Tile-based radiation therapy (TBRT) implanted at the time of resection outperformed standard postoperative stereotactic radiation therapy (SRT) in the phase III Radiation One and Done Study (ROADS), improving surgical bed control and overall survival while maintaining a comparable safety profile, investigators reported at the 2026 ASCO Annual Meeting.¹

“Patients undergoing tumor resection currently face an inherent gap in care: Radiation is delayed for weeks after surgery, which not only can result in worse local control, but can also delay systemic treatment,” said presenting author Jeffrey S. Weinberg, MD, FAANS, FACS, Professor of Neurosurgery, Deputy Chair, and Vice Chair of Clinical Operations in the Department of Neurosurgery, The University of Texas MD Anderson Cancer Center, Houston, in an ASCO press release. “Patients needed a better option that could result in better outcomes and easier logistics.”

Not only does this technique seem to work better, but there is the added benefit of guaranteeing that a patient receives the radiation….

— JEFFREY S. WEINBERG, MD, FAANS, FACS

Tweet this quote

ROADS evaluated an approach in which 2 × 2-cm collagen-based tiles embedded with four titanium-encapsulated cesium-131 brachytherapy seeds are implanted into the surgical cavity immediately after tumor resection. The system delivers a highly focal, limited-penetration radiation dose of approximately 100 to 120 Gy to the cavity wall (80 Gy at 3 mm and 60 Gy at 5 mm), with about 90% delivered within 5 weeks. In this final analysis, the U.S. Food and Drug Administration–cleared therapy was associated with median time to surgical bed recurrence and surgical bed recurrence–free survival that were not reached during follow-up, as well as a more than twofold increase in median overall survival compared with SRT.

“If this proof of concept holds, [TBRT] would largely supplant cavity radiosurgery for this indication,” said David Schiff, MD, Harrison Distinguished Teaching Professor of Neurology, Neurological Surgery, and Medicine at the University of Virginia School of Medicine, Charlottesville, and an ASCO Expert in tumors of the central nervous system, in the press release.

Study Details

The noninferiority randomized controlled trial enrolled patients from 32 sites with a newly diagnosed surgical brain metastasis (index lesion measuring 2–7 cm; ≤ 5 nonindex lesions allowed). Of 230 randomized (1:1) patients, 204 met the modified intention-to-treat (mITT) population criteria (ie, maximal safe resection, pathologically confirmed metastasis, and available follow-up data), including 101 assigned to postoperative SRT (delivered 2 to 4 weeks after surgery) and 103 to TBRT; 18 patients in the SRT arm ultimately did not receive treatment. All new or nonindex lesions were treated with SRT in both arms.

The primary endpoints were time to surgical bed recurrence and surgical bed recurrence–free survival. Secondary endpoints included overall survival, functional status, quality of life, adverse events, leptomeningeal disease, radiation necrosis, and factors contributing to delays in SRT. All reported surgical bed recurrence and leptomeningeal disease events were centrally reviewed by two independent neuroradiologists.

Median follow-up was 12.9 months—“about 2 months [longer] than most similar practice-changing trials,” according to Dr. Weinberg. In the SRT arm, 7.8% of patients received one fraction and 92.2% received three to five fractions. Baseline disease extent and other characteristics appeared to be balanced between the treatment arms.

Surgical Bed Recurrence

Surgical bed recurrence was documented in 1.0% of the TBRT arm vs 11.9% of the SRT arm.

Median time to surgical bed recurrence was not reached with TBRT vs 17.4 months with SRT (hazard ratio [HR] = 0.06; P = .007). The 12-month cumulative incidence of surgical bed recurrence was 1.3% with TBRT vs 15.4% with SRT (HR = 0.07; P = .012).

Surgical bed recurrence–free survival was also found to be significantly improved with TBRT compared with SRT, with median not reached vs 10.9 months, respectively (HR = 0.48; P = .0021).

Based on a sensitivity analysis excluding patients who were assigned to but did not undergo SRT, the superiority of TBRT was maintained for both primary endpoints. Dr. Weinberg noted that, “Not only does this technique seem to work better, but there is the added benefit of guaranteeing that a patient receives the radiation, overcoming the risk of not returning in a timely manner for their radiation,” a challenge observed in approximately 20% of patients assigned to SRT in this trial who ultimately did not receive the planned treatment, consistent with findings from previous trials.

Overall Survival and Other Secondary Endpoints

Median overall survival more than doubled in the TBRT arm—42.5 vs 17.6 months with SRT (HR = 0.59; P = .032). The estimated 24-month overall survival rates were 61.7% and 35.7%, respectively.

“We were quite surprised, and we’re extremely excited by this,” Dr. Weinberg remarked, noting that the trial was not powered for overall survival. “This is the impetus for current analysis, which is ongoing to define the reason for this encouraging finding.”

Other secondary endpoints—functional status, quality of life, distant brain failure, adverse events, leptomeningeal disease, and radiation necrosis—appeared similar between the arms, he reported. Prolonged recovery, insurance issues, weather events, patient or investigator decisions, and rehospitalization were among the most frequently cited reasons for delays in SRT, with a median time from resection of 27 days.

