2624 - Clinical Outcomes of Stereotactic Body Radiotherapy for Spinal Metastases with Paraspinal Extension

N. K. Vuppala¹, M. C. LeCompte², A. Bydon³, K. Kebaish⁴, N. Theodore³, B. Wu², M. Khan⁵, L. R. Kleinberg², S. H. Lee⁶, D. Lubelski³, and K. J. Redmond⁶; ¹Alabama College of Osteopathic Medicine, Dothan, AL, ²Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University School of Medicine, Baltimore, MD, ³Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, MD, ⁴Department of Orthopedic Surgery, Johns Hopkins University School of Medicine, Baltimore, MD, ⁵Department of Radiology, Thomas Jefferson University, Philadelphia, PA, ⁶Johns Hopkins University School of Medicine, Baltimore, MD

Purpose/Objective(s): Stereotactic body radiotherapy (SBRT) for spine metastases has been increasingly utilized, given its excellent local control and pain management. However, emerging data suggests that tumors with paraspinal extension may have poorer local control. The purpose of the current study is to evaluate local control after SBRT for spine metastases with paraspinal extension. Materials/

Methods: Patients with spinal metastases with paraspinal extension were treated with SBRT at a single institution from 2009-2021. Pain response was assessed at 3-month intervals post-SBRT, with complete response defined as a pain level of 0 and partial response as a decrease of pain by =2 or =25% in oral morphine equivalents. Local failure (LF) was defined per Spine response assessment in Neuro-Oncology criteria. Variables examined included paraspinal mass measurements, extent of involvement, and dosimetric/volumetric characteristics. Fractionation schemes were converted to 3-fraction equivalent doses (3fxED) using the linear quadratic model with an alpha/beta of 3. Univariate Cox proportional hazard models for LF were constructed using the Fine and Gray competing risk method.
Results: 76 paraspinal masses from 69 patients underwent SBRT, with a median follow-up of 11.8 months. Common primary tumor types were non-small cell lung cancer (17%), sarcoma (16%), and gastrointestinal (16%). Cases spanned 1 (21%), 2 (34%), 3 (33%), or =4 (12%) vertebral segments. 67% occurred in the thoracic region, 76% of which involved adjacent ribs. 21% of masses abutted gastrointestinal organs/kidney, and 18% had muscle invasion. The median craniocaudal extent was 3.6 cm (1.0 – 8.8 cm), and the median planning target volume was 133 cc (22 – 753 cc). Cases were treated in 2 (12%), 3 (45%), and 5 (43%) fractions to median prescriptions doses of 2400, 2700, and 3000 cGy, respectively. At 3 months post-SBRT, complete and partial pain responses were 39% and 15%, respectively. The 12-month incidence of LF was 30%, with a 39% overall incidence. 43% of LF had an epidural component, and 40% had a paraspinal component. 27% of LF were infield, 33% were marginal, and 40% were infield and marginal. On univariate analysis, clinical and dosimetric data were not significantly associated with LF.
Conclusion: High rates of LF were observed for spinal metastases with paraspinal extension treated with SBRT. While a significant proportion of progression occurred in the epidural space as anticipated, the rate of paraspinal progression, including marginal failures, was notably high. Further investigation is necessary to understand whether these patterns reflect aggressive biology or limited barriers to microscopic spread. Our findings emphasize the need to optimize treatment for patients with paraspinal extension while considering challenges in dose escalation due to large volumes and proximity to critical structures. Potential strategies include larger expansions in paraspinal involvement, spacers, and a combination of SBRT and ablation.