P17.17.A RADIATION NECROSIS INDUCED BY STEREOTACTIC RADIOTHERAPY. LONG TERM IMPACT OF BEVACIZUMAB

Abstract BACKGROUND Brain radiation necrosis (RN) is a frequent and severe complication of stereotactic radiotherapy, which is an effective and common treatment for brain metastases (BM). There is no standard treatment for RN and Bevacizumab (Bev) has emerged as a potential treatment, due to its anti-angiogenic properties. Data regarding the long-term efficacy of Bev are scarce. We performed a retrospective study to evaluate the immediate and long-term impact of Bev on patients with RN following stereotactic therapy. MATERIAL AND METHODS We reviewed all adult patients treated with Bev between 01/2018 and 01/2023 in our institution. Only patients treated for RN following either stereotactic radiosurgery (SRS) or fractionated stereotactic radiotherapy (FSR) for BM were included in the study. Clinical and radiographic data were collected at baseline, immediately after completion of Bev, 6, 12, and 24-months post-treatment. All MRI tests were reviewed to assess response rate using RANO criteria. Changes in volume were evaluated for each lesion using BrainLab software. RESULTS Twenty-four patients were included in the study, among them 6 patients presented with more than one RN lesion. Subsequently a total of 32 distinct target lesions were analyzed. Median age at diagnosis was 64 (range 56-70). Median time of follow-up was 14 months (range 11.2-25.2). At first evaluation (median time 3.5 months after Bev), we observed 67% (16/24) response rate according to RANO criteria (either partial or complete response). 25% (6/24) had stable disease and 8% (2/24) experienced progressive disease (PD). A reduction of at least 50% of the lesion volume was observed in 84% of the patients, 58% of patients improved clinically, and steroid dose was reduced. A year after Bev treatment, 15 patients were available for follow up (8 patients were deceased due to systemic progression and 1 was lost to follow-up). Three patients continued to improve and 5 remained stable. Two years after Bev treatment, among 7 patients available for evaluation, 1 patient continued to improve while 6 remained stable, 3 of them no longer requiring steroid therapy. During follow-up, 12 patients experienced regrowth of the RN (PD according to RANO). The median time to PD was 8 months (range 5 to 9 months). Six of them underwent Bev re-challenge therapy, but only two displayed a response. CONCLUSION Our study supports Bev as an effective and sustainable treatment for brain RN following stereotactic radiothe