Spine Stereotactic Radiosurgery for Primary and Metastatic Osteosarcoma

Abstract

Osteosarcoma is difficult to control due to its high propensity for metastasis and resistance to local and systemic therapies. High doses of radiation therapy (RT) may confer local control (LC) in some settings but for lesions involving the vertebral bodies, proximity to the spinal cord may limit the ability to deliver an adequate dose. In this analysis, we investigate the role of spine stereotactic radiosurgery (SSRS) to overcome this barrier and enable efficacious treatment of primary or metastatic osteosarcoma of the spine. We retrospectively reviewed all patients treated with SSRS for osteosarcoma of the vertebrae between 2006 and 2022 at a single large tertiary cancer center. We utilized the Kaplan-Meier method to estimate overall survival (OS) and LC. We identified 18 patients treated with SSRS for 25 lesions of spinal osteosarcoma. Median follow-up was 17.2 months. Two patients and three separate lesions were treated with SSRS for primary osteosarcoma of the vertebrae. The remaining 16 patients and 22 lesions received SSRS to the spine for metastatic disease. Lesions were treated to a dose of 24Gy in one fraction (n = 20) 27Gy in 3 fractions (n = 4) or 50Gy in 5 fractions (n = 1). Treatment sites included the cervical spine alone (n = 4), thoracic spine alone (n = 12), lumbar spine alone (n = 4), sacrum alone (n = 3), or both the thoracic and lumbar spine (n = 2). At latest follow up, local failure was observed in 9/25 (36%) treated lesions and median LC was 22.5 months (95% CI 6-43 months). Per-lesion LC at 1 year was 64% (95% CI 35-83%). Per-patient median OS was 14 months (95% CI 7-68 months) and OS estimates at 1 and 2 years were 60% (95% CI 32-80%) and 35% (11-60%), respectively. Among 15 patients who received 24 Gy in one fraction, at 1 year per-lesion LC was 72% (95% CI 41-88%) and per-patient OS was 60% (95% CI 28-81%). The most common acute treatment related toxicity was pain flare (12%). Four patients (16%) developed compression fractures in the treated vertebrae after radiation, with incidence between 57 and 578 days after radiation. Two of these fractures required intervention and two were incidental findings on imaging. No patients developed CTCAE Grade 3 or higher adverse events including neurological toxicities. SSRS appears to be safe and effective in the treatment of metastatic or primary osteosarcoma involving spinal bone. Future work should include further investigation of this technique with pooled multi-institutional studies and randomized trials.