Construction of Metabolic Molecular Classification and Immune Characteristics for the Prognosis Prediction of Ovarian Cancer

Abstract

Background. Ovarian cancer (OC) is a malignant tumor that seriously threatens women’s health. Molecular classification based on metabolic genes can reflect the deeper characteristics of ovarian cancer and provide support for prognostic evaluation and the guidance of individualized treatment. Method. The metabolic subtypes were determined by consensus clustering and CDF. We used the ssGSEA method to calculate the IFNγ score of each patient. The CIBERSORT method was used to evaluate the score distribution and differential expression of 22 immune cells, and LDA was applied to establish a subtype classification feature index. The Kaplan-Meier and ROC curves were generated to validate the prognostic performance of metabolic subtypes in different cohorts. WGCNA was used to screen the coexpression modules associated with metabolic genes. Results. We obtained three metabolic subtypes (MC1, MC2, and MC3). MC2 had the best prognosis, and MC1 and MC3 had poor prognoses. Consistently, MC2 subtype had higher T cell lytic activity and lower angiogenesis, IFNγ, T cell dysfunction, and rejection scores. TIDE analysis showed that MC2 patients were more likely to benefit from immunotherapy; MC1 patients were more sensitive to immune checkpoint inhibitors and traditional chemotherapy drugs. The multiclass AUCs based on the RNASeq and GSE cohorts were 0.93 and 0.84, respectively. Finally, we screened 11 potential gene markers related to the metabolic characteristic index that could be used to indicate the prognosis of OC. Conclusion. Molecular subtypes related to metabolism are crucial to comprehensively understand the molecular pathological characteristics related to metabolism for OC development, explore reliable markers for prognosis, improve the OC staging system, and guide personalized treatment.