Brain metastases are a common and severe complication potentially leading to death in patients with metastatic melanoma. Immunotherapy and targeted therapy have significantly improved progression-free survival (PFS) and overall survival (OS) in patients with advanced melanoma. Few studies focus on patients with central nervous system (CNS) metastases, and these patients are often excluded and have a poor prognosis. It has been suggested that immunotherapy could reduce the incidence of brain metastases. We tested this hypothesis in a retrospective bicentric study. We performed a retrospective, bicentric descriptive analysis on a cohort of 293 patients treated for metastatic melanoma between May 2014 and October 2017 (Toulouse, N = 202; Limoges, N = 91). Patients with brain metastasis at diagnosis were excluded from the analysis. Patients were separated into two groups according to the first line of treatment: immunotherapy [immune checkpoint inhibitor (ICI)] vs other and anti-PD-1 vs other. The primary endpoint was the cumulative incidence of brain metastases, and secondary endpoints were OS and PFS. At 12 months, the cumulative incidence of brain metastases was 13.78% in the ICI group [95% confidence interval (CI) 9.14-19.36] and 27.26% in the other group (95% CI 19.38-35.71), P = 0.004. The cumulative incidence was 9.49% in the anti-PD-1 group (95% CI 5.43-14.90) vs 30.11% in the other group (95% CI 22.59-37.97), P < 0.0001. In multivariable analysis (model with 277 patients), anti-PD-1 reduced the risk of brain metastases by almost 70% (hazard ratio = 0.29, 95% CI 0.15-0.56, P < 0.0001). The use of ICI (anti-PD-1/PD-L1) in advanced melanomas without initial brain metastasis shows a protective effect and prevents their occurrence.