Concurrent Radiosurgery and Immunotherapy is Associated with Improved Intracranial Tumor Control in Patients with Metastatic Melanoma

Abstract

The anti-CTLA-4 and anti-programmed cell death-1 (PD-1) antibody therapies have significantly improved survival of patients with metastatic melanoma. However, little is known about the interaction of immunotherapy and stereotactic radiosurgery (SRS) for brain metastases and how the combination may affect toxicity and intracranial tumor control. We retrospectively reviewed 26 patients with 106 lesions who received immunotherapy and SRS for brain metastases from 2011-2016. 13 patients received Ipilimumab alone, 2 patients received Pembrolizumab alone, and the remaining patients received combination therapy. Treatment response was evaluated based on imaging and clinical follow-up. Treatment toxicities were recorded. To study the impact of timing and sequencing of immunotherapy on SRS, we categorized the lesions into two groups. The concurrent group included patients treated with SRS within 30 days of immunotherapy. The non-concurrent group included patients treated with SRS at least 30 days before or after immunotherapy. Kaplan-Meier methods were used to determine overall survival (OS), local control (LC) and regional control (RC). OS was analyzed per patient, from the date of diagnosis of brain metastasis to date of death or last patient contact. Local and regional recurrence-free duration were analyzed per treated lesion. Regional recurrence was defined as intracranial failure outside of the treatment field. The median follow-up was 14.5 (range 3 to 54) months. The median age was 59 (range 19-92) years. The median KPS and melanoma specific GPA were 80 and 2, respectively. Median survival was 26.1 months. The 1, 2 and 3 yr OS rates were 67%, 62%, and 45% respectively. The 1, 2 and 3 yr LC rates were 100%, 90% and 84%. The 1, 2, and 3 yr RC rates were 37%, 18% and 5%. 2.8% of the lesions developed a local recurrence. There was no difference in LC in SRS lesions treated with concurrent vs. non-concurrent immunotherapy. However, there was a significant improvement in regional recurrence free duration in lesions treated concurrently with immunotherapy compared to those treated without concurrent immunotherapy. The median regional recurrence-free duration was 15 mo vs 2 mo, P=0.0067. Treatment related adverse effects were noted in 35% of the SRS treated lesions, including focal neurologic deficits, seizure, headaches, and steroid dependence. No grade 4 or 5 toxicities were noted. SRS delivered concurrently with either single or dual check point blockade therapy is safe and well tolerated. The increased regional control following concurrent SRS and immunotherapy suggests radiotherapy may augment the regional effects of immunotherapy. Our findings may influence clinical decisions on how to optimize intracranial disease control to further improve treatment outcomes.