Clinical Outcomes of Stereotactic Body Radiation Therapy for Bone and Non-Spine Bone Metastases: A Large Single Institution Experience.

PURPOSE/OBJECTIVE(S) Improved oncologic therapies have expanded the role of stereotactic body radiation therapy (SBRT) for bone metastases; however, the impact of SBRT on local control (LC), progression-free survival (PFS), overall survival (OS) and long-term risks remains poorly understood. We characterized predictive factors associated with clinical outcomes through a large single institution experience. MATERIALS/METHODS We retrospectively reviewed all patients with bone metastases treated with SBRT at our institution and characterized patient, tumor, and treatment factors and subsequent outcomes. Common treatment indications included synchronous (S) or metachronous (M) oligometastatic cancer, re-irradiation, or radioresistant histology. Common fractionations included 30-35Gy/5F, 9-10Gy/3F, 16-20Gy/1F. Outcomes were assessed on a per patient and per lesion basis using Cox regression to evaluate predictive factors for LC, PFS, and OS. Toxicities were graded by CTCAEv5. RESULTS From 2009 to 2020, 434 patients (median age 64, range 19-93, 65% male, 88% Caucasian, ECOG PS 0-1 88%) were treated with SBRT primarily for oligometastatic cancer (82% M, 5% S; 96% with 1-3 lesions) of various origin (39% prostate, 21% breast/lung). 651 lesions (66% spine, 31% non-spine bone, 3% spine/paraspinal) were treated. 96% of lesions had ≥95% PTV covered by 90% of prescription dose (V90) (median 100, range 42-100). Median PTV volume was 41.3cc, and 87% of lesions were < 150cc. 46% lesions were prescribed a biologically equivalent dose (BED) ≥50 (α/β = 10). LC was significantly worse if PTV V90 < 95% (HR 2.06, CI 1.18-3.62, P < 0.05) and improved if prescribed BED≥50 (HR 0.59, CI 0.42-0.83, P < 0.01) on multivariate analysis (MVA). Larger lesions ≥150cc predicted worse LC on univariate analysis (UVA) only (HR 1.80, CI 1.22-2.64, P < 0.005). Improved PFS and OS were associated with better performance status (P < 0.0001) and prostate histology (P < 0.01) on MVA. Metachronous presentation with met-free interval ≥1 year had improved PFS and OS on UVA only (P < 0.001). Common acute toxicities were pain flare (9% vs 12%) and grade 1-2 fatigue (14% vs 19%), among oligometastatic vs re-irradiation lesions, respectively. Late toxicities included treatment-related fracture (1.2%) for oligometastatic lesions and radiation-related myelopathy (2.3%) and nerve pain (1.2%) for re-irradiation/other lesions. CONCLUSION SBRT may be an effective means of improving LC among patients with oligometastatic bone disease if adequate target coverage and a BED≥50 can be achieved, independent of fraction number, and with minimal risk of acute or late toxicity. Prostate histology and performance status were patient factors predictive of PFS and OS. Our experience highlights important factors that can help guide appropriate patient selection and use of SBRT in this unique population.