

c.41330-7T>A† Predicted to not alter splicing 2:179500975 Splice I-band 2 (1) 226 PP1, BP4; 0.0003 VCV000137789.3 Conflicting: Benign VUS (3), LB (2), VUS (6)

c.42088delC p.(Gln14030Lysfs*18) 2:179499513 Frameshift I-band 1 (1) 229 PVS1, PM2, Absent Not previously

PP4; reported

Pathogenic

c.42667C>T p.(Gln14223*) 2:179498559 Nonsense I-band 2 (1) 231 PVS1, PM2, Absent Not previously

PP1; reported

Pathogenic

c.42823delT p.(Tyr14275Ilefs*12) 2:179498263 Frameshift I-band 1 (1) 232 PVS1, PM2, Absent Not previously

PP4; reported

Pathogenic

c.44272C>T p.(Arg14758*) 2:179494977 Nonsense I-band 1 (1) 240 PVS1; VUS 8.0E-6 VCV000202367.5

Conflicting: LP (4),

VUS (3)

c.47314C>T p.(Arg15772*) 2:179482764 Nonsense A-band 4 (4) 253 PVS1, PM2, Absent Not previously

PP1; reported

Pathogenic

c.48181G>T p.(Glu16061*) 2:179481337 Nonsense A-band 3 (2) 257 PVS1, PM2, Absent Not previously

PP1; reported

Pathogenic

c.48379_48382delTTTA p.(Phe16127Lysfs*20) 2:179480446 Frameshift A-band 1 (1) 258 PVS1, PM2, Absent VCV000423289.2

PP4; Pathogenic

Pathogenic

c.48877_48878insACTG p.(Thr16293Asnfs*5) 2:179479363 Frameshift A-band 1 (0) 260 PVS1, PM2; Absent Not previously reported Likely

pathogenic

c.52975C>T p.(Gln17659*) 2:179472539 Nonsense A-band 1 (1) 276 PVS1, PM2; Absent Not previously

Likely reported

pathogenic

c.53918delG p.(Gly17973Glufs*18) 2:179469986 Frameshift A-band 1 (0) 280 PVS1, PM2, Absent Not previously

PP1; reported

Pathogenic

c.54037delG p.(Ala18013Hisfs*4) 2:179469867 Frameshift A-band 1 (0) 280 PVS1, PP4; 4.2E-6 Not previously

VUS reported

c.54166C>T p.(Arg18056*) 2:179469738 Nonsense A-band 1 (0) 280 PVS1, PP5; 8.3E-6 VCV000202509.2

VUS Pathogenic

c.54777delA p.(Glu18259Aspfs*33) 2:179468636 Frameshift A-band 1 (0) 282 PVS1, PM2; Absent Not previously reported Likely

pathogenic

c.58117dupT p.(Cys19373Leufs*10) 2:179459104 Frameshift A-band 3 (1) 296 PVS1, PM2, Absent Not previously

PP1; reported

Pathogenic

c.59351_59352delCT p.(Pro19784Argfs*5) 2:179457380 Frameshift A-band 1 (1) 301 PVS1, PP5, 4.0E-6 VCV000264060.1

PP4; Likely pathogenic

Pathogenic

c.61769G>A p.(Trp20590*) 2:179454683 Nonsense A-band 1 (0) 304 PVS1, PM2, Absent Not previously

PP4; reported

Pathogenic

c.62231_62237del p.(Gln20744Argfs*10) 2:179454214 Frameshift A-band 2 (0) 304 PVS1, PM2, Absent Not previously AAACAAA PP1; reported

Pathogenic

c.64094-2A>G Broken acceptor site 2:179451536 Splice A-band 2 (1) 308 PVS1, PM2, Absent VCV000228302.2

PP1, PP3; Likely pathogenic

Pathogenic

c.72333dupT p.(Lys24112*) 2:179438525 Nonsense A-band 1 (1) 326 PVS1, PM2; Absent Not previously

Likely reported

pathogenic

c.72730_72785dup p.(Lys24264Ilefs*36) 2:179438129 Frameshift A-band 1 (1) 326 PVS1, PM2; Absent Not previously [56bp] Likely reported

pathogenic

c.74904dupT p.(Lys24969*) 2:179435954 Nonsense A-band 1 (1) 326 PVS1, PM2; Absent Not previously

Likely reported

pathogenic

c.79407delT p.(Glu26471Asnfs*27) 2:179431452 Nonsense A-band 1 (1) 326 PVS1, PM2, Absent Not previously

PP1; reported

Pathogenic

c.79844_79848del p.(Pro26615Glnfs*7) 2:179431015 Frameshift A-band 2 (2) 326 PVS1, PM2, Absent Not previously CATTC PP1; reported

