Integrated multiomics characterization reveals cuproptosis-related hub genes for predicting the prognosis and clinical efficacy of ovarian cancer

Abstract

Background: As a prevalent malignancy in women, ovarian cancer (OC) presents a challenge in clinical practice because of its poor prognosis and poor therapeutic efficacy. The mechanism by which cuproptosis activity is accompanied by immune infiltration in OC remains unknown. Here, we investigated cuproptosis-related OC subtypes and relevant immune landscapes to develop a risk score (RS) model for survival prediction.

Methods: Cuproptosis-related genes (CRGs) were identified to construct molecular subtypes via an unsupervised clustering algorithm based on the expression profiles of survival-related CRGs in the GEO database. Single-cell datasets were used to estimate immune infiltration among subtypes. The RS oriented from molecular subtypes was developed via LASSO Cox regression in the TCGA OC dataset and independently validated in the GEO and TCGA datasets. Hub markers from RS were identified in tissues and cell lines. The function of the key gene from RS was identified in vitro.

Results: We investigated cuproptosis activity and immune infiltration to establish three clinical subtypes of OC based the differentially expressed genes (DEGs) from CRGs to create an RS model validated for clinical efficacy and prognosis. Six hub genes from the RS served as ongenic markers in OC tissues and cell lines. The function of GAS1 in the RS model revealed that it exerts oncogenic effects.

Conclusions: Our study provides a novel RS model including 6 hub genes associated with cuproptosis and immune infiltration to predict OC prognosis as well as clinical efficacy.

Keywords: RiskScore; bulk RNA sequencing; cuproptosis; ovarian cancer; single-cell sequencing.