Polygenic risk score for dilated cardiomyopathy and the relationship to myocardial reserve

Background Dilated cardiomyopathy (DCM) is characterized by left ventricular (LV) enlargement and reduced ejection fraction (LVEF), and recent studies have linked higher polygenic risk score (PRS) for DCM to such phenotypic characteristics. Myocardial reserve, defined as the heart’s ability to increase cardiac output during increased demand, is expected to have a stronger link to DCM PRS than measures derived solely from rest, but the relationship between DCM PRS and myocardial reserve remains unexplored. LV end-diastolic volume (EDV) indexed to absolute peak oxygen uptake (VO2peak, L/min) provides a physiologically grounded ratio reflecting myocardial reserve by combining LV size and VO2peak. Thus, the LVEDV/ VO2peak indicates whether the LV can scale up cardiac output in line with expected by its size and provides a scalable alternative for assessing myocardial reserve. Purpose To investigate the association between a PRS for DCM and LVEDV/VO2peak in a large population sample and compare this to resting echocardiographic measures of LV size and function. Methods The PRS was derived from an independent genome-wide association study based on >9k DCM cases and 1.1M genetic variants and applied to a sample of participants from a large health study undergoing resting echocardiography and a cardiopulmonary exercise test for directly measuring VO2peak. The associations between DCM PRS with LVEDV/VO2peak, measures of cardiac size and function, and VO2peak were tested using general linear models both adjusted for age, sex, principal components and genomic batch. We calculated partial explained variance (R2) for each predictor in models with the PRS as outcome by comparing models with and without each predictor. Sensitivity analyses were performed excluding clinical DCM (n=8). Results A total of 1838 participants (50% female) with mean age 59 (SD 12) years were included in primary analyses. There were significant associations between the DCM PRS and all measures of LV size and function (end-diastolic internal dimension (LVIDd) LVEDV, LVEF, LV global longitudinal strain (GLS) and LVEDV/VO2peak, Table 1). The strongest association with the highest standardized beta and explained partial variance was found for LVEDV/VO2peak (Figure 1). There were no significant associations for VO2peak. Non-linear models using restricted cubic splines did not improve the models, and sensitivity analyses showed unchanged results. The PRS was significantly associated with clinical DCM in the health study population where the 1838 participants were sampled from (535 cases/86,342 controls), with an odds ratio for DCM per SD of 1.42 (p=6.1×10). Conclusion This study indicates that a PRS for DCM has a stronger association to myocardial reserve as measured by LVEDV/VO2peak than to resting measures of LV size and function alone. These findings also suggest that a larger LV due to a high PRS for DCM is not likely to yield benefits for athletic performance.