Abstract 4363833: Impact of Damaging Genetic Variants on Right Ventricular Function in Tetralogy of Fallot

Background: Right ventricular (RV) dysfunction is an important determinant of outcomes in many forms of CHD. However, clinical and imaging biomarkers explain only 25% of the variability in long-term outcomes in patients with repaired tetralogy of Fallot (rTOF). The contribution of genetic factors to RV dysfunction in rTOF is a knowledge gap. Hypothesis: Genetic factors contribute to variation in RV function in patients with rTOF. Methods: We studied the relationship of rare damaging genetic variants (RDV) and RV function, assessed by cardiac MRI in a cohort of 223 rTOF patients with genome or exome sequencing, recruited under the auspices of the Pediatric Cardiac Genomics Consortium at Boston Children’s Hospital. We characterized demographics, genotypes, clinical and postoperative variables, and cardiac MRI measurements. Rare variants (gnomAD allele frequency <10 -4 ) were considered damaging if predicted to be loss-of-function (nonsense, frameshift, or read-through), missense (REVEL score >0.5), or splice altering (SpliceAI delta score >0.8). Association was determined by comparing the proportion of participants with RV dysfunction (RV ejection fraction <45%) and RDV in 70 genes associated with pediatric onset cardiomyopathy, as reported in ClinVar. We performed multivariable logistic regression to adjust for independent predictors of RV dysfunction. Results: Patients with rTOF and RV dysfunction were older at MRI (15.3 years vs. 12.6 years, p=0.01), and more likely to be male (71% vs. 49%, p=0.002), have a history of arrhythmia (26% vs. 11%, p=0.008), or repaired prior to 1985 (17% vs 6%, p = 0.01). 22q11 deletion syndrome was not associated with RV dysfunction. Heterozygous RDV in genes associated with pediatric cardiomyopathy were more common in patients with RV dysfunction (11% vs. 1%; OR 8.63, p =0.007). In a multivariable model (C statistic = 0.71), presence of a pediatric cardiomyopathy variant remained associated with RV dysfunction (OR 1.44, p=0.01). Conclusions: RDV in genes associated with pediatric cardiomyopathy are associated with RV dysfunction in patients with rTOF. These genes would not be expected to be causal for CHD but instead modify myocardial function. While future larger multicenter studies should validate these findings, these results suggest that pediatric cardiomyopathy variants may affect outcomes, improve risk-stratification, and provide more precise personalized therapies for CHD.