Neurologically relevant immune-related adverse events associated with immunotherapy for lung cancer in clinical trials: a systematic review and meta-analysis

Abstract Background Immune checkpoint inhibitors (ICIs) are crucial in lung cancer treatment but pose risks of immune-related adverse events (irAEs), including neurological manifestations. To address conceptual ambiguity, we conducted a systematic review and meta-analysis focusing on "neurologically relevant irAEs" (n-irAEs). Methods This term encompasses a broad spectrum of clinically reported neuro-related toxicities, including primary neurological disorders and secondary manifestations with neuroanatomical or clinical relevance (e.g., hypophysitis), rather than being limited to strictly defined primary neurological diseases. We included prospective clinical trials assessing immunotherapy safety in lung cancer patients with at least one arm utilizing ICIs and reporting treatment-related n-irAEs. The primary outcome was n-irAE incidence and rate ratio, categorized into central and peripheral subtypes. Results Compared to chemotherapy, PD-1 therapy showed a higher incidence of central n-irAEs, though this finding was not robust in sensitivity analysis, alongside lower rates of peripheral n‑irAEs. Single ICI had significantly lower central and overall n-irAE incidence compared to dual ICIs. However, combination therapy (ICI + Chemotherapy) increased n-irAE rates compared to chemotherapy alone; PD-1 + Chemotherapy had a significantly higher incidence of central n-irAEs, while PD-L1 + Chemotherapy showed higher peripheral n-irAE rates. Peripheral neuropathy was the most common Grade ≥ 3 n-irAE. Conclusion This study provides comprehensive insights into n-irAEs in lung cancer. While the overall incidence of n-irAEs especially peripheral subtypes is lower with ICIs than chemotherapy, the risk of central n-irAEs is notably higher, although this finding was attenuated and no longer statistically significant in a sensitivity analysis restricted to classical neurological events. These findings emphasize the need for vigilant monitoring of neuro-related symptoms to optimize the management of lung cancer patients undergoing ICI therapy. Systematic review registration : PROSPERO registration number CRD42023460190.