Objectives: Comparing to standard radiotherapy, SBRT allows to deliver high doses to the spinal tumor volume by its ability to respect the spinal cord dose tolerance. The comparison between different stereotactic fractionation schedules, leading to several BED, has not yet been evaluated. The aim of the study is to evaluate the impact of BED on local control in case of SBRT for spinal metastases.

Methods: From November 2010 to June 2016, 152 patients were treated on 183 metastases in a single institution with a robotic stereotactic device. We retrospectively reviewed patient, tumor and treatment characteristics as well as local control and complications. Dose constraints from Timmerman et al. (2008) were respected. The total prescribed dose was converted into BED with an alpha/beta ratio of 10 (BED10). The Kaplan Meier method was applied to calculate the local control rate. Impact of all characteristics on local control was assessed using univariate and multivariate Cox proportional-hazards models.

Results: Median follow up was 10.8 months (range: 0.3-53.9 months). Median age was 65 (range: 7-90). Main primary tumor sites were lung (31%), kidney (25%), breast (15%) and prostate (7%). 90% of fractionation schedules were 3 fractions (8 to 10 Gy per fraction) or 5 fractions (6 to 8 Gy per fraction). Median BED10 was 59 Gy (range: 29-100 Gy). The 1-year local control rate was 94% (95%CI: 89-98%). Among clinical and dosimetric parameters, the BED10 was found to be the only predictor of local control in multivariate analysis: RR 0.90 (95%CI: 0.81-0.99, p=0.047). Patients with a BED10 > 50 Gy had a significantly better 1-year control rate than those with a BED10 < 50 Gy: 96% (95%CI: 92-100) versus 79% (95%CI: 60-98), respectively, p = 0.034 (log-rank test). Pain control or improvement was achieved in 91% and pain flare in 9%. 7% of patients developed a vertebral compression fracture and there was no radiation-induced myelitis.

Conclusions: Upfront spinal SBRT is an efficient therapeutic option for patients with spinal metastases. Local control and pain relief rates were high while side effects remained low and comparable to the literature. BED is an important independent prognostic factor of local control. A BED > 50 Gy seems to lead to excellent local control rates.