Dr. Weinberg noted that surgical bed recurrence and radiation necrosis are radiographic findings and therefore often difficult to distinguish, describing this as “a clinical scenario faced by most clinicians dealing with patients with metastatic lesions.” To address this challenge, the investigators analyzed time to either surgical bed recurrence or radiation necrosis as a composite endpoint; TBRT was found to significantly prolong this endpoint compared with SRT (median, not reached vs 18.3 months; HR = 0.28; P = .004). The benefit appeared to be consistent in cumulative incidence analysis, which also showed a reduced risk of surgical bed recurrence or radiation necrosis over time (HR = 0.36; P = .016).

Contextualizing Clinical Outcomes

A key limitation of the study is its open-label design. As Dr. Weinberg noted, “there are many nuances and complications that make it somewhat infeasible to conduct a blinded or placebo-controlled trial,” and, to the investigators’ knowledge, no such radiation trials exist in oncology.

Despite this limitation, he concluded that among patients who underwent resection of a new brain metastasis, TBRT compared with standard-of-care postoperative SRT was associated with a lower likelihood of recurrence in the treated area, fewer negative changes on follow-up brain MRI, and longer survival, without increased toxicity.

Dr. Schiff raised some cautionary notes. Approximately 20% of randomized patients were not included in the outcomes analysis, and postrandomization exclusions can favor the investigational therapy. Additionally, the improvement in overall survival with TBRT is quite surprising, both because the great majority of patients with brain metastases succumb to systemic disease rather than CNS progression, and previous studies comparing resection cavity SRT to either observation or whole brain radiotherapy showed no overall survival benefit with SRT.

Further analyses of the data are ongoing to explore additional outcomes, including the impact of treatment on mental processes such as cognition, memory, and attention, as well as the influence of concomitant oncologic therapies on treatment efficacy. In addition, a separate trial (BRIDGES; ClinicalTrials.gov Identifier: NCT07195591) is evaluating TBRT in patients with glioblastoma.

DISCLOSURE: The study was funded by GT Medical Technologies. Dr. Weinberg has received honoraria from GT Medical Technologies; has served as a consultant or advisor for OsteoMed; has received research funding from GT Medical Technologies and InSightec; and has received reimbursement for travel expenses from GT Medical Technologies. Dr. Schiff has served as a consultant or advisor for Curis and Exelixis; and has received royalties for submissions to UpToDate.

REFERENCE

1. Weinberg JS, Lin HY, McAleer MF, et al: 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. 2026 ASCO Annual Meeting. Abstract LBA2000. Presented May 30, 2026.

Expert Point of View

During a press briefing ahead of the 2026 ASCO Annual Meeting, David Schiff, MD, Harrison Distinguished Teaching Professor of Neurology, Neurological Surgery, and Medicine at the University of Virginia School of Medicine, Charlottesville, and an ASCO Expert in central nervous system (CNS) tumors, commented on the phase III Radiation One and Done Study (ROADS), noting the biologic plausibility that tile-based radiation therapy implanted at the time of brain metastasis resection may improve local control, particularly for larger metastases in which standard radiosurgery may be less effective.¹

David Schiff, MD

“These results certainly are very applicable to patients undergoing brain metastasis resection, and they may represent a new standard of care,” Dr. Schiff remarked, while emphasizing “there are nonetheless a few caveats.”

One concern was that a “pretty high figure” of randomized patients (~ 11%) was excluded from the modified intention-to-treat analysis. “… Postrandomization exclusions can favor the investigational therapy,” he remarked.

“Related to that, of the 101 patients randomly assigned to radiosurgery, as opposed to tiles [n = 103 treated], 18 of them never received cavity radiosurgery,” he added. “So, it’s not completely a fair comparison, although you could argue that that makes it more of a real-world comparison….”

Dr. Schiff also noted that it was not reported whether the Independent Review Committee evaluating surgical bed recurrence was blinded to treatment assignment, although he presumed that it was.

The final issue raised was overall survival. Overall survival in most patients with brain metastases is driven by systemic disease rather than CNS-related death, with fewer than 25% dying from intracranial disease, leading him to describe the overall survival benefit as “surprising.” He added that prior studies—including one from The University of Texas MD Anderson Cancer Center comparing postoperative stereotactic radiosurgery vs observation², which established this approach as the standard of care—showed no overall survival benefit, concluding that “these overall survival results aren’t easily explained.”

Despite these caveats, Dr. Schiff offered praise, stating, “These are exciting results, and I congratulate Dr. Weinberg and the rest of the study team on completing a randomized surgical neuro-oncology trial—those can really be a challenge to complete accrual.”

DISCLOSURE: Dr. Schiff has served as a consultant or advisor for Curis and Exelixis; and has received royalties for submissions to UpToDate.

REFERENCES

1. Weinberg JS, Lin HY, McAleer MF, et al: 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. 2026 ASCO Annual Meeting. Abstract LBA2000. Presented May 30, 2026.

2. Mahajan A, Ahmed S, McAleer MF, et al: Post-operative stereotactic radiosurgery versus observation for completely resected brain metastases: A single-centre, randomised, controlled, phase 3 trial. Lancet Oncol 18:1040-1048, 2017.