Ma Qi, Hui Yuan, Huang Bang-Rong, Yang Bin-Feng, Li Jing-Xian, Fan Ting-Ting, Gao Xiang-Chun, Ma Da-You, Chen Wei-Fu, Pei Zheng-Xue
School of Integrative Medicine, Gansu University of Traditional Chinese Medicine, Lanzhou, China.
Department of Oncology, Gansu Provincial Hospital of Traditional Chinese Medicine, Lanzhou, China.
Transl Cancer Res. 2023 Jan 30;12(1):46-64. doi: 10.21037/tcr-22-2203. Epub 2022 Dec 19.
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).
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.
[hazard ratio (HR) =1.477, P<0.001], (HR =1.373, P=0.001), (HR =1.650, P=0.004), (HR =1.576, P=0.002), (HR =1.728, P=0.008), (HR =1.826, P=0.002), (HR =1.596, P=0.005), (HR =1.290, P=0.002), and (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.
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.
肝细胞癌（HCC）是一种常见的恶性肿瘤。铁死亡和铜死亡促进HCC的扩散和增殖。然而，将铁死亡和铜死亡结合起来构建HCC预后特征的研究较少。本研究试图基于铁死亡相关基因（FRGs）和铜死亡相关基因（CRGs）建立一种用于预测HCC预后、免疫治疗及药物敏感性的新型评分系统。
利用FerrDb和以往文献鉴定FRGs。CRGs来自原始研究。癌症基因组图谱（TCGA）和国际癌症基因组联盟（ICGC）数据库包含HCC转录谱和临床信息[生存时间、生存状态、年龄、性别、肿瘤淋巴结转移（TNM）分期等]。采用相关性分析、Cox分析和最小绝对收缩和选择算子（LASSO）回归分析来筛选预后基因并建立HCC风险模型。使用R语言中的“caret”包将TCGA-HCC样本分为训练风险集和内部测试风险集。作为外部验证，我们使用了ICGC样本。采用Kaplan-Meier分析和受试者工作特征（ROC）曲线评估模型的临床疗效。使用CIBERSORT和TIMER测量高风险和低风险人群中的免疫细胞浸润情况。
[风险比（HR）=1.477，P<0.001]，（HR =1.373，P=0.001），（HR =1.650，P=0.004），（HR =1.576，P=0.002），（HR =1.728，P=0.008），（HR =1.826，P=0.002），（HR =1.596，P=0.005），（HR =1.290，P=0.002），以及（HR =1.306，P<0.001）被用于构建预测模型。在模型队列（P<0.001）和验证队列（P<0.05）中，低风险患者的总生存期（OS）均更优。计算了训练队列（1年、3年和5年AUC分别为：0.751、0.727和0.743）、内部验证队列（1年、3年和5年AUC分别为：0.826、0.624和0.589）以及ICGC队列（1年、3年和5年AUC分别为：0.699、0.702和0.568）中ROC曲线的曲线下面积（AUC）。免疫细胞浸润和免疫检查点也与我们构建的特征相关。使用BI.2536、埃坡霉素B、吉西他滨、丝裂霉素C、甲磺酸奥巴托克斯和舒尼替尼进行治疗可能使高风险患者获益。
我们分析了HCC中的FRGs和CRGs图谱，并建立了一个用于治疗和预后的独特风险模型。我们的数据突出了FRGs和CRGs在临床实践中的意义，并表明铁死亡和铜死亡可能是HCC患者的治疗靶点。为验证该模型的临床疗效，需要更多的HCC病例和前瞻性临床评估。
