Identification of an immune signature predicting prognosis risk of patients in lung adenocarcinoma

Abstract

Background Lung cancer has become the most common cancer type and caused the most cancer deaths. Lung adenocarcinoma (LUAD) is one of the major type of lung cancer. This study aimed to establish a signature based on immune related genes that can predict patients’ OS for LUAD. Methods The expression data of 976 LUAD patients from The Cancer Genome Atlas database (training set) and the Gene Expression Omnibus database (four testing sets) and 1534 immune related genes from the ImmPort database were used for generation and validation of the signature. The glmnet Cox proportional hazards model was used to find the best gene model and construct the signature. To assess the independently prognostic ability of the signature, the Kaplan–Meier survival analysis and Cox’s proportional hazards model were performed. Results A gene model consisting of 30 immune related genes with the highest frequency after 1000 iterations was used as our signature. The signature demonstrated robust prognostic ability in both training set and testing set and could serve as an independent predictor for LUAD patients in all datasets except GSE31210. Besides, the signature could predict the overall survival (OS) of LUAD patients in different subgroups. And this signature was strongly associated with important clinicopathological factors like recurrence and TNM stage. More importantly, patients with high risk score presented high tumor mutation burden. Conclusions This signature could predict prognosis and reflect the tumor immune microenvironment of LUAD patients, which can promote individualized treatment and provide potential novel targets for immunotherapy. Electronic supplementary material The online version of this article (10.1186/s12967-019-1824-4) contains supplementary material, which is available to authorized users.

An immune infiltration signature to predict the overall survival of patients with colon cancer

“…To date, however, this viewpoint has not been well‐translated into clinical implications and significance. In addition, a number of existing studies have focused on the role of immune gene signatures, rather than of immune cells composition, in the initiation and progression of cancer . In the present study, we first constructed a PIS model for predicting the overall survival of patients based on the fractions of 17 immune cell types using LASSO penalized regression.…”

Prognostic signature of ovarian cancer based on 14 tumor microenvironment genes

“…The study describes the comprehensive features of gastric cancer TME-related genes characteristics and provide new strategies for cancer treatment [9]. Another present study found that many elements of TME other than tumor epithelial cells in uence the progression of non-small cell lung cancer [10]. Consequently, new treatment strategies and paradigms are of great need for these patients.…”

IRGS: an immune-related gene classifier for lung adenocarcinoma prognosis

“…Although it has been shown in different studies that a specific gene set can predict the prognosis of cancer patients [20,21], the number of genes in a model is often beyond practical need [22] or of immune irrelevance [23], making translational feasibility low. With the development of tumour immunotherapy, the role of the immune system in the development of cancer has once again been taken into consideration.…”