Session Item

Stereotactic Body Radiotherapy (SBRT) is increasingly used in uncomplicated spine metastases to palliate symptoms and prolong disease control. However, criteria for patient selection are not available. The aim of this study is to identify determinants of local failure and progression-free interval in patients treated with SBRT to spinal metastases.

Data from consecutive patients treated with Cyberknife-based spine SBRT between January 2019 and March 2020 were retrospectively collected. Dose was expressed as Biological Effective Dose for α/β=10 (BED₁₀). Kaplan-Meyer method was used to calculate Local Control (LC) and Disease Progression-free Survival (DPFS) from date of SBRT to event. Univariate (UVA) and Multivariate analysis (MVA) were performed using log-rank and Cox model, respectively. 

Sixty-two patients accounting for 70 spinal metastases were included. Median age was 66 (range 32-87) years. Disease was metastatic at diagnosis in 21 patients (34%) : an active primary tumor was present in 17 patients (27%). Among treated sites, most represented primary malignancies were prostate (n=28, 40%) and breast (n=21, 30%). SBRT was delivered to cervical, thoracic, lumbar and sacral vertebrae in respectively 15(21%), 21(30%), 29(41%), and 5(8%) cases. Dose regimens consisted of 25-30 Gy in 5 fractions and 21-30 Gy in 3 fractions in respectively 61 (87%) and 9 (13%) cases, resulting in a median BED of 43.2 (range 37.5-60) Gy₁₀. Concurrent chemotherapy (including cytotoxic or targeted agents) was administered in 43% of cases (n=30). After a median follow up of 10 months (range 1-24 months), 9 local relapses and 40 distant progressions were observed. One year LC was 87% (Fig.1A): non-prostate primary tumor ( p=0.003, Fig.1B) and concurrent chemotherapy (p=0.006, Fig.1C) were associated to poorer LC at UVA, and an independent correlation was confirmed at MVA (respectively p=0.017 and p=0.024). One-year DPFS was 43% (Fig.1D). UVA showed a correlation between impaired DPFS and active primary tumor (p=0.003), metastatic dissemination at diagnosis (p=0.02) and non-prostate primary tumor (p=0.009), although only an active primary tumor site was independently associated to DPFS at MVA (p=0.007, Fig.1E). Acute toxicity consisted of G2 pain flare and G2 nausea in respectively 5 (7%) and 4 (6%) cases: no clinical variable was significantly correlated with increased acute toxicity. No late toxicity, in particular vertebral fracture, was reported.

Spine SBRT results in high LC rates and durable progression-free survival with modest incidence of mild toxicity. A better knowledge of determinants of radioresistance (both intrinsically inherent to tumor biology  or acquired following repeated administration of systemic therapy) is desirable due to potential correlation with impaired LC. A controlled primary site is associated with longer time to further distant progression following SBRT, supporting local treatment of primary tumor in selected metastatic patients.