A Population-Level Real-World Analysis and Single-Center Validation of Melanoma Brain Metastasis Epidemiology following Dual-Agent Immunotherapy.
Publication Title
Clinical cancer research : an official journal of the American Association for Cancer Research
Document Type
Article
Publication Date
8-1-2025
Keywords
california; sjci; santa monica
Abstract
Purpose: Melanoma brain metastases (MBM) are common in advanced melanoma and linked to poor prognosis. Preventing MBM can improve survival and reduce morbidity. Although dual-agent immunotherapy (dIT) improves survival, its role in MBM prevention is unclear. We compared MBM incidence, overall survival (OS), and brain metastasis-free survival (BMFS) between dIT and single-agent immunotherapies.
Experimental design: A real-world multi-institutional database identified patients with melanoma without MBM at immunotherapy initiation. Patients were stratified by anti-CTLA4, anti-PD1, and combination anti-CTLA4/anti-PD1 (dIT) treatment. MBM incidences were measured within 5 years after immunotherapy initiation and compared with risk ratios (RR). In a complementary single-institution cohort, the median OS and BMFS were compared between dIT, anti-CTLA4, and anti-PD1 via log-rank tests and multivariate Cox proportional hazards models.
Results: TriNetX identified 8,287 patients receiving anti-CTLA (3,205), anti-PD1 (3,218), and dIT (1,864). MBM incidence was significantly lower in dIT (8.6%) and anti-PD1 (7.8%) versus anti-CTLA4 (12.2%) cohorts, with RR = 0.72 [95% confidence interval (CI), 0.61-0.86] and 0.63 (95% CI, 0.57-0.70), respectively. There was no significant difference in MBM incidence between anti-PD1 (7.8%) and dIT (8.6%; RR = 1.13; 95% CI, 0.93-1.36). In the single-institution analysis (n = 119), 2-year OS probabilities were 90%, 80%, and 95%, and 2-year BMFS probabilities were 72.7%, 80%, and 95.7%, in the dIT, anti-CTLA4, and anti-PD1 cohorts, respectively. DIT and anti-PD1 showed improved early-phase protection against MBM development. The number of metastatic sites was significantly associated with MBM development (HR = 2.36; 95% CI, 1.22-4.58; P = 0.01).
Conclusions: These findings highlight dIT's potential role in primary prophylaxis against MBM, with anti-PD1 as the likely workhorse agent. Prospective studies are warranted.
Area of Special Interest
Cancer
Area of Special Interest
Neurosciences (Brain & Spine)
Specialty/Research Institute
Neurosciences
Specialty/Research Institute
Oncology
Specialty/Research Institute
Epidemiology
DOI
10.1158/1078-0432.CCR-24-2681
