Stereotactic radiosurgery combined with immune checkpoint inhibition for the treatment of melanoma brain metastases is associated with high levels of extracranial disease control and survivorship - an abscopal effect?

Aims Melanoma brain metastases (MBM) are a common presentation to the neuro-oncology MDT. Stereotactic radiosurgery (SRS) is a highly effective treatment for cerebral metastases, with at least 70% control rates of individual metastases,[1] whilst immune checkpoint blockade has revolutionised the management of metastatic melanoma in recent years.[2] Recent studies have demonstrated that immune checkpoint inhibition alone also has activity in the brain, with MBM response rates of 50% or more.[3, 4] When MBM are treated with combination immunotherapy and SRS together, 12-month intracranial progression free survival (PFS) rates of 85% have been achieved.[4, 5] The aim of the current study was to evaluate the local control of MBM treated at our tertiary referral centre, which benefits from specialist neuro-radiology peer review of SRS contour volumes, and further to investigate whether overall survival is also improved, and what the mechanism of this may be. Method A retrospective analysis of all patients treated with SRS for brain metastases at our teriary SRS centre between June 2017 – January 2020 was performed. Inclusion criteria included patients treated for MBM, who received at least 2 doses of any combination of immune checkpoint inhibition concurrently with (defined as at the time of or commenced within 3 months of) SRS. The primary endpoints were the intracranial and extracranial response rates and survival rate at 12 months. Response was defined as complete response, partial response or stable disease. Secondary endpoints included the rate of imaging-defined radionecrosis, median lesional progression free survival (mPFSlesion), non-lesional intracranial PFS (mPFSintracranial), extracranial PFS (mPFSextracranial) and overall survival (mOS), measured from the start date of SRS to the date of event or censored at the start date of data collection. Kaplan-Meier curves and survival statistics were generated using SPSS v26. Results 33 MBM from 18 patients were identified. The median follow up was 25.8 months (minimum 12 months). Of the 18 patients: the median age was 60 (IQR 48 – 72); 17 (94%) patients were ECOG performance status 0-1; the median number of extracranial disease sites was 2 (pre-immunotherapy) and 1 (pre-SRS); the median duration of immunotherapy treatment was 17.6 (12.9 – 28.5) months, and the median number of metastases treated per patient was 2. Of the 33 metastases: 31 (94%) were supratentorial; 6 (18%) underwent prior neurosurgical resection; the median GTV volume (cc) of unresected metastases was 0.5cc (0.1 – 2.7), and 21 (64%) were treated with single fraction SRS. The median OS and PFS for all subtypes were not reached. The rates of OS, PFSlesion, PFSintracranial and PFSextracranial at 12 months were 93.9%, 87.9%, 81.8% & 75.8% respectively. Conclusion Our cohort of MBM patients appear to perform favourably when compared with the current literature. When compared to a recent extensive systematic review of modern management of MBM, our lesional control rate is as good as the weighted average of concurrent SRS + immunotherapy studies (87.9% vs 85.4% 12-month PFS), however we demonstrate a significantly improved 12-month OS rate (93.9% vs 52.8%) compared to the same (mOS of 15.8 – 17.4 months in other studies).[6,7] Our extra-lesional PFS is high and, compared to extracranial PFS rates from 51% at 6-months to 70.4% at 9-months in the literature,[3,4] our 75.8% control at 12 months suggests that extracranial control could drive the OS benefit. This suggests a benefit of SRS beyond the local control of MBM and questions whether patients without brain metastases may benefit from body SABR to extracranial metastases, to elicit a similar, potentially abscopal type effect.