Abstract
BACKGROUND: The shortened life expectancy in schizophrenia (SCZ) patients may be correlated with most cancers, yet there is heterogeneity in the studies examining these correlations. This study explored the expression of SCZ-related genes (HTR2A, COMT, and PRODH) in pan-cancer analysis. It helped to enhance the mechanistic understanding of the SCZ-cancer relationship and their immune mechanisms at the genetic level. Additionally, this study established a survival prediction model for glioblastoma and low-grade glioma (GBMLGG). METHODS AND RESULTS: SCZ-associated genes (HTR2A, COMT, and PRODH) were subjected to pan-cancer analysis. COX regression analysis and survival analysis were carried out for differentially expressed genes in multiple cancers, and finally, GBMLGG was derived as the focus for further detailed analysis. The immune scores and immune cell infiltration analyses were performed. All three genes were considerably linked with immune infiltration in GBMLGG, consistent with survival analysis. Based on the immunocyte analysis, it was observed that CD8(+) T cells might be critically involved in the survival of GBMLGG. Genomic heterogeneity studies identified correlations of three genes with GBMLGG in tumor mutational burden (TMB) and mutant-allele tumor heterogeneity (MATH). HTR2A and COMT were significantly negatively correlated in TMB. Furthermore, it was found that HTR2A had a significant positive correlation with MATH, whereas PRODH had a significant negative correlation with MATH. Accordingly, a survival prediction model was constructed for GBMLGG using these three genes and clinical data, with better results obtained when evaluated in two separate datasets. Finally, gene expression validation and further immunocyte analysis were carried out in the single-cell RNA sequencing (scRNA-seq) data of glioma. CONCLUSION: SCZ-associated genes (HTR2A, COMT, and PRODH) were significantly differentially expressed in the carcinogenesis and survival of multiple cancers. The up or downregulation of gene expression varied across cancer types. In the GBMLGG analysis, upregulation of HTR2A and COMT was significantly positively correlated with carcinogenesis, while the opposite was noted for PRODH. Furthermore, a negative correlation was found between the upregulation of HTR2A and COMT and the survival of GBMLGG, and the opposite was also noted for PRODH. As reflected in the immunocyte analysis, abnormal expression of the three genes might be linked with CD8(+) T cell infiltration, which might be critically involved in the survival of GBMLGG patients. The expression of HTR2A and COMT may inversely affect the efficacy of immunotherapy through the TMB pathway and further affect the prognosis of patient survival. The expression of HTR2A might positively indicate the degree of tumor heterogeneity through MATH and further affect the survival and prognosis of patients. The negative correlation of PRODH led to the opposite effect. Finally, the constructed survival prediction model demonstrated good predictive value, which was well validated in scRNA-seq analysis.