Abstract
Antibodies against programmed death-1 (PD-1), and its ligand, (PD-L1) have been approved recently for the treatment of small-cell lung cancer (SCLC). Although there are previous reports that addressed PD-L1 detection on tumour cells in SCLC, there is no comprehensive meta-analysis on the prevalence of PD-L1 expression in SCLC. We performed a systematic search of the PubMed, Cochrane Library and EMBASE databases to assess reports on the prevalence of PD-L1 expression and the association between PD-L1 expression and overall survival (OS). This meta-analysis included 27 studies enrolling a total of 2792 patients. The pooled estimate of PD-L1 expression was 26.0% (95% CI 17.0-37.0), (22.0% after removing outlying studies). The effect size was significantly heterogeneous (I2 = 97.4, 95% CI: 95.5-98.5, p < 0.0001).Positive PD-L1 expression was a favourable prognostic factor for SCLC but not statistically significant (HR = 0.86 (95% CI (0.49-1.50), p = 0.5880; I2 = 88.7%, p < 0.0001). Begg's funnel plots and Egger's tests indicated no publication bias across included studies (p > 0.05). Overall, there is heterogeneity in the prevalence of PD-L1 expression in SCLC tumour cells across studies. This is significantly moderated by factors such as immunohistochemistry (IHC) evaluation cut-off values, and assessment of PD-L1 staining patterns as membranous and/or cytoplasmic. There is the need for large size, prospective and multicentre studies with well-defined protocols and endpoints to advance the clinical value of PD-L1 expression in SCLC.