UNRAVELING SEX DIFFERENCES IN THE GENETIC BASIS OF SCHIZOPHRENIA

Schizophrenia (SCZ) shows clear sex differences in age at onset, symptoms, disease course, and treatment response, yet the genetic basis of these differences remains poorly understood. This study applied advanced genomic analyses to investigate sex-specific genetic contributions to SCZ risk.
Using the latest sex-stratified SCZ-GWAS summary statistics, a CIBERSAM sample of 1,826 SCZ patients (33.2% female) and 1,372 healthy controls (45.8% female), we conducted SNP-based heritability (h²), global and local genetic correlation, gene-based analyses, transcriptome-wide association studies (TWAS), and polygenic risk score (PRS) analyses.
SNP-based liability h² was higher in males than females (0.23±0.01 vs 0.19±0.01, p=0.002). Global male-female genetic correlation was high (rg=0.99), but local genetic correlation analyses revealed four genomic regions with significant negative correlations. Genes within these regions were enriched for processes related to hypothalamus cell differentiation, cellular response to gonadotropin stimulus, and ovulation cycle regulation. Gene-based analyses revealed sex-dependent contributions, and TWAS highlighted female-specific associations (e.g., SEPTIN4, RP11-572O6.1, CELSR3) and male-specific associations (e.g., SNX19, SLTM, LETM2).
PRS analyses further showed asymmetry: in females, a female-specific GWAS improved prediction by 2% compared to the male-specific GWAS (11.12% vs. 9.04%), whereas in males, the male-specific GWAS explained only slightly more variance than the female-specific one (12.46% vs. 11.97%). This suggests that female-derived PRS generalizes better across sexes than male-derived PRS.
Our findings provide new insights into the sex-specific genetics of SCZ, suggesting the need to consider sex in genetic studies. Our results may have implications for personalized approaches to SCZ prevention and treatment based on sex.