Radionecrosis (RN) in patients with brain metastases treated with stereotactic radiosurgery (SRS) and immunotherapy

Abstract

Objectives Limited data exist regarding radionecrosis (RN) rates when patients receive immunotherapy (IT) and SRS for brain metastases. This study assesses the influence of such treatments on the rate of RN. Methods We retrospectively reviewed 352 lesions from 105 patients with metastatic melanoma or NSCLC treated with SRS and IT from 2012 to 2018. Lesions were excluded from analysis if patients had received WBRT or prior GK to the same lesion, if RN occurred before IT, or if IT had been discontinued >6 months pre-SRS or initiated >1 year post-SRS. IT was delivered concurrently (±30 days of SRS) or sequentially. Overall survival and RN rates were assessed with Kaplan-Meier analysis. Univariate analysis and multivariate analysis were performed to identify characteristics predicting RN. Results Of 195 lesions from 63 patients included in analysis, the median prescription dose, IDL, lesion volume, and maximum tumor dimension (MTD) were 19 Gy, 50%, 0.15 cc and 0.8 cm, respectively. RN rates at 1, 2, and 3 years were 7.3%, 10.4% and 10.4%. On UVA, RN risk increased with, isodose volume (IDV), MTD, and tumor volume (TV) whereas conformity index was associated with a trend toward decreased RN risk. Two-year RN rates increased with TV ≥ 0.3 cc (16% vs 1.1% p = 0.001), MTD ≥ 1.3 cm (19.1% vs 1.8% p < 0.003), and IDV ≥ 1.5 cc (19.6% vs 1.7% p = 0.001). Concurrent vs sequential timing of IT did not predict for RN. Conclusions Patients who received IT and SRS had acceptably low rates of RN. Timing of IT did not predict for RN. Further investigation is warranted to define RN risk with combined SRS and IT.