Final results of a randomized, controlled, phase 3 trial comparing resection and post-operative stereotactic radiation versus resection and cesium-131 tile-based radiation for treatment of newly diagnosed brain metastases.

Abstract

LBA2000 Background: Post-operative stereotactic radiation (SRT) is the standard to reduce recurrence of resected brain metastases (BM). Cesium-131 tile-based radiation therapy (TBRT) (GT Medical Technologies) delivers focal radiation (RT) immediately upon resection and may offer logistical and therapeutic advantages. ROADS (NCT04365374) compared TBRT vs SRT in patients with newly diagnosed resectable BM. Methods: In this open label phase 3 trial across 32 US centers, patients requiring resection of one newly diagnosed BM were randomized 1:1 to surgery + SRT or surgery + TBRT. Up to 5 additional unresected BM were allowed and treated with SRT in both arms. Co-primary endpoints were time to surgical bed recurrence (SBR; independently centrally reviewed) and surgical bed recurrence free survival (SB-RFS), analyzed using Cox proportional hazards models with stratification factors as covariates. Multiplicity was controlled by hierarchical testing. Key secondary endpoints were overall survival (OS), leptomeningeal disease (LMD; independently centrally reviewed), radiation necrosis (RN) and treatment related adverse events (TRAEs). Exploratory endpoints were time to SBR or RN and time to total cranial management (TTCM, time from surgery to completion of all RT). Analyses used the modified intent to treat (mITT) population, defined as patients who underwent surgery and had follow up information. Results: 230 patients with balanced patient and tumor characteristics were randomized (115 per arm). 204 patients (101 SRT, 103 TBRT) comprised the mITT population. Median follow-up time was 12.9 mo (IQR:5.9-22.8 months). SBR occurred in 11.9% (SRT) vs 1.0% (TBRT). Median time to SBR was 17.4 mo in SRT and not reached in TBRT (HR: 0.06, 95% CI: 0.01-0.46, p=0.007). SB-RFS was significantly improved with TBRT: median SB-RFS was 10.9 mo in SRT vs not reached in TBRT (HR: 0.48, 95% CI: 0.30-0.76, p=0.002). OS was significantly improved with TBRT (HR: 0.59, 95% CI: 0.37-0.96; p=0.032); estimated 24-mo survival was 35.7% (SRT) vs 61.7% (TBRT). LMD occurred in 3.0% (SRT) vs 9.7% (TBRT) (p=0.146). The 24-mo probability of being alive without LMD was 35.9% (SRT) vs 57.7% (TBRT) (HR: 0.68, 95% CI: 0.43-1.08, p=0.102). RN occurred in 6.9% of patients (SRT) vs 7.8% (TBRT) (p=0.495). TBRT extended time to SBR or RN: median time was 18.3 mo in SRT vs not reached in TBRT (HR: 0.28, 95% CI: 0.12-0.66, p=0.004). Median TTCM was 30 days (SRT) vs 1 day (TBRT) (p<0.001). Grade ≥3 TRAEs were 19.3% (SRT) vs 18.1% (TBRT). Conclusions: TBRT improved surgical bed control and OS compared to SRT for patients with a newly diagnosed resectable BM. Safety was equivalent. Completing total cranial treatment faster with TBRT, coupled with fewer SBR occurrences may lead to fewer gaps in systemic therapy perhaps resulting in the observed OS benefit. Clinical trial information: NCT04365374 .