AUTHOR=Shu Pei , Liu Ning , Luo Xu , Tang Yuanling , Chen Zhebin , Li Dandan , Miao Dong , Duan Jiayu , Yan Ouying , Sheng Leiming , Ouyang Ganlu , Wang Sen , Jiang Dan , Deng Xiangbing , Wang Ziqiang , Li Qingyun , Wang Xin 

TITLE=An immune-related gene prognostic prediction risk model for neoadjuvant chemoradiotherapy in rectal cancer using artificial intelligence

JOURNAL=Frontiers in Oncology

VOLUME=Volume 14 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2024.1294440

DOI=10.3389/fonc.2024.1294440

ISSN=2234-943X

ABSTRACT=Background: To develop and validate an immune-related gene prognostic model (IRGPM) that can predict disease-free survival (DFS) in patients with locally advanced rectal cancer (LARC) who received neoadjuvant chemoradiotherapy and to clarify the immune characteristics of patients with different prognostic risks.We obtained transcriptomic and clinical data from the Gene Expression Omnibus (GEO) database and database of West China Hospital. Genes in the RNA immune-oncology panel were extracted. Elastic net was used to identify the immune-related genes that significantly affected the DFS of patients. A prognostic risk model (IRGPM) for rectal cancer was constructed with the random forest method. The prognostic risk score was calculated by the model, and the patients were divided into high-and low-risk groups according to the risk score. Immune characteristics were analyzed and compared between the high-and low-risk groups.Results: A total of 407 LARC samples were used in this study. A 20-gene signature was identified by elastic net. The IRGPM was constructed on the basis of the 20 immune-related genes. Kaplan-Meier survival analysis showed poorer 5-year DFS in the high-risk group than in the low-risk group, and the receiver operating characteristic (ROC) curve suggested good model prediction. The model was validated in the GSE190826 cohort and the cohort from our institution. The differentially expressed genes between the high-and low-risk groups were enriched in cytokine-cytokine receptor interactions. The patients in the low-risk group had higher immune scores than the patients in the high-risk group. Subsequently, we found that activated B cells, activated CD8 T cells, central memory CD8 T cells, macrophages, T follicular helper cells and type 2 helper cells were more abundant in the low-risk group. Moreover, we compared the expression of immune checkpoints and found that the low-risk group had a higher PDCD1 expression level.The IRGPM, which was constructed based on the random forest and elastic net methods, is a promising method to distinguish DFS in LARC patients treated with a standard strategy. The low risk group identified by IRGPM was characterized by the activation of adaptive immunity in tumor microenvironment.