Evaluation of polygenic scores for hypertrophic cardiomyopathy in the general population and across clinical settings

Abstract

Hypertrophic cardiomyopathy (HCM) is an important cause of morbidity and mortality, with pathogenic variants found in about a third of cases. Large-scale genome-wide association studies (GWAS) demonstrate that common genetic variation contributes to HCM risk. Here we derive polygenic scores (PGS) from HCM GWAS and genetically correlated traits and test their performance in the UK Biobank, 100,000 Genomes Project, and clinical cohorts. We show that higher PGS significantly increases the risk of HCM in the general population, particularly among pathogenic variant carriers, where HCM penetrance differs 10-fold between those in the highest and lowest PGS quintiles. Among relatives of HCM probands, PGS stratifies risks of developing HCM and adverse outcomes. Finally, among HCM cases, PGS strongly predicts the risk of adverse outcomes and death. These findings support the broad utility of PGS across clinical settings, enabling tailored screening and surveillance and stratification of risk of adverse outcomes.

Fig. 1. Study overview.

Bayesian genome-wide PGS were generated from a published European-ancestry HCM GWAS meta-analysis of seven case–control studies (comprising 5,900 cases and 68,359 controls; PGS_GWAS), and MTAG (analyzing HCM with three genetically correlated quantitative traits measured using CMR imaging in 36,083 European ancestry UKB participants—LV concentricity (LVconc), LEVSV and left ventricular circumferential strain (strain_circ); PGS_MTAG). The value of PGS to support clinical decision-making was evaluated across three key settings: in the general population (including among carriers of pathogenic rare variants in HCM-causing genes (sarcomere-positive) that might be returned as secondary findings), in relatives of HCM probands currently recommended to undergo cascade screening and surveillance, and in confirmed HCM cases under longitudinal follow-up. The figure is created with BioRender.com. SARC-PLP, pathogenic or likely pathogenic variant in sarcomeric HCM genes.

Fig. 2. HCM PGS is associated with HCM disease status in the UKB.

To validate the PGS, we analyzed associations with PGS in the UKB population. a, PGS_MTAG distribution in 374,845 UKB participants with and without HCM, demonstrating higher PGS in those with HCM. b, Cumulative curve of HCM cases ranked across PGS centiles. For example, approximately 75% of HCM cases have a PGS above the population 50th centile. Dashed lines represent mean, ±1 PGS s.d. and ±2 PGS s.d. Shaded line indicates 95% CI surrounding the cumulative estimate. c, Manhattan plot of HCM PGS phenome-wide association study in UKB, showing associations with cardiovascular and metabolic phenotypes. ICD-9 and ICD-10 diagnostic codes are mapped to Phecode Map (v1.2). Mapped phenotypes exceeding phenome-wide significance threshold (P = 2.7 × 10⁻⁵, red line) are labeled. Blue line indicates nominal significance (P < 0.05). Direction of triangle indicates the direction of effect of the PGS association. d, HCM prevalence and risk in UKB across the spectrum of PGS, demonstrating significantly higher HCM prevalence in individuals with the highest PGS (top centile, n = 3,394), compared with the median (n = 68,587) and lowest groups (bottom centile, n = 3,431). Effect estimates generated using logistic regression adjusting for age, age², sex and top ten genetic principal components (PCs), with unadjusted two-sided P value. Data are presented as effect estimates with 95% CI. e, Cumulative hazards for lifetime diagnosis of HCM stratified by high (highest centile, red) and median (middle quintile, orange) PGS risk in UKB. HR calculated using Cox proportional hazards model, adjusted for age, age², sex and first ten genetic PCs, with two-sided P value.

Fig. 3. PGS modulates HCM penetrance in carriers of rare pathogenic variants in HCM-associated genes.

a–c, UKB represents a broadly unselected population, as participants were not recruited based on genotype or phenotype. d–f, 100,000 Genomes Project (GeL) comprises a mix of participants recruited based on cardiomyopathy and participants recruited with other rare diseases, cancer or as relatives of patients with a rare disease. a,d, The PGS distribution is shown in 640 sarcomere-positive UKB participants (a) and 599 GeL participants (d) with and without HCM, validating that PGS is higher in cases than controls. b,e, Among sarcomere-positive individuals, the highest PGS quintile (UKB, n = 136; GeL, n = 116) was associated with increased HCM diagnosis compared with median (UKB, n = 111; GeL, n = 116) and lowest quintiles (UKB, n = 133; GeL, n = 118). Effect estimates generated using logistic regression adjusting for age, age², sex and top ten genetic PCs, with unadjusted two-sided P value. Data are presented as effect estimates with 95% CI. c,f, The time to HCM diagnosis in highest, median and lowest quintiles, shows that those with higher PGS are at increased risk of HCM, and develop disease earlier, which is important for lifetime burden of disease morbidities. HR calculated using Cox proportional hazards model, adjusted for age, age², sex and first ten genetic PCs, with two-sided P value.

Fig. 4. PGS associate with HCM risk and adverse outcomes in relatives of HCM cases.

To evaluate applications of PGS in families undergoing screening and surveillance for HCM, we studied the relatives of HCM cases in two cohorts, GeL and EMC cohort. a, OR for HCM among relatives of HCM probands in the two cohorts (GeL, n = 288; EMC, n = 214), stratified by PGS. b, Violin and box and whisker plot of maxLVWT in sarcomere-positive relatives stratified by highest (n = 40) and lowest (n = 38) PGS_EMC quintiles. Box plot indicate median and interquartile range, whiskers denote 1.5× the interquartile range, outliers shown separately, and the edges of violin plots indicate minimum and maximum values. Dashed line indicates a 13-mm cutoff used for guideline diagnosis of HCM in relatives of individuals with HCM. c, Cumulative major adverse cardiovascular events (MACE) among 214 sarcomere-positive relatives of HCM index patients stratified by PGS_EMC above or below the median. MACE was defined as a composite of septal reduction therapy, cardiac transplantation, aborted cardiac arrest, appropriate defibrillator shock or sudden cardiac death. To avoid inflation of PGS performance resulting from sample overlap, PGS were rederived from GWAS leaving out the cohort that the PGS was being evaluated in (GeL–PGS_GeL, EMC–PGS_EMC). HR calculated using Cox proportional hazards model, adjusted for sex, first four genetic PCs, and genetic relatedness matrix, with two-sided P value.

Fig. 5. PGS stratifies the risk of death and adverse outcomes in individuals with HCM.

a–c, Cumulative all-cause mortality and adverse HCM outcomes after HCM diagnosis in 382 HCM cases from the UKB (a,b) and cumulative all-cause mortality in 683 HCM cases from GeL (c), stratified by PGS in the highest and lowest quintiles. Adverse HCM outcomes include death, stroke, cardiac arrest, implantable cardioverter-defibrillator implantation, septal reduction therapy (alcohol–septal ablation or surgical myectomy), LVAD implantation or cardiac transplantation. HR calculated using Cox proportional hazards model, adjusted for age, age², sex and first ten genetic PCs, with two-sided P value.