Arrhythmic Risk Stratification of Carriers of Filamin C Truncating Variants

, Marta Gigli1,2, Davide Stolfo1,3

  • 1Cardiothoracovascular Department, Azienda Sanitaria-Universitaria Giuliano Isontina, Trieste, Italy.

JAMA Cardiology
|February 12, 2025
PubMed

Insights

Filamin C truncating variants (FLNCtv) increase risk for sudden cardiac death (SCD) and major ventricular arrhythmias (MVA). A new 5-variable model helps predict risk in FLNCtv carriers, aiding decisions on prophylactic ICD implantation.

Area of Science:

  • Cardiology
  • Genetics
  • Clinical Research

Background:

  • Filamin C truncating variants (FLNCtv) are a rare cause of cardiomyopathy.
  • FLNCtv carriers face a high risk of life-threatening ventricular arrhythmias and sudden cardiac death (SCD).
  • Reliable risk predictors for stratifying FLNCtv carriers are currently lacking.

Purpose of the Study:

  • To identify factors that predict SCD and major ventricular arrhythmias (MVA) in individuals with FLNCtv.
  • To develop a risk prediction model for SCD/MVA in FLNCtv carriers.

Main Methods:

  • International, multicenter, retrospective cohort study (February 2023 - June 2024) involving 308 FLNCtv carriers from 19 centers.
  • Primary outcome: composite of SCD and MVA (including aborted SCD, sustained ventricular tachycardia, appropriate ICD interventions).
  • Multivariable analysis used to derive a 5-variable predictive model.

Main Results:

  • During a median follow-up of 34 months, 19% of individuals experienced SCD/MVA.
  • A predictive model incorporating age, male sex, syncope, nonsustained ventricular tachycardia, and LVEF demonstrated good accuracy (AUC 0.76-0.78).
  • Left ventricular ejection fraction (LVEF) showed a non-linear association with risk, with lower risk above 58%.

Conclusions:

  • The risk of SCD/MVA is high in phenotype-positive FLNCtv carriers.
  • A 5-variable predictive model can aid clinicians in risk stratification and decisions regarding prophylactic ICD implantation.
  • External validation of the predictive model is recommended.
Abstract