Identification of Prognostic and Predictive Biomarkers and Druggable Targets among 205 Antioxidant Genes in 21 Different Tumor Types via Data-Mining

Abstract

(1) Background: Oxidative stress is crucial in carcinogenesis and the response of tumors to treatment. Antioxidant genes are important determinants of resistance to chemotherapy and radiotherapy. We hypothesized that genes involved in the oxidative stress response may be valuable as prognostic biomarkers for the survival of cancer patients and as druggable targets. (2) Methods: We mined the KM Plotter and TCGA Timer2.0 Cistrome databases and investigated 205 antioxidant genes in 21 different tumor types within the context of this investigation. (3) Results: Of 4347 calculations with Kaplan–Meier statistics, 84 revealed statistically significant correlations between high gene expression and worse overall survival (p p ALOX5, ALOX5AP, EPHX4, G6PD, GLRX3, GSS, PDIA4, PDIA6, PRDX1, SELENOH, SELENON, STIP1, TXNDC9, TXNDC12, TXNL1, TXNL4A, and TXNRD1). (4) Conclusions: We concluded that a sub-set of antioxidant genes might serve as prognostic biomarkers for overall survival and as druggable targets. Renal and liver tumors may be the most suitable entities for this approach.

Keywords

• drug resistance
• Kaplan–Meier analysis
• oxidative stress
• prognostic value
• survival analysis
• the cancer genome atlas