Ferroptosis and cuproptosis prognostic signature for prediction of prognosis, immunotherapy and drug sensitivity in hepatocellular carcinoma: development and validation based on TCGA and ICGC databases
Qi Ma,Yuan Hui,Bang-Rong Huang,Bin-Feng Yang,Jing-Xian Li,Ting-Ting Fan,Xiang-Chun Gao,Da-You Ma,Wei-Fu Chen,Zheng-Xue Pei
DOI: https://doi.org/10.21037/tcr-22-2203
2023-01-30
Abstract:Background: Hepatocellular carcinoma (HCC) is a common malignancy. Ferroptosis and cuproptosis promote HCC spread and proliferation. While fewer studies have combined ferroptosis and cuproptosis to construct prognostic signature of HCC. This work attempts to establish a novel scoring system for predicting HCC prognosis, immunotherapy, and medication sensitivity based on ferroptosis-related genes (FRGs) and cuproptosis-related genes (CRGs). Methods: FerrDb and previous literature were used to identify FRGs. CRGs came from original research. The Cancer Genome Atlas (TCGA) and International Cancer Genome Consortium (ICGC) databases included the HCC transcriptional profile and clinical information [survival time, survival status, age, gender, Tumor Node Metastasis (TNM) stage, etc.]. Correlation, Cox, and least absolute shrinkage and selection operator (LASSO) regression analyses were used to narrow down prognostic genes and develop an HCC risk model. Using "caret", R separated TCGA-HCC samples into a training risk set and an internal test risk set. As external validation, we used ICGC samples. We employed Kaplan-Meier analysis and receiver operating characteristic (ROC) curve to evaluate the model's clinical efficacy. CIBERSORT and TIMER measured immunocytic infiltration in high- and low-risk populations. Results: TXNRD1 [hazard ratio (HR) =1.477, P<0.001], FTL (HR =1.373, P=0.001), GPX4 (HR =1.650, P=0.004), PRDX1 (HR =1.576, P=0.002), VDAC2 (HR =1.728, P=0.008), OTUB1 (HR =1.826, P=0.002), NRAS (HR =1.596, P=0.005), SLC38A1 (HR =1.290, P=0.002), and SLC1A5 (HR =1.306, P<0.001) were distinguished to build predictive model. In both the model cohort (P<0.001) and the validation cohort (P<0.05), low-risk patients had superior overall survival (OS). The areas under the curve (AUCs) of the ROC curves in the training cohort (1-, 3-, and 5-year AUCs: 0.751, 0.727, and 0.743), internal validation cohort (1-, 3-, and 5-year AUCs: 0.826, 0.624, and 0.589), and ICGC cohort (1-, 3-, and 5-year AUCs: 0.699, 0.702, and 0.568) were calculated. Infiltration of immune cells and immunological checkpoints were also connected with our signature. Treatments with BI.2536, Epothilone.B, Gemcitabine, Mitomycin.C, Obatoclax. Mesylate, and Sunitinib may profit high-risk patients. Conclusions: We analyzed FRGs and CRGs profiles in HCC and established a unique risk model for treatment and prognosis. Our data highlight FRGs and CRGs in clinical practice and suggest ferroptosis and cuproptosis may be therapeutic targets for HCC patients. To validate the model's clinical efficacy, more HCC cases and prospective clinical assessments are needed.