Immune signature-based risk stratification and prediction of immune checkpoint inhibitor’s efficacy for lung adenocarcinoma
Ming Yi, Anping Li, Linghui Zhou, Qian Chu, Suxia Luo, Kongming Wu
Abstract
BACKGROUND: Lung adenocarcinoma (LUAD) is a common pulmonary malignant disease with a poor prognosis. There were limited studies investigating the influences of the tumor immune microenvironment on LUAD patients' survival and response to immune checkpoint inhibitors (ICIs). METHODS: Based on TCGA-LUAD dataset, we constructed a prognostic immune signature and validated its predictive capability in the internal as well as total datasets. Then, we explored the differences of tumor-infiltrating lymphocytes, tumor mutation burden, and patients' response to ICI treatment between the high-risk score group and low-risk score group. RESULTS: This immune signature consisted of 17 immune-related genes, which was an independent prognostic factor for LUAD patients. In the low-risk score group, patients had better overall survival. Although the differences were non-significant, patients with low-risk scores had more tumor-infiltrating follicular helper T cells and fewer macrophages (M0), which were closely related to clinical outcomes. Additionally, the total TMB was markedly decreased in the low-risk score group. Using immunophenoscore as a surrogate of ICI response, we found that patients with low-risk scores had significantly higher immunophenoscore. CONCLUSION: The 17-immune-related genes signature may have prognostic and predictive relevance with ICI therapy but needs prospective validation.