The clinicopathological and prognostic significance of PD-L1 expression assessed by immunohistochemistry in lung cancer: A meta-analysis of 50 studies with 11,383 patients

AME Lung Cancer Collaborative Group

Research output: Contribution to journalArticle

1 Scopus citations

Abstract

Background: We conducted a meta-analysis to systematically evaluate the relationship between programmed death-ligand 1 (PD-L1) expression and survival in patients with lung cancer. Methods: The electronic databases PubMed, Embase, Cochrane, and Web of Science were searched up to January 2nd, 2018, for articles relating to PD-L1 expression detected by immunohistochemistry (IHC) and lung cancer patient prognosis. Results: Fifty studies including 11,383 patients published between 2011 and 2017 were enrolled in this meta-analysis. The pooled hazard ratios (HRs) and 95% confidence intervals (CIs) suggested that PDL1 IHC expression was related to poor overall survival (OS) (HR =1.45, 95% CI: 1.24–1.68). In subgroup analysis categorized according to sample type, cut-off value, ethnicity and TNM stage, the pooled results demonstrated inferior survival in the PD-L1 positive group when the PD-L1 expression was detected by resection specimens (P=0.000), 5% was taken as the cutoff value (P=0.000), the patients were in early stage (I–III) (P=0.000), and the geographic setting of the study was in Asia (P=0.000). Besides, patients with high PDL1 expression had shorter OS in NSCLC (P=0.000), ADC (P=0.000), SCC (P=0.353) and LELC (P=0.810), while no significant difference was observed in SCLC (P=0.000). The pooled odds ratios (ORs) suggested that PD-L1 expression was associated with male (P<0.001), smoker (P<0.001), poor tumor differentiation (P=0.014), large tumor size (P=0.132), positive lymph nodal metastasis (P=0.002), EGFR wild-type status (P<0.001) and KRAS mutations (P=0.393). However, age (P=0.15) and ALK rearrangements (P=0.567) had no bearing on PD-L1 expression. Conclusions: PD-L1 expression that is associated with several clinicopathological feactures may serve as a poor prognostic biomarker for patients with lung cancer.

Original languageEnglish (US)
Pages (from-to)429-449
Number of pages21
JournalTranslational Lung Cancer Research
Volume8
Issue number4
DOIs
StatePublished - Jan 1 2019

    Fingerprint

Keywords

  • Lung cancer
  • Meta-analysis
  • Prognosis
  • Programmed cell death ligand 1

ASJC Scopus subject areas

  • Oncology

Cite this