ROADS

Background

Randomised, open-label, non-inferiority phase 3 trial (NCT04365374) at 32 US centres in patients with a newly diagnosed brain metastasis requiring resection. N=230 randomised (115 per arm); mITT n=204. Tile-based RT (TBRT; cesium-131 collagen tile implanted at resection) delivers immediate focal radiation vs standard post-operative SRT.

Results

Interventions and follow up: Arm A: resection + intraoperative Cs-131 collagen tile implant (R+TBRT)
Arm B: resection + post-operative SRT (R+SRT)
Non-resected metastases received post-operative SRT in both arms.
Primary endpoint: co-primary: time to surgical bed recurrence (SBR) and SB-RFS (hierarchical testing)
mFollow up: 12.9 mo
Results: Time to SBR: median NR vs 17.4 mo; HR 0.06 (95% CI 0.01–0.46), P=.0070
SB-RFS: median NR vs 10.9 mo; HR 0.48 (95% CI 0.30–0.76), P=.0021
OS: NR
Non-inferiority: met; superiority also shown

Adverse events

Any adverse event: 79.0% (95% CI 70.0–86.4) vs 80.7% (95% CI 70.6–88.6)
Grade ≥3 / radiation necrosis / discontinuation: NR

Conclusions

Resection plus Cs-131 tile brachytherapy met non-inferiority and superiority for surgical-bed control vs post-operative SRT, with comparable overall adverse-event rates. It is a practical alternative that delivers adjuvant RT at the time of surgery.

Key Limitations

Open-label. Short median follow-up (12.9 mo) with few SBR events; HR 0.06 has a very wide CI. Analysis on mITT excluding 26 of 230 randomised pts. No OS result; radiation necrosis and late toxicity not detailed in abstract. Co-primary endpoints concern local control, not distant or intracranial progression, and SRT comparator delivered in a separate post-operative step.

Clinical Context

Post-operative SRT is the current standard of care for resected brain metastases (NCCN, ASCO-SNO-ASTRO guidance); pre-operative SRS is an emerging alternative. ROADS supports TBRT as an alternative, presented at ASCO 2026 per the ASCO Post; I am not claiming a guideline update.

References