Vertebral compression fractures after stereotactic body radiation therapy: a large, multi-institutional,...

OBJECTIVE The purpose of this study was to identify factors contributing to an increased risk for vertebral compression fracture (VCF) following stereotactic body radiation therapy (SBRT) for spinal tumors. METHODS A total of 594 tumors were treated with spinal SBRT as primary treatment or re-irradiation at 8 different institutions as part of a multi-institutional research consortium. Patients underwent LINAC-based, image-guided SBRT to a median dose of 20 Gy (range 8-40 Gy) in a median of 1 fraction (range 1-5 fractions). Median patient age was 62 years. Seventy-one percent of tumors were osteolytic, and a preexisting vertebral compression fracture (VCF) was present in 24% of cases. Toxicity was assessed following treatment. Univariate and multivariate analyses were performed using a logistic regression method to determine parameters predictive for post-SBRT VCF. RESULTS At a median follow-up of 10.1 months (range 0.03-57 months), 80% of patients had local tumor control. At the time of last imaging follow-up, at a median of 8.8 months after SBRT, 3% had a new VCF, and 2.7% had a progressive VCF. For development of any (new or progressive) VCF following SBRT, the following factors were predictive for VCF on univariate analysis: short interval from primary diagnosis to SBRT (less than 36.8 days), solitary metastasis, no additional bone metastases, no prior chemotherapy, preexisting VCF, no MRI used for target delineation, tumor volume of 37.3 cm(3) or larger, equivalent 2-Gy-dose (EQD2) tumor of 41.8 Gy or more, and EQD2 spinal cord Dmax of 46.1 Gy or more. Preexisting VCF, solitary metastasis, and prescription dose of 38.4 Gy or more were predictive on multivariate analysis. The following factors were predictive of a new VCF on univariate analysis: solitary metastasis, no additional bone metastases, and no MRI used for target delineation. Presence of a solitary metastasis and lack of MRI for target delineation remained significant on multivariate analysis. CONCLUSIONS A VCF following SBRT is more likely to occur following treatment for a solitary spinal metastasis, reflecting a more aggressive treatment approach in patients with adequately controlled systemic disease. Higher prescription dose and a preexisting VCF also put patients at increased risk for post-SBRT VCF. In these patients, pre-SBRT cement augmentation could be considered to decrease the risk of subsequent VCF.