Single-fraction versus two-fraction stereotactic radiosurgery for brain oligometastases: a multicenter propensity score-matched analysis with exploratory assessment of biologically effective dose and primary tumor control status

Abstract

To compare survival and intracranial outcomes between single- and two-fraction SRS for brain oligometastases, and to explore whether primary tumor control status and biologically effective dose (BED) are associated with outcomes. In this multicenter, retrospective, propensity score-matched (PSM) study, 280 patients treated with single-fraction (16 Gy) or two-fraction (9–14 Gy/fraction) SRS were evaluated using 1:1 nearest-neighbor matching. The primary endpoint was overall survival (OS). Secondary endpoints included local control (LC), distant brain failure, and symptomatic radionecrosis (SRN). Primary tumor control status was predefined using standardized criteria and included as a matching covariate to ensure balanced comparison of this key effect modifier. Pre-specified exploratory subgroup analyses assessed effect modification by primary tumor control status and tumor border clarity. BED was calculated using the linear-quadratic model (α/β = 10 Gy). In the matched cohort ( n = 280; 140 per arm), overall median OS was 10.0 months (95% CI 9.4–10.9), the 12-month LC rate was 87.5% (95% CI 83.3%–90.8%), and the SRN rate was 14.7% (39/266 evaluable patients). OS did not differ significantly between single-fraction and two-fraction SRS (10.2 vs. 9.6 months; HR = 1.13, 95% CI 0.88–1.45; P = 0.33). LC was comparable (88.6% vs. 86.4%; P = 0.49); SRN was numerically lower with two-fraction SRS (11.4% vs. 17.9%; P = 0.13). Pre-specified exploratory subgroup analyses suggested shorter OS with two-fraction versus single-fraction SRS in patients with uncontrolled primary tumors (HR = 1.83, 95% CI 1.26–2.66; interaction FDR-adjusted P = 0.002) and unclear tumor borders (HR = 2.39, 95% CI 1.10–5.16; subgroup P = 0.023; interaction FDR-adjusted P = 0.041). Stratification by BED revealed that survival disparities were predominantly seen in the low-dose two-fraction schedule (9 Gy×2, BED10=32.4 Gy10), whereas high-dose two-fraction SRS (10–14 Gy×2, BED10≥40 Gy10) provided survival outcomes comparable to single-fraction SRS (Bonferroni-adjusted P = 1.000). Overall outcomes were equivalent between the two SRS regimens. Exploratory analyses raise the hypothesis that primary tumor control status and lower BED may modify survival in patients with uncontrolled systemic disease. These hypothesis-generating observations warrant prospective validation with subgroup-specific design before informing clinical practice.