Insights on Titin cardiomyopathy: from truncating variants site to phenotypic expression and outcomes

Abstract

Abstract Background Titin truncating variants (TTNtv) represent the most prevalent genotype underlying dilated cardiomyopathy (DCM). Recent molecular studies provided initial evidence about distinct pathogenic mechanisms of different protein truncating sites (A-band vs non-A-band), possibly related to the presence or absence of truncated peptides. Yet, clinical studies on TTNtv-DCM failed to demonstrate any relevant clinical difference according to the band affected. We hypothesized this lack of distinction could be due to the selection bias of including only DCM patients. Thus, we studied the clinical phenotypes and outcomes of our entire population of TTNtv carriers, regardless of phenotypic expression at onset. Methods We retrospectively analysed a population of pathogenic TTNtv carriers, enrolled in our referral centre. TTNtv were classified according to their site (Z disk, I band, A band, A-M junction, M band) and further grouped for analysis in A-band and non-A band. Kaplan-Meier curves were estimated for the primary composite endpoint of all-cause mortality and heart transplant (HTx); while cumulative incidence functions were estimated for the two secondary endpoints (i) heart failure-related death, heart transplantation, or destination left ventricular assist device implantation (LVAD) and (ii) first major ventricular arrhythmia (MVA, defined as sudden death or life-threatening ventricular arrhythmia), compared by Gray’s test. Analyses were then repeated adding an independent cohort of TTVtv carriers from another tertiary centre. Results 161 TTNtv carriers were included in the study (79% A-band, 21% non-A band), 63% were probands. The most prevalent phenotype was DCM (92%). Mean age at diagnosis was 43±16 years and 68% were males. There was no difference between groups in the primary composite outcome nor in the secondary composite endpoint of heart failure-related death, HTx or destination LVAD. Instead, MVA were significantly more frequent in the non-A-band group, in which 45% (n=15) of TTNtv carriers experienced MVA as compared to 17% (n=20) of subjects in the A-band group (p=0.001). Interestingly, we observed a tendency in MVA occurring at a higher mean left ventricular ejection fraction (LVEF) in the non-A-band group as compared to the A-band group (42±13% vs 35±9%, respectively). Furthermore, there was no significant association between MVA and LVEF at the event in the non-A group, while this association was significant in A-band (p=0.03). When an external population of 22 TTNtv carriers (95% DCM, 77% A-band) was added for analysis, this secondary outcome was confirmed (p<0.001). Conclusions Among TTNtv carriers referred to two tertiary centres, those with a mutation in the non-A band seem characterized by higher arrhythmic risk, occurring independently from EF. This novel finding highlights the importance of considering the site of TTN truncating variants to guide the most appropriate clinical management for each patient.Fig 1.Secondary Outcome, arrhythmic risk