A GWAS META-ANALYSIS OF PSYCHOTIC DEPRESSION LEVERAGING HEALTH-RECORD, PRESCRIPTION AND SELF-REPORT DATA

Abstract

Depression with psychotic features is poorly understood, and the extent to which its biology is shared with other psychosis disorders is unclear. There is evidence for a genetic basis for MDD, but there has not been a major GWAS of the phenotype. We are conducting the first GWAS meta-analysis of psychotic depression. Given that this condition may be under-diagnosed, we investigate three case definitions: health record–based diagnosis of psychotic depression (> 5,000 European cases), co-prescription of antipsychotics and antidepressants alongside depression diagnosis (> 3,500 European cases,), and self-report psychotic symptoms alongside a depression diagnosis (> 5,000 European cases) across individuals of European and African ancestries. In the UK Biobank, the prescription data required extensive quality control to identify prescriptions likely reflecting psychiatric indications. The three definitions show low overlap in the UK Biobank, suggestive of substantial phenotypic heterogeneity. This limited concordance highlights the challenge of capturing psychotic depression and underscores the need to evaluate how these definitions influence genetic signal. We compare the genetic associations across case definitions and assess whether clinical heterogeneity in depression reflects underlying biological heterogeneity.