Brain metastases and immune checkpoint inhibitors in non-small cell lung cancer: a systematic review and meta-analysis

Background The use of immune checkpoint inhibitors (ICIs) for brain metastases (BMs) from non-small cell lung cancer (NSCLC) remains debatable. This study aimed to explore the efficacy of ICIs for NSCLC with BMs. We also evaluated the effect of BMs on outcomes of ICIs. Methods A systematic search of PubMed, Embase, Web of Science, and Cochrane databases was conducted to identify studies where the efficacy of ICIs against BMs from NSCLC, or the association between BMs and outcomes of ICIs were evaluated. Outcomes included intracranial objective response rate (icORR), intracranial disease control rate (icDCR), systemic ORR and DCR. Results Overall, 33 studies were included in this meta-analysis. The pooled icORR was 13% (95%CI 6-23%) and icDCR was 50% (95%CI 40-63%) for programmed cell death-ligand 1 (PD-L1) unselected patients with any BMs. For active BMs, pooled icORR was 15% (95%CI 6-28%) and icDCR was 47% (95% CI 36-57%). For PD-L1 >= 50% patients with any BMs, pooled icORR and icDCR were 68% (95%CI 57-80%) and 82% (95%CI 73-92%), respectively. Additionally, pooled systemic ORR and DCR for any BMs were 22% (95%CI 15-30%) and 41% (95%CI 18-67%), respectively. Patients with BMs had inferior progression-free survival (HR 1.19, 95%CI 1.07-1.33, P = 0.0016) and overall survival (HR 1.14, 95%CI 1.03-1.25, P = 0.011) when applying ICIs compared to those without BMs. However, no significant difference in systemic ORR between patients with and without BMs was observed (OR 0.94, 95%CI 0.72-1.20, P = 0.629). Conclusion ICIs may be clinically active in NSCLC patients with BMs. More effective treatments for BMs from NSCLC are needed.