Genotype is associated with left ventricular reverse remodelling and early events in recent-onset dilated cardiomyopathy

Abstract

Aims: Recent-onset dilated cardiomyopathy (RODCM) is characterized by heterogeneous aetiology and diverse clinical outcomes, with scarce data on genotype-phenotype correlates. Our aim was to correlate individual RODCM genotypes with left ventricular reverse remodelling (LVRR) and clinical outcomes.

Methods and results: In this prospective study, a total of 386 Czech RODCM patients with symptom duration ≤6 months underwent genetic counselling and whole-exome sequencing (WES). The presence of pathogenic (class 5) or likely pathogenic (class 4) variants in a set of 72 cardiomyopathy-related genes was correlated with the occurrence of all-cause death, heart transplantation, or implantation of a ventricular assist device (primary outcome) and/or ventricular arrhythmia event (secondary outcome). LVRR was defined as an improvement of left ventricular ejection fraction to >50% or ≥10% absolute increase, with a left ventricular end-diastolic diameter ≤33 mm/m² or ≥10% relative decrease. Median follow-up was 41 months. RODCM was familial in 98 (25%) individuals. Class 4-5 variants of interest (VOIs) were identified in 125 (32%) cases, with 69 (18%) having a single titin-truncating variant (TTNtv) and 56 (14%) having non-titin (non-TTN) VOIs. The presence of class 4-5 non-TTN VOIs, but not of TTNtv, heralded a lower probability of 12-month LVRR and proved to be an independent baseline predictor both of the primary and the secondary outcome. The negative result of genetic testing was a strong protective baseline variable against occurrence of life-threatening ventricular arrhythmias. Detection of class 4-5 VOIs in genes coding nuclear envelope proteins was another independent predictor of both study outcomes at baseline and also of life-threatening ventricular arrhythmias after 12 months. Class 4-5 VOIs of genes coding cytoskeleton were associated with an increased risk of life-threatening ventricular arrhythmias after baseline assessment. A positive family history of dilated cardiomyopathy alone only related to a lower probability of LVRR at 12 months and at the final follow-up.

Conclusions: RODCM patients harbouring class 4-5 non-TTN VOIs are at higher risk of progressive heart failure and life-threatening ventricular arrhythmias. Genotyping may improve their early risk stratification at baseline assessment.

Keywords: Genetics; Left ventricular reverse remodelling; Prognosis; Recent‐onset dilated cardiomyopathy; Whole‐exome sequencing.

Figure 1

Timeline illustrates prediction of the study outcomes both from baseline and 12 months of follow‐up.

Figure 2

(A) Spectrum of affected genes among class 4–5 non‐titin (non‐TTN) variants detected in the study group. Their functional annotation is described in panel (B).

Figure 3

Diagrams in the upper part of the figure list variables predicting the primary and secondary outcome from baseline. The white subsets inside each of both lists summarize variables, which remained independent in most multivariate models. The Kaplan–Meier curves in the lower part show an increased occurrence of both outcomes in carriers of class 4–5 non‐titin VOIs compared to genotype‐negative individuals. Abbreviations are shown in Table 1 .