Os06.1.a Intraoperative Radiotherapy After Resection of Brain Metastases (Intramet)- Final Results of a Prospective, Single-Arm, Open-Label Phase 2 Trial

BACKGROUND Brain metastases represent the most common intracranial malignancy. The standard therapeutic approach for space-occupying lesions is neurosurgical resection followed by adjuvant radiotherapy of the resection cavity. We here present the final analysis of the INTRAMET trial (NCT03226483), investigating efficacy and safety of intraoperative radiotherapy (IORT) following the resection of brain metastases. MATERIAL AND METHODS Patients with a Karnofsky score of ≥50 and aged ≥18 years old were eligible for inclusion in the study. The histological diagnosis was confirmed by frozen section. The IORT procedure was performed using a mobile irradiation unit that emits low-energy X-rays. The delivered dose was 30 Gy to the surface of the resection cavity. The primary endpoint of the study was defined as local tumor control (LCR), with secondary endpoints including regional brain control (RBC), overall survival (OS), time to further anticancer therapy, and adverse events (AEs). RESULTS The study population comprised 35 patients (54.3% male). Mean follow-up was 25.7 (0.8-64.5) months. Mean age was 65 (45-85)years. At the time of analysis, 51.4% of patients were alive. The most prevalent primary tumor site was lung (68.6%). LCR was 94.3% (95% CI 82.9-98.8%), while RCR was 57.1% (95% CI 40.7-72.4%). Leptomeningeal progression occurred in 8.6% of patients, with all cases distant from the IORT-treated metastasis. Median OS was 43.6 months (95% CI 8.8-78.4). Seven patients (20%) received salvage whole brain RT, with a median initiation time of 147 (20-601) days. A significant survival benefit was observed for patients who did not require WBRT (p=0.027). With respect to OS, no statistically significant survival differences were observed according to primary histology (p=0.618), immunotherapy (p=0.928), seizures at baseline (p=0.169), Karnofsky performance score (p=0.056), or radionecrosis (RN) appearance (p=0.214). The initiation of anticancer therapy occurred at a median of 45.0 days (95% CI 35.1-54.8). No grade 4-5 AEs related to IORT were observed. RN was observed in 20% of patients (n = 7). Of these, five patients exhibited grade 1 RN, one patient exhibited grade 2 RN, and one patient exhibited grade 3 RN. No risk factors could be identified that were associated with RN. Postoperative seizures were observed in 10 patients (28.6%), with eloquent localization identified as the sole predictive factor (p=0.008). CONCLUSION In the context of BM resection, IORT has been shown to yield optimal local control rates and a toxicity profile analogous to that observed in postoperative SRS trials. The initiation of subsequent systemic treatments is facilitated by IORT, which has the potential to reduce the duration of cancer treatment.