Pathogenic

c.81726delA p.(Glu27242Aspfs*4) 2:179429133 Frameshift A-band 1 (1) 326 PVS1, PM2; Absent Not previously

Likely reported

pathogenic

c.85519dupA p.(Met28507*) 2:179425340 Nonsense A-band 4 (3) 326 PVS1, PM2, Absent VCV000660709.1

PP1; Likely pathogenic

Pathogenic

c.89197+2T>G Creates new exonic splicing 2:179418639 Splice A-band 15 (11) 333 PM2, PP1, Absent VCV000642902.1 silencer site PP3; Likely pathogenic

Pathogenic

c.90572_90573delTTins p.(Ile30191Argfs*23) 2:179417055 Frameshift A-band 1 (1) 336 PVS1, PM2 Absent Not previously

GAAGAAAAAAAAAAAAA Likely reported

AAAAG pathogenic

c.91715dupA p.(Asn30572Lysfs*15) 2:179414849 Frameshift A-band 6 (5) 337 PVS1, PM2, Absent VCV000223353.2

PP1; Conflicting: LP (1)

Pathogenic VUS (2)

c.93166C>T p.(Arg31056*) 2:179413187 Nonsense A-band 1 (1) 339 PVS1, PP5; 2.5E-5 VCV000223326.4 Pathogenic VUS

c.95341C>T p.(Arg31781*) 2:179410622 Nonsense A-band 1 (1) 343 PVS1, PM2, Absent Not previously reported PP4;

Pathogenic

c.98299_98300delAG p.(Arg32767Glyfs*2) 2:179404492 Frameshift A-band 1 (0) 352 PVS1, PP4, 4.0E-6 VCV000047599.4

PP5; Pathogenic

Pathogenic

c.99994G>T p.(Glu33332*) 2:179401842 Nonsense A-band 2 (1) 356 PVS1, PM2, Absent Not previously reported PP1;

Pathogenic

c.100825C>T p.(Arg33609*) 2:179400517 Nonsense M-band 1 (1) 358 PVS1, PP5; 4.0E-6 VCV000373074.3

VUS Pathogenic

c.101188delG p.(Ala33730Glnfs*10) 2:179400153 Frameshift M-band 1 (1) 358 PVS1, PM2; Absent Not previously

Likely reported

pathogenic

c.101809delC p.(His33937Thrfs*22) 2:179399533 Frameshift M-band 2 (2) 358 PVS1, PM2, Absent Not previously

PP1, reported

Pathogenic

c.101932_101935dupG p.(Pro33979Argfs*8) 2:179399410 Frameshift M-band 3 (2) 358 PVS1, PM2, Absent Not previously CCC PP1; reported

Pathogenic

c.102290delT p.(Leu34097Argfs*27) 2:179399051 Frameshift M-band 3 (1) 358 PVS1, PM2, Absent Not previously

PP1; reported

Pathogenic

c.102510G>A p.(Trp34170*) 2:179398832 Nonsense M-band 6 (4) 358 PVS1, PM2, Absent Not previously

PP1; reported

Pathogenic

c.102800dupA p.(Thr34268Aspfs*6) 2:179398542 Frameshift M-band 4 (3) 358 PVS1, PM2, Absent Not previously

PP1; reported

Pathogenic

c.104835_104836delAC p.(Leu34946Glufs*22) 2:179396506 Frameshift M-band 1 (1) 358 PVS1, PM2; Absent Not previously

Likely reported

pathogenic

c.105423C>G p.(Tyr35141*) 2:179395919 Nonsense M-band 2 (1) 358 PVS1, PP1, 8.0E-6 VCV000593024.2

PP5; Likely pathogenic

Pathogenic

Abbreviations: ACMG: American College of Medical Genetics and Genomics; DCM: dilated cardiomyopathy; ExAC: Exome Aggregation Consortium; gnomAD: The Genome Aggregation

Database; GRCH37: Genome research council build 37; MAF: minor allele frequency.

ACMG criteria reported: PVS1: predicted null variant in gene where loss of function is a known mechanism of disease; PM2: Variant absent in population databases; PP1: Cosegregated with

disease in multiple affected family members; PP3:Multiple lines of computational evidence support a deleterious effect on the gene/gene product; PP4: Patient’s phenotype or family history

highly specific for gene; PP5: Reputable source reported variant as pathogenic; BP4: Multiple lines of computational evidence suggest no impact on gene/gene product.

†: These two variants were shared by two family members with DCM. c.31763-1G>A is a splice acceptor variant which affects an exon with a PSI of 0.78 while c.41330-7T>A is an intronic

variant which has not been tested in functional studies, but it is predicted to not affect splicing by in silico algorithms.

