Landscape of PCOS co-expression gene and its role in predicting prognosis and assisting immunotherapy in endometrial cancer

Abstract Background Endometrial carcinoma (EC) is the sixth most frequent malignancy in women and is often linked to high estrogen exposure. Polycystic ovarian syndrome (PCOS) is a known risk factor for EC, but the underlying mechanisms remain unclear. Methods We investigated shared gene signals and potential biological pathways to identify effective therapy options for PCOS- and EC-related malignancies. Weighted gene expression network analysis (WGCNA) was used to identify genes associated with PCOS and EC using gene expression data from the Gene Expression Omnibus (GEO) and Cancer Genome Atlas (TCGA) datasets. Enrichment analysis using Cluego software revealed that the steroid hormone biosynthetic process was a critical feature in both PCOS and EC. A predictive signature encompassing genes involved in steroid hormone production was developed using multivariate and least absolute shrinkage and selection operator (LASSO) regression analysis to predict the prognosis of EC. Then, we conducted further experimental verification. Results Patients in the TCGA cohort with high predictive scores had poorer outcomes than those with low scores. We also investigated the relationship between tumor microenvironment (TME) features and predictive risk rating and found that patients with low-risk scores had higher levels of inflammatory and inhibitory immune cells. Also, we found that immunotherapy against anti-CTLA4 and anti-PD-1/PD-L1 was successful in treating individuals with low risk. Low-risk individuals were more responsive to crizotinib therapy, according to further research performed using the “pRRophetic” R package. We further confirmed that IGF2 expression was associated with tumor cell migration, proliferation, and invasion in EC cells. Conclutions By uncovering the pathways and genes linking PCOS and EC, our findings may provide new therapeutic strategies for patients with PCOS-related EC