Abstract 6559: Model constructed by ferroptosis-related genes and m6A genes predicts the sensitivity to different drug treatments

Abstract Background: Immune checkpoint blockade (ICB) therapies exhibit distinct response in hepatocellular carcinoma (HCC) patients, therefore biomarkers to screen beneficial patients remains an urgent need in clinical practice. Ferroptosis, an unique form of programmed cell death (PCD), has been implicated in the carcinogenesis and considered as one of the underlying mechanism of targeted and chemotherapeutic therapies in HCC. N6-methyladenosine (m6A), the most dominant form of RNA post-transcriptional modification, has been reported to regulat cancer-associated genes, including ferroptosis-related genes. Methods: Differentially expressed genes (DEGs) between tumor and normal tissues were screened from TCGA-LIHC dataset. The overlap of DEGs and ferroptosis-related genes (FRGs) resulted in a set of 51 genes. These genes together with 22 m6A-related genes were next subjected to univariate Cox regression for their association with survival. In total, 35 genes showed significant association with survival in this analysis, and they were then used to construct the regression model by Lasso-Cox regression analysis. We then validated the performance of the signature derived from the model in TCGA-LIHC and an external dataset (GSE14520). The signature was also examined for their association with immune-related characters, and their performance in predicting drug response using datasets of immunotherapy cohort studies and cellular experimental studies. Results: We constructed this prognosis model by expression level of four genes (G6PD, YTHDF1, STMN1, and SLC7A11). Patients with a high-risk score of the signature have a shortened survival time, and the signature is an independent predictor of the prognosis. We also found that the difference between the risk score before and after ICB treatment was significantly associated with sensitivity to the treatment: an increase in the risk score after the treatment correlated with an adverse response. Using the in vitro cellular experimental dataset, we detected a positive association between the risk score and the sensitivity to 5-fluorouracil and sorafenib. Conclusion: The prognosis signature constructed by four ferroptosis- and m6A- related genes exhibited promising predictive value for OS of HCC patients. The signature could also indicate the sensitivity for different kinds of clinical treatment. A higher risk score indicated a higher resistance for immunotherapy but a higher sensitivity for chemotherapy and RTKs inhibitor. The results of this study not only help to improve the clinical practice of HCC, but also provide new insights into unveiling the underlying mechanisms of tumorigenesis and progression. Citation Format: Tao Han, Mengyuan Liu. Model constructed by ferroptosis-related genes and m6A genes predicts the sensitivity to different drug treatments [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6559.