Exploring the genetic architecture of multimorbidity and its impact on long COVID risk

Abstract

Multimorbidity, the co-occurrence of multiple long-term conditions, represents a major challenge for ageing populations, yet its genetic architecture and relationship to long COVID remain unclear, despite shared epidemiological risk factors. We analysed multimorbidity patterns in 86,756 White British UK Biobank participants aged [≥]65 years, identifying six clusters spanning neurodegenerative, cardiac, gastrointestinal, musculoskeletal, vascular, and cancer & eye disease domains. Genome-wide association studies and post-GWAS analyses revealed significant loci in five clusters, including APOE, LPA, and CDKN2B-AS1, with patterns of genetic correlation consistent with known disease relationships. Notably, a shared variant within the APOE-APOC1 locus showed opposite effect directions for the musculoskeletal and vascular clusters, consistent with their negative genetic correlation. Investigating the multimorbidity-long COVID relationship via genetic correlation and Mendelian randomisation revealed no evidence of significant shared genetic architecture or causal effects. These findings indicate that multimorbidity clusters represent biologically structured, partly heritable phenotypes, whereas genetic overlap with long COVID appears limited.