A signature of estimate-stromal-immune score-based genes associated with the prognosis of lung adenocarcinoma

Qianli Ma; Yang Chen; Fei Xiao; Yang Hao; Zhiyi Song; Jin Zhang; Katsuhiro Okuda; Sang-Won Um; Mario Silva; Yoshihisa Shimada; Chaozeng Si; Chaoyang Liang

doi:10.21037/tlcr-21-223

A signature of estimate-stromal-immune score-based genes associated with the prognosis of lung adenocarcinoma

Transl Lung Cancer Res. 2021 Mar;10(3):1484-1500. doi: 10.21037/tlcr-21-223.

Authors

Qianli Ma¹, Yang Chen², Fei Xiao¹, Yang Hao¹, Zhiyi Song¹, Jin Zhang¹, Katsuhiro Okuda³, Sang-Won Um⁴, Mario Silva⁵, Yoshihisa Shimada⁶, Chaozeng Si⁷, Chaoyang Liang¹

Affiliations

¹ Department of Thoracic Surgery, China-Japan Friendship Hospital, Beijing, China.
² Department of Biochemistry and Molecular Biology, The State Key Laboratory of Medical Molecular Biology, Institute of Basic Medical Sciences, School of Basic Medicine, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.
³ Department of Oncology, Immunology and Surgery, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan.
⁴ Division of Pulmonary and Critical Care Medicine, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, South Korea.
⁵ Section of Scienze Radiologiche, Department of Medicine and Surgery (DiMeC), University of Parma, Parma, Italy.
⁶ Department of Thoracic Surgery, Tokyo Medical University, Tokyo, Japan.
⁷ Department of Information Management, China-Japan Friendship Hospital, Beijing, China.

Abstract

Background: Immune and stromal component evaluation is necessary to establish accurate prognostic markers for the prediction of clinical outcomes in lung adenocarcinoma (LUAD). We aimed to develop a gene signature based on the Estimation of STromal and Immune cells in MAlignant Tumor tissues using Expression data (ESTIMATE)-stromal-immune score in LUAD.

Methods: The transcriptomic profiles of patients with LUAD were obtained from The Cancer Genome Atlas (TCGA), and the immune and stromal scores were derived using the ESTIMATE algorithm. The prognostic signature genes were selected from the differentially expressed genes (DEGs) using the robust partial likelihood-based cox proportional hazards regression method. The negative log-likelihood and the Akaike Information Criterion (AIC) were used to identify the optimal gene signature. The validation was carried out in 2 independent datasets from the Gene Expression Omnibus (GSE68571 and GSE72094).

Results: Patients with high ESTIMATE, stromal, and immune scores had better overall survivals (P=0.0035, P=0.066, and P=0.0077). The expression of thirty-seven genes was related to ESTIMATE-stromal-immune score. A risk stratification model was developed based on a gene signature containing CD74, JCHAIN, and PTGDS. The ESTIMATE-stromal-immune risk score was revealed to be a prognostic factor (P=0.009) after multivariate analysis. Four groups were classified based on this risk stratification model, yielding increasing survival outcomes (log-rank test, P=0.0051). This risk stratification model and other clinicopathological factors were combined to generate a nomogram. The calibration curves showed perfect agreement between the nomogram-predicted outcomes and those actually observed. Similar observations were made in 2 independent cohorts.

Conclusions: The gene signature based on the ESTIMATE-stromal-immune score could predict the prognosis of patients with LUAD.

Keywords: Lung adenocarcinoma (LUAD); immune; prognosis; stromal.