Prognostic implications of programmed death ligand 1 expression in resected lung adenocarcinoma: a systematic review and meta-analysis

Abstract

Abstract OBJECTIVES Conflicting results have been reported about the prognostic value of programmed death ligand 1 (PD-L1) protein and gene expression in lung adenocarcinoma. METHODS We performed a comprehensive online search to explore the association between PD-L1 expression (protein and messenger RNA) and overall survival (OS) or disease-free survival. Outcomes also included pooled rates of high PD-L1 protein expression in different cell types, per threshold used and per antibody used. A pooled gene expression analysis was also performed on 3 transcriptomic data sets that were obtained from The Cancer Genome Atlas database and the Gene Expression Omnibus database. RESULTS A total of 6488 patients from 25 studies were included. The pooled results suggested that high PD-L1 expression was associated with shorter OS [hazard ratio (HR) 1.57; P < 0.001] and disease-free survival (HR 1.341; P = 0.037) in the overall population. The overall pooled rate of high PD-L1 protein expression was 29% (95% confidence interval 23–34%) in tumour cells. In subgroup analysis, high PD-L1 protein expression in tumour cells predicted worse OS and disease-free survival. A pooled analysis of The Cancer Genome Atlas and Gene Expression Omnibus data sets revealed that higher levels of PD-L1 messenger RNA predicted poorer OS in the entire population. CONCLUSIONS This study is, to our knowledge, the largest pooled analysis on the subject to shed light on the high expression rate of PD-L1 and the prognostic value of high PD-L1 expression in resected lung adenocarcinomas. PD-L1 gene expression is a promising prognostic factor for patients with surgically resected lung adenocarcinoma. Standardization of staining should be underscored prior to routine implementation.