Programmed cell death (PCD) is an overwhelming factor affecting tumor cell metastasis, but the mechanism of PCD in ovarian cancer (OV) is still uncertain.
To define the molecular subtypes of OV, we performed unsupervised clustering based on the expression level of prognosis related PCD genes in the Cancer Genome Atlas (TCGA)-OV. COX and least absolute shrinkage and selection operator (LASSO) COX analysis were used to identify the OV prognostic related PCD genes, and the genes identified according to the minimum Akaike information criterion (AIC) were the OV prognostic characteristic genes. According to the regression coefficient in the multivariate COX analysis and gene expression data, the Risk Score of OV prognosis was constructed. Kaplan-Meier analysis was conducted to assess the prognostic status of OV patients, and receiver operating characteristic (ROC) curves were conducted to assess the clinical value of Risk Score. Moreover, RNA-Seq date of OV patient derived from Gene Expression Omnibus (GEO, GSE32062) and the International Cancer Genome Consortium (ICGC) database (ICGC-AU), verifying the robustness of the Risk Score
9-gene composition Risk Score system was finally determined by COX and LASSO COX analysis. Patients in the low Risk Score group possessed improved prognostic status, immune activity. PI3K pathway activity was increased in the high Risk Score group. In the chemotherapy drug sensitivity analysis, we found that the high Risk Score group might be more suitable for treatment with PI3K inhibitors Taselisib and Pictilisib. In addition, we found that patients in the low-risk group responded better to immunotherapy.
Risk Score of 9-gene composition of PCD signature possesses promising clinical potential in OV prognosis, immunotherapy, immune microenvironment activity, and chemotherapeutic drug selection, and our study provides the basis for an in-depth investigation of the PCD mechanism in OV.