Two Phase 3 Trials Show Radiation Advances for Brain Metastases
Two phase 3 trials tested radiation approaches for brain metastases. Tile-based brachytherapy cut 1-year recurrence to 1.3% and improved median overall survival. Stereotactic radiation reduced symptom burden versus whole brain radiation.
Researchers at The University of Texas MD Anderson Cancer Center reported that implanting collagen tiles during brain surgery to deliver targeted radiation therapy — tile-based radiation therapy (TBRT) — dramatically improved tumor control, lowered the risk of recurrence and improved overall survival compared with current standard of care for patients with newly diagnosed brain metastases in need of surgical resection. In a separate phase 3 trial, stereotactic radiation focusing on individual brain metastases reduced symptom burden and interference with daily functioning compared with hippocampal-avoidance whole brain radiation in patients with 5 to 20 brain metastases.
The ROADS trial, the first randomized controlled Phase 3 trial comparing cesium-131 collagen tile-based radiation therapy against standard-of-care postoperative stereotactic radiation therapy (SRT), was presented at the 2026 American Society of Clinical Oncology (ASCO) Annual Meeting. After one year, patients treated with TBRT had a 1.3% rate of recurrence at the surgical site compared to 15.4% of patients in the SRT arm. Median overall survival, a key secondary endpoint of the trial, was 42.5 months with TBRT – more than double the 17.6 months seen with standard SRT.
TBRT uses a Food and Drug Administration (FDA)-cleared low-dose brachytherapy device developed by GT Medical Technologies, Inc. The small tiles, about the size of a postage stamp, contain evenly spaced seeds filled with cesium-131 embedded in a collagen matrix that essentially gets 'wallpapered' to the surrounding cavity left after surgery. This ensures that radiation is evenly distributed across the cavity surface, where most remaining microscopic tumor cells are located. The seeds disperse low-dose therapeutic radiation over the course of several weeks while limiting exposure to healthy tissue.
There were no differences in serious treatment-related side effects between TBRT and SRT, and the rate of radiation necrosis, an important late risk for patients treated with radiation for brain metastases, was nearly identical between the two groups. Patients receiving TBRT were able to complete cranial radiation faster, most in just one day, compared with a median of 32 days for those needing to schedule postoperative SRT.
In the second trial, a Phase 3, open-label, randomized clinical trial conducted at four hospital-based centers in the US, 196 patients with 5 to 20 brain metastases were randomized 1:1 to stereotactic radiation or hippocampal-avoidance whole brain radiation with memantine; 173 were analyzed for the primary end point. The stereotactic radiation group had greater improvement in mean symptom severity and interference score (MD Anderson Symptom Inventory–Brain Tumor change: –0.32) vs hippocampal-avoidance whole brain radiation (change: 0.74); between-group mean difference: –1.06. Overall survival was similar (median 8.3 vs 8.5 months). Stereotactic radiation improved functional independence, performance status, and cognitive test scores over time. New brain metastases developed more frequently after stereotactic radiation; however, the need for subsequent whole brain radiation was low. Related serious adverse events occurred in 12% of the stereotactic radiation group and 13% of the hippocampal-avoidance whole brain radiation group, most commonly fatigue.
The ROADS trial investigators suggested that immediate placement of radioactive tiles during surgery could become the new standard of care for patients with brain metastases requiring surgery. The second trial's results suggest that stereotactic radiation may reduce symptom burden and preserve daily function better than hippocampal-avoidance whole brain radiation for patients with 5 to 20 brain metastases, with a similar risk of serious adverse events.