Supplementary Table 2: Association of clinical variables with combined end-point (implantation of LVAD, heart transplantation or death) in univariable and multivariable cox regression analysis Parameter Univariate 95% CI for p-value Multivariable HR

HR univariate HR [CI 95%]

Demographic data

Age at DCM diagnosisa*† 1.07 [0.94-1.22] 0.369 1.07 [0.89-1.28]

Male sex* 2.44 [0.93-6.38] 0.069 NA

Family history of heart failure 0.16 [0.08-0.34] 1.7E-6 NA

Proband*† 3.07 [1.17-8.02] 0.022 2.14 [0.65-7.07]

Medical history

History of diabetes* 2.24 [0.85-5.92] 0.102 NA

History of hypertension 1.44 [0.55-3.80] 0.461 NA

Atrial fibrillation 1.10 [0.54-2.26] 0.792 NA

Ventricular arrhythmias 1.03 [0.46-2.32] 0.947 NA

AV-block* 4.10 [1.59-10.57] 0.004 NA

QRS-duration > 120 ms* 2.37 [1.08-5.24] 0.032 NA

Echocardiography

LVEF at onseta*† 0.85 [0.72-1.02] 0.088 0.88 [0.71-1.09]

Indexed LVEDD at onseta 1.05 [0.97-1.14] 0.2226 NA

Occurrence of LVRR*†‡ 0.06 [0.02-0.14] 2.5E-10 0.05 [0.02-0.14]

Medial therapy

Betablocker use* 0.25 [0.12-0.52] <0.001 NA

Diuretics use* 7.15 [2.16-23.63] 0.001 NA

ACE-I/ARB use 1.14 [0.39-3.31] 0.809 NA

Aldosterone-antagonist use 1.18 [0.76-2.45] 0.657 NA

Genetic findings

Amino acid position in full-length titina 0.99 [0.85-1.16] 0.904 NA

TTNtv in 1st quartile of titin (reference) 1.0 NA NA NA

TTNtv in 2nd quartile of titin 1.31 [0.40-4.29] 0.659 NA

TTNtv in 3rd quartile of titin 1.70 [0.52-5.58] 0.380 NA

TTNtv in 4th quartile of titin 0.96 [0.34-2.69] 0.935 NA

TTNtv in Z-disk/I-band 0.97 [0.40-2.39] 0.954 NA

TTNtv in A-band (reference) 1.0 NA NA NA

TTNtv in M-band 1.27 [0.53-3.00] 0.593 NA

Abbreviations: ACE-I: ace-inhibitor; ARB: Angiotensin receptor blocker; AV: atrioventricular; DCM: Dilated

cardiomyopathy; LVEDD: left ventricular end-diastolic dimensions; LVEF: Left ventricular ejection fraction; LVRR:

Left ventricular reverse remodeling; NA: not applicable; TTNtv: truncating titin variant;

Symbols: *Variable investigated in multivariate analysis. † Variables included in the multivariate model. ‡Variables

included in the multivariate model with p < 0.05 which are .a Continuous variables: Increment of 5 years for age, 5

%-points for LVEF, 1 mm for LVEDD, 1 mm/m2 for indexed LVEDD and 5000 amino acids for amino acid position in

full-length titin.

Supplementary Table 3: Association of clinical variables with ventricular arrhythmia in univariable and multivariable cox regression analysis. Parameter Univariate 95% CI for p-value Multivariable

HR univariate HR HR [CI 95%]

Demographic data

Age at DCM diagnosisa *† 1.04 [0.90-1.20] 0.611 1.03 [0.86-1.25]

Male sex*†‡ 2.75 [0.95-7.97] 0.063 3.05 [1.00-9.32]

Family history of heart failure 0.32 [0.14-0.76] 0.009 NA

Proband*†‡ 3.99 [1.37-11.60] 0.011 5.62 [1.61-19.61]

Medical history

History of diabetes 0.43 [0.06-3.15] 0.403 NA

History of hypertension 0.56 [0.13-2.35] 0.423 NA

Atrial fibrillation* 1.97 [0.90-4.35] 0.092 NA

AV-block 1.40 [0.60-3.41] 0.417 NA

QRS-block*†‡ 2.38 [1.04-5.44] 0.040 2.44 [1.05-5.71]

Echocardiography

LVEF at onseta 0.86 [0.72-1.04] 0.113 NA

Indexed LVEDD at onseta 1.02 [0.94-1.12] 0.624 NA

Occurrence of LVRR 0.67 [0.26-1.72] 0.405 NA

Medial therapy

Betablocker use 1.55 [0.53-4.54] 0.427 NA

Diuretics use 1.70 [0.76-3.83] 0.197 NA

ACE-I/ARB use 0.57 [0.23-1.44] 0.233 NA

Aldosterone-antagonist use 1.91 [0.87-4.18] 0.105 NA

Genetic findings

Amino acid position in full-length 0.89 [0.76-1.05] 0.157 NA

titina

TTNtv in 1st quartile of titin 1.0 NA NA NA

(reference)

