Predictive effect of PD-L1 expression for immune checkpoint inhibitor (PD-1/PD-L1 inhibitors) treatment for non-small cell lung cancer: A meta-analysis

Abstract Background Programmed death-ligand-1 (PD-L1) is a well-known predictive biomarker in non-small cell lung cancer (NSCLC) patients, however, its accuracy remains controversial. Here, we investigated the correlation between PD-L1 expression level and efficacy of its inhibitors, and hence assessed the predictive effect of PD-L1 expression. Methods Studies that evaluated the efficacy of programmed death-1 (PD-1)/ PD-L1 inhibitors in advanced NSCLC patients according to tumor PD-L1 expression levels were searched for on Medline, Cochrane Library, and Embase. The pooled risk ratio (RR) and 95% confidence intervals (95% CIs) were calculated for the objective response rate (ORR) with overall survival (OS) and progression-free survival (PFS) were measured in terms of hazard ratio (HR) and the corresponding 95% CIs. Results 1432 NSCLC patients from six randomized controlled trials (RCTs) were included and three PD-1/PD-L1 inhibitors (atezolizumab, nivolumab, and pembrolizumab) were used to treat the patients. A significantly higher ORR was observed in the high PD-L1 expression group compared to the low expression group (0.35 [95% CI, 0.30–0.40] vs 0.11 [95% CI, 0.09–0.14]). The results of the subgroup analysis, grouped by the type of drugs and antibodies which assess immune checkpoint inhibitors were identical with the pooled result. However, our study showed that PD-L1 expression was neither prognostic nor predictive of overall survival (OS) or progression-free survival (PFS) in patients treated with PD-1/PD-L1 inhibitors compared to chemotherapy. Conclusions PD-L1 can be a predictive biomarker for ORR. Nevertheless, PD-L1 expression is not a good predictive tool for OS and PFS.