Machine learning-based construction of a ferroptosis and necroptosis associated lncRNA signature for predicting prognosis and immunotherapy response in hepatocellular cancer

Lei Zhao; Zhixuan You; Zhixun Bai; Jian Xie

doi:10.3389/fonc.2023.1171878

Machine learning-based construction of a ferroptosis and necroptosis associated lncRNA signature for predicting prognosis and immunotherapy response in hepatocellular cancer

Front Oncol. 2023 Apr 20:13:1171878. doi: 10.3389/fonc.2023.1171878. eCollection 2023.

Authors

Lei Zhao¹, Zhixuan You¹, Zhixun Bai², Jian Xie³

Affiliations

¹ The Third Clinical lnstitute, Guangzhou Medical University, Guangzhou, China.
² Department of Nephrology, The Second Affiliated Hospital of Zunyi Medical University, Zunyi, China.
³ Department of Medical Genetics, Zunyi Medical University, Zunyi, China.

Abstract

Introduction: Liver hepatocellular carcinoma (LIHC), one of the most common malignancies worldwide, occurs with high incidence and mortality. Ferroptosis and necroptosis are critically associated with LIHC prognosis. Some long non-coding RNAs (lncRNAs) have been found to induce ferroptosis and necroptosis in hepatocellular carcinoma cells.

Methods: Cox regression analysis was used to construct a risk model for LIHC based on differentially expressed ferroptosis and necroptosis related lncRNAs (F-NLRs), and their expression in SMMC7721, HepG2 and WRL68 cells was detected by qPCR.

Results: Five F-NLRs were associated with LIHC prognosis, including KDM4A-AS1, ZFPM2-AS1, AC099850.3, MKLN1-AS, and BACE1-AS. Kaplan-Meier survival analysis indicated that patients with LIHC in the high-risk group were associated with poor prognosis. The combined F-NLR signature model demonstrated a prognostic AUC value of 0.789 and was more accurate than standard clinical variables for predicting LIHC prognosis. T cell functions and immunotherapy responses differed significantly between patients in the low- and high-risk groups. Additionally, immune checkpoints and m6A-related genes were differentially expressed between patients in the two risk groups. Furthermore, proteins encoded by the five F-NLRs were overexpressed in four liver cancer cell lines compared to that in human liver cell line WRL68. Pan-cancer examination revealed that expression levels of the five F-NLRs differed between most common tumor types and normal tissues.

Conclusion: F-NLRs identified in this study provide a predictive signature representing ferroptosis and necroptosis in LIHC, which correlated well with patient prognosis, clinicopathological characteristics, and immunotherapy responses. The study findings help to elucidate the mechanisms of F-NLRs in LIHC and provide further guidance for the selection and development of immunotherapeutic agents for LIHC.

Keywords: LIHC; ferroptosis; immunotherapy; lncRNA; necroptosis.

Grants and funding

This research was funded by Science and Technology Program of Guizhou Province (QKHRCPT[2019]1332), Doctor science foundation of Zunyi Medical University (F-941), Science and Technology Fund of Guizhou Provincial Health Commission (gzwkj2021-539), and the National Natural Science Foundation of China (82260106).