Genetic architecture of schizophrenia clinical subtype

Clinical heterogeneity in schizophrenia (SZ) presents a significant challenge to genetic research, as diverse symptom profiles may stem from distinct genetic risk factors. Although previous studies have stratified patients into symptom-based subtypes, and preliminary evidence suggests the presence of distinct architectures, these findings remain limited. This study aimed to identify the clinically defined SZ subtypes and investigate the genetic differences underlying different subtypes. In a Chinese Han cohort of 2,416 SZ patients, we applied K-means cluster analysis to symptom profiles to identify clinical subtypes. The identified subtype structure was validated in an independent cohort of 480 patients. Subsequently, subtype-specific genome-wide association studies (GWASs) were conducted to identify genetic risk loci associated with individual subtypes. Three stable subtypes were identified: Cluster-L (low severity), Cluster-S (severe), and Cluster-N (predominant negative symptoms). Reproducibility of this classification was confirmed in the independent cohort. The three subtypes also exhibited significantly different symptom network structures. In GWAS analysis, three independent genome-wide significant loci were detected, including a locus within the CNTN2 gene that showed significance in Cluster-N (lead SNP rs3767295, P = 6.40 × 10 -10 , OR = 0.62). Gene-based analyses identified eight significant risk genes, seven of which were only significant in Cluster-N. In addition, distinct patterns emerged in both pathway-specific polygenic risk scores (pPRS) and cell type-specific PRS (ctPRS). Specifically, GABA-PRS (OR = 1.19 [1.06–1.34], Padjusted = 0.040) and oligodendrocyte progenitor cells-PRS (OR = 1.25 [1.11–1.41], Padjusted = 0.004) were significantly elevated in Cluster-N, Astrocyte-PRS was elevated only in Cluster-S (OR = 1.30 [1.10–1.53], Padjusted = 0.033), and Oligodendrocyte-PRS was elevated in both Cluster-S (OR = 1.28 [1.10–1.50], Padjusted = 0.036) and Cluster-N (OR = 1.19 [1.06–1.34], Padjusted = 0.047). These findings underscore the value of reducing clinical heterogeneity in improving statistical power to detect genetic risk loci. They further demonstrate that SZ patients with distinct symptom profiles harbor differential genetic liabilities, with certain genetic effects appearing more pronounced in specific symptom-defined subtypes.