Comparative safety and survival outcomes of immune checkpoint inhibitors in brain metastases from lung cancer: a propensity-matched cohort study

PURPOSE: Brain metastases from lung cancer are associated with high morbidity and mortality. Immune checkpoint inhibitors (ICIs), including PD-1/PD-L1 and CTLA-4 inhibitors, are increasingly used, but their safety profile in this population remains poorly defined. We aim at exploring their safety and outcomes in this population.

METHODS: We used TriNetX to identify patients with diagnoses of lung cancer and secondary brain metastases treated at the Mount Sinai Health System. Three cohorts were formed: PD-1/PD-L1 inhibitors (n = 330), CTLA-4 inhibitors (n = 30), and no immunotherapy (n = 800). Propensity score matching was applied to create balanced subgroups for each comparison. Outcomes assessed included hematologic toxicity, respiratory and neurologic complications, and survival, with a 3-year follow-up window. Risk ratios and Kaplan-Meier survival analysis were performed.

RESULTS: PD-1/PD-L1 inhibitors were associated with a significantly higher incidence of anemia (40.0% vs. 13.3%; RR 3.0, OR 4.3, p < 0.001) and shorter anemia-free survival (HR 3.44, 95% CI 2.36-5.02, p < 0.001). CTLA-4 inhibitors were associated with a higher risk of seizures (33.3% vs. 10%; RR 3.33, OR 4.67, p = 0.04), anemia (66.7% vs. 33.3%, p = 0.01), and respiratory complications including intubation (26.7% vs. 10%, p = 0.04) and respiratory failure (36.7% vs. 13.3%, p = 0.01). Kaplan-Meier analysis showed a survival probability of 46.7% vs. 82.6% at 3 years for CTLA-4 (p < 0.01).

CONCLUSION: Checkpoint inhibitors are associated with increased hematologic, neurologic, and respiratory complications in patients with brain metastases from lung cancer. CTLA-4 blockade appears to carry higher toxicity than PD-1/PD-L1 inhibitors. These findings show the need for careful patient selection and close monitoring when initiating ICIs in this population.