TTNtv in 2nd quartile of titin 0.22 [0.05-1.01] 0.052 NA

TTNtv in 3rd quartile of titin 0.59 [0.18-1.92] 0.380 NA

TTNtv in 4th quartile of titin 0.46 [0.19-1.10] 0.082 NA

TTNtv in Z-disk/I-band 1.40 [0.60-3.28] 0.439 NA

TTNtv in A-band (reference) 1.0 NA NA NA

TTNtv in M-band 0.69 [0.22-2.14] 0.521 NA

Abbreviations: ACE-I: ace-inhibitor; ARB: Angiotensin receptor blocker; AV: atrioventricular; DCM: Dilated

cardiomyopathy; LVEDD: left ventricular end-diastolic dimensions; LVEF: Left ventricular ejection fraction;

LVRR: Left ventricular reverse remodeling; NA: not applicable; TTNtv: truncating titin variant;

Symbols: *Variable investigated in multivariate analysis. † Variables included in the multivariate model.

‡Variables included in the multivariate model with p < 0.05 which are .a Continuous variables: Increment of 5

years for age, 5 %-points for LVEF, 1 mm for LVEDD, 1 mm/m2 for indexed LVEDD and 5000 amino acids for

amino acid position in full-length titin.

Supplementary Table 4: Association of clinical variables with atrial fibrillation in univariable cox regression analysis.

Parameter Univariate 95% CI for p- Multivariable HR [CI

HR univariate HR value 95%]

Demographic data

Age at DCM diagnosisa* 1.16 [1.05-1.29] 0.004 1.12 [0.99-1.27]

Male sex 1.37 [0.75-2.52] 0.306 NA

Family history of heart failure 0.69 [0.33-1.44] 0.323 NA

Proband*† 1.16 [0.65-2.07] 0.614 0.97 [0.48-1.96]

Medical history

History of diabetes 0.73 [0.23-2.38] 0.609 NA

History of hypertension 1.67 [0.81-3.46] 0.167 NA

Ventricular arrhythmias* 1.73 [0.95-3.14] 0.071 NA

AV-block*†‡ 2.76 [1.38-5.49] 0.004 2.51 [1.22-5.16]

QRS-duration > 120 ms 1.53 [0.79-2.96] 0.206 NA

Echocardiography

LVEF at onseta 0.98 [0.87-1.10] 0.721 NA

Indexed LVEDD at onseta 0.97 [0.91-1.04] 0.352 NA

Occurrence of LVRR 1.05 [0.50-2.21] 0.905 NA

Medical therapy

Betablocker use 1.31 [0.63-2.72] 0.467 NA

Diuretics use 1.50 [0.85-2.65] 0.166 NA

ACE-I/ARB use 0.91 [0.42-1.94] 0.801 NA

Aldosterone-antagonist use 1.43 [0.82-2.49] 0.211 NA

Genetic findings

Amino acid position in full- 0.94 [0.84-1.07] 0.356 NA

length titina

TTNtv in 1st quartile of titin 1.00 NA NA NA

(reference)

TTNtv in 2nd quartile of titin 0.59 [0.23-1.53] 0.276 NA

TTNtv in 3rd quartile of titin 0.98 [0.41-2.37] 0.964 NA

TTNtv in 4th quartile of titin 0.62 [0.30-1.29] 0.203 NA

TTNtv in Z-disk/I-band 1.57 [0.82-3.00] 0.171 NA

TTNtv in A-band (reference) 1.00 NA NA NA

TTNtv in M-band 1.07 [0.52-2.20] 0.858 NA

Abbreviations: ACE-I: ace-inhibitor; ARB: Angiotensin receptor blocker; AV: atrioventricular; DCM:

Dilated cardiomyopathy; LVEDD: left ventricular end-diastolic dimensions; LVEF: Left ventricular

ejection fraction; LVRR: Left ventricular reverse remodeling; NA: not applicable; TTNtv: truncating titin

variant;

Symbols: *Variable investigated in multivariate analysis. † Variables included in the multivariate model.

‡Variables included in the multivariate model with p < 0.05 which are .a Continuous variables:

Increment of 5 years for age, 5 %-points for LVEF, 1 mm for LVEDD, 1 mm/m2 for indexed LVEDD and

5000 amino acids for amino acid position in full-length titin.

Vissing CR, et al. J Med Genet 2020;0:1–10. doi: 10.1136/jmedgenet-2020-107178