Surgery for Control of Brain Metastases After Previous Checkpoint Inhibitor Immunotherapy

Background Despite reports on the efficacy of checkpoint inhibitor (CPI) immunotherapies for metastatic cancers, there are limited data on the effectiveness of surgery for brain metastases (BMs) that have progressed after previous CPI treatment. We sought to evaluate surgical outcomes for patients undergoing BM resection after failing CPI immunotherapy. Methods A single-center series of patients with BM that had progressed after previous CPI treatment and who underwent surgery was retrospectively reviewed. Outcomes of interest included local tumor progression, leptomeningeal dissemination, and overall survival. Cox proportional hazard models were applied to determine factors associated with outcomes of interest. Results Over a 16-year period, 26 patients underwent resection of 32 BMs at a median of 1.2 months (range, 2 days–41.1 months) from their last CPI dose. Median censored survival was 7.6 months from surgery and was shorter than the survival of patients without previous CPI exposure (21.9 months; log-rank P = 0.001). Four BMs had local central nervous system progression (16%), and 75% of procedures were associated with distant central nervous system progression within a median time of 3.3 months. Leptomeningeal disease developed after 33.3% of surgeries. Increased time from first BM diagnosis to surgery was associated with increased risk of leptomeningeal disease (hazard ratio by month, 1.07; 95% confidence interval, 1.01–1.14; P = 0.03). Conclusions Patients who require BM resection after previous CPI treatment have a poor overall prognosis compared with patients without previous CPI exposure. Although local control rates are acceptable, these patients are at high risk for developing distant progression and leptomeningeal disease postoperatively.