Implanted Radiation Tiles Outperform Standard Therapy for Brain Metastases

A surgical team at MD Anderson Cancer Center has been sitting with a problem for years: patients who need radiation after brain tumor removal often don't get it on time, or don't get it at all. The delay matters. Cancer grows back in the surgical cavity. Systemic treatment gets pushed back. About one in five patients assigned to radiation therapy in the past simply never showed up for their scheduled sessions—delayed by insurance snags, weather, rehospitalization, or the simple friction of life. Jeffrey Weinberg and his colleagues wondered if there was a way to eliminate that gap entirely.

Their answer was to implant the radiation at the moment of surgery itself. The ROADS trial, presented at the 2026 ASCO Annual Meeting, tested a technique called tile-based radiation therapy: small collagen tiles embedded with four cesium-131 seeds, each tile measuring 2 by 2 centimeters, placed directly into the surgical cavity immediately after the tumor is removed. The tiles deliver a highly concentrated dose—100 to 120 Gy to the cavity wall, with about 90 percent of that dose delivered within five weeks. The radiation is focal and limited in penetration, designed to sterilize the surgical bed without flooding the surrounding brain.

The trial enrolled 230 patients across 32 sites with newly diagnosed brain metastases measuring 2 to 7 centimeters. Of those, 204 met the criteria for analysis: 103 received the implanted tiles, and 101 received standard stereotactic radiation therapy delivered two to four weeks after surgery. The results were stark. Surgical bed recurrence—cancer returning in the treated area—occurred in just 1 percent of the tile group versus 11.9 percent of the standard therapy group. The median time to recurrence was not reached in the tile arm, meaning most patients never experienced it during follow-up, compared to 17.4 months in the standard group. At 12 months, cumulative recurrence was 1.3 percent with tiles versus 15.4 percent with standard therapy.

But the finding that surprised even the investigators was overall survival. Median overall survival more than doubled: 42.5 months with tiles versus 17.6 months with standard radiation. At 24 months, 61.7 percent of tile patients were alive compared to 35.7 percent in the standard group. Weinberg acknowledged the trial was not powered to detect an overall survival difference, yet there it was. The mechanism remains unclear—ongoing analysis is attempting to explain why implanted radiation would extend life so substantially when most patients with brain metastases ultimately die from systemic disease, not brain recurrence.

The practical advantage was equally important. By implanting the radiation at surgery, the technique guaranteed that patients received treatment. There was no appointment to miss, no insurance delay, no weather event that could derail the plan. Among the 101 patients assigned to standard radiation, 18 never received it. Among the 103 assigned to tiles, all received the treatment. Weinberg noted that this "one and done" approach also addressed a clinical reality: distinguishing between true cancer recurrence and radiation necrosis on follow-up MRI is notoriously difficult. When the investigators analyzed the composite endpoint of either recurrence or necrosis, tiles still showed significant benefit.

David Schiff, a neuro-oncologist at the University of Virginia and an ASCO expert in central nervous system tumors, offered cautious enthusiasm. He noted the biologic plausibility—larger metastases are often less responsive to standard radiosurgery, and immediate, focal radiation makes sense. But he raised legitimate concerns: the open-label design meant both doctors and patients knew which treatment they were receiving, which can bias perception of outcomes. About 11 percent of randomized patients were excluded from the final analysis, and such postrandomization exclusions can inadvertently favor the new therapy. Most puzzling to Schiff was the overall survival benefit itself. Prior trials comparing postoperative radiation to observation showed no survival advantage, so why would this one? The answer may lie in ongoing analyses examining the interaction between the implanted radiation and systemic therapies, or in subtle differences in patient selection or follow-up.

Despite these caveats, the results are substantial enough that neurosurgeons are already reconsidering standard practice. If the findings hold, tile-based radiation could become the default approach for patients undergoing brain metastasis resection. A separate trial, called BRIDGES, is now testing the same technique in patients with glioblastoma, a more aggressive primary brain tumor. The tiles themselves are FDA-cleared, and the question now is not whether they work, but how broadly they should be deployed and what additional benefits or risks emerge as the technique moves into wider use.