# Outcomes after SRS and ipilimumab plus nivolumab for melanoma brain metastases following prior immune checkpoint inhibitor or targeted therapy

**Authors:** Ian Messing, Lauren Linkowski, Matthew D Riina, Melanie Berger, Jonathan Baron, Xingmei Wang, Michael Kallan, Harper G Hubbeling, Robert A Lustig, Jay F Dorsey, Goldie Kurtz, Suneel Nagda, John N Lukens, Ahron Flowers, John Y K Lee, Nduka M Amankulor, Christina Jackson, Suyash Mohan, Ravi K Amaravadi, Lynn M Schuchter, Alexander C Huang, Tara C Mitchell, Michelle Alonso-Basanta, Emily S Lebow

PMC · DOI: 10.1093/oncolo/oyag043 · 2026-02-16

## TL;DR

This study finds that combining stereotactic radiosurgery with ipilimumab and nivolumab offers good brain cancer control, but prior treatments worsen outcomes.

## Contribution

First real-world evaluation of SRS plus ipi/nivo for melanoma brain metastases after prior therapies.

## Key findings

- Median overall survival was 24 months with better outcomes for ICI-naive patients.
- Prior ICI or BRAF/MEKi therapy predicted worse survival and progression-free survival.
- Local and distant progression rates were 11% and 49% at 24 months, respectively.

## Abstract

Melanoma brain metastases (BM) carry high morbidity and mortality despite advances in systemic therapy. Combined immune checkpoint inhibition (ICI) with ipilimumab and nivolumab (ipi/nivo) demonstrates intracranial activity, but the influence of prior systemic therapy exposure is poorly defined. This is the first real-world study evaluating outcomes of melanoma BM treated with stereotactic radiosurgery (SRS) and concurrent ipi/nivo, focusing on the impact of prior ICI or targeted therapy.

We retrospectively analyzed 68 patients with 413 melanoma BM treated with concurrent SRS and ipi/nivo from 2015 to 2025. Primary endpoints were overall survival (OS) and intracranial progression-free survival (iPFS). Secondary endpoints included local and distant control, radionecrosis, and leptomeningeal disease. Univariable and multivariable Cox models identified predictors of outcome.

Median OS was 24.0 months (12- and 24-month OS: 64% and 50%). ICI-naive patients had longer OS (50.5 vs. 17.6 months; P = 0.007) and iPFS (15.1 vs. 5.9 months) than those with prior ICI. On multivariable analysis, prior ICI (HR 2.23, 95% confidence interval [CI] 1.13-4.41), prior BRAF/MEKi (HR 2.26, 95% CI 1.01-5.04), and ≥11 SRS-treated lesions (HR 3.22, 95% CI 1.43-7.21) predicted worse outcomes, while higher graded prognostic assessment (GPA) favored OS (HR 0.46, 95% CI 0.29-0.75). At 24 months, local progression was 11%, distant 49%, radionecrosis 7%, and leptomeningeal disease 4%.

Concurrent SRS with ipi/nivo provides durable intracranial control with low toxicity. Patients with prior ICI or targeted therapy represent a high-risk subgroup with poorer outcomes, supporting exploration of intensified or novel strategies.

## Linked entities

- **Diseases:** melanoma (MONDO:0005105)

## Full-text entities

- **Genes:** BRAF (B-Raf proto-oncogene, serine/threonine kinase) [NCBI Gene 673] {aka B-RAF1, B-raf, BRAF-1, BRAF1, NS7, RAFB1}
- **Diseases:** leptomeningeal disease (MESH:D008577), melanoma (MESH:D008545), BM (MESH:D001932), toxicity (MESH:D064420)
- **Chemicals:** MEKi (-), Ipilimumab (MESH:D000074324), Nivolumab (MESH:D000077594)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Figures

2 figures with captions in the complete paper: https://tomesphere.com/paper/PMC13010309/full.md

---
Source: https://tomesphere.com/paper/PMC13010309