GEO combined with quantitative protein trait loci identify causative proteins in hypertrophic cardiomyopathy

Abstract Aims Hypertrophic cardiomyopathy (HCM) is a rare genetic heart disease characterized by a limited patient population and scarce research and treatment resources. This study aimed to identify HCM‐associated proteins by integrating cardiac tissue data from the Gene Expression Omnibus (GEO) database with the latest protein quantitative trait locus (pQTL) dataset. Methods and results We analysed data from the GEO database. The GSE36961 dataset included 106 HCM samples and 39 healthy controls. The GSE180313 dataset included 13 HCM samples and 7 healthy controls. pQTL data were obtained from the plasma of 54 000 UK Biobank participants, covering 1463 proteins. HCM genome‐wide association study (GWAS) data were sourced from the FinnGen study, which included 1125 HCM cases and 411 056 controls. We analysed the GEO dataset of cardiac tissue from HCM patients to identify differentially expressed genes (DEGs). These DEGs were compared with pQTL data to identify protein phenotypes suitable for Mendelian randomization (MR) analysis. A two‐sample MR analysis was performed to assess the causal association between these protein phenotypes and HCM. The robustness of the study results was further assessed through sensitivity analysis of heterogeneity and horizontal pleiotropy tests. Two proteins were identified as causally associated with HCM risk: carbonic anhydrase 3 (CA3) [inverse variance weighted (IVW): odds ratio (OR) = 1.292, 95% confidence interval (CI) = 1.021–1.636, P = 0.033] and serpin family E member 1 (SERPINE1) [IVW: OR = 1.313, 95% CI = 1.063–1.621, P = 0.011]. Both proteins were associated with increased HCM risk, with no significant heterogeneity ( P > 0.05) or evidence of horizontal pleiotropy ( P > 0.05). Conclusions CA3 and SERPINE1 proteins may exert causal effects on HCM and may serve as characteristic markers and therapeutic targets for this condition.