Three-fraction preoperative stereotactic radiotherapy for resectable brain metastases: the phase 2 STEP trial

Postoperative stereotactic radiotherapy (SRT) is the standard adjuvant treatment after resection of brain metastases. It nevertheless requires delineation of a dynamic postsurgical cavity, delays local treatment by 3 to 6 weeks (and thus resumption of systemic therapy) and carries substantial risks of radiation necrosis and leptomeningeal disease. Preoperative SRT targets an intact lesion and may address some of these limitations.

In this open-label, single-arm, multicenter phase 2 trial, eligible adults had a Karnofsky performance status ≥70 and up to four brain metastases, including one resectable lesion ≤5 cm, from breast, non-small cell lung, kidney, or digestive cancer or melanoma. Patients received preoperative SRT (three fractions of 7.7 Gy prescribed to the encompassing 70% isodose line, equivalent to 11Gy per fraction at the isocenter) followed by surgical resection within 3 days of the last fraction. The primary outcome was 6-month local control on centrally reviewed MRI. Secondary outcomes included index-lesion radiation necrosis, leptomeningeal disease, distant intracranial control, overall survival, toxicity, cognition, and quality of life.

Of 73 patients assessed, 67 were eligible and 63 received both SRT and surgery. Median age was 62 years. At a median follow-up of 12.7 months, the 6-month local control rate was 100% (95% CI, 95%-100%), meeting the primary endpoint; the 12-month rate was 96.4% (95% CI, 87%-99%). No leptomeningeal disease occurred. The 1-year cumulative incidence of index-lesion radiation necrosis was 7% (95% CI, 2%-16%), meeting the key secondary endpoint. Distant intracranial failure occurred in 22 patients (35%); 12-month overall survival was 81.2% (95% CI, 70%-89%). Postoperative complications occurred in 4 patients (6.3%); cognition and quality of life remained stable over the 12 months of protocol follow-up.

Three-fraction preoperative SRT followed by resection was associated with high local control, low radiation necrosis, and no leptomeningeal disease, supporting evaluation in ongoing randomized trials.