358 - Is Reirradiation Spine SBRT with Spinal Cord Doses Exceeding HyTEC Constraints Safe and Effective?
Presenter(s)
T. Tang1, C. Udovicich2, K. L. Zeng1, E. Atenafu3, J. Mullins1, C. L. Tseng1, H. Chen1, D. Dinakaran1, H. Soliman4, A. Sahgal4, and J. Detsky4; 1Department of Radiation Oncology, Odette Cancer Centre, Sunnybrook Health Sciences Centre, Toronto, ON, Canada, 2GenesisCare Melbourne, Malvern, Australia, 3Department of Biostatistics, University Health Network, University of Toronto, Toronto, ON, Canada, 4Department of Radiation Oncology, Odette Cancer Centre, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, ON, Canada
Purpose/Objective(s): Data on spinal cord tolerance in the context of reirradiation stereotactic body radiotherapy (SBRT) for spinal metastases remains limited. Hy Dose per Fraction, Hypofractionated Treatment Effects in the Clinic (HyTEC) recommends a reirradiation SBRT thecal sac point maximum dose (Dmax) in equivalent dose in 2-Gy fractions with a/ß = 2 (EQD22) =25 Gy following prior conventional radiation, along with other caveats. We report outcomes after reirradiation spine SBRT stratified by adherence to vs. exceeding HyTEC constraints.
Materials/Methods: A prospectively maintained institutional database of patients treated with spine SBRT between January 2008 and December 2022 was retrospectively reviewed. Patients with at least one vertebral segment reirradiated with spine SBRT were included. Those with only adjacent segments reirradiated were excluded. Segments were classified based on reirradiation dose to the central neural structure (CNS), defined as the spinal cord planning organ at risk volume or thecal sac. Those with reirradiation CNS EQD22 Dmax =25 Gy were considered adherent, while those >25 Gy were considered CNS-escalated. Rates of local failure (LF), overall survival (OS), vertebral compression fracture (VCF), and neurologic adverse events (AE) were compared between the two cohorts.
Results: A total of 211 patients/487 vertebral segments were analyzed. Of these, 80 patients/195 segments were adherent and 131 patients/292 segments were CNS-escalated. Most patients had a radiosensitive primary tumor (n=150, 71%) and no baseline neurologic deficits (n=178, 84%). Most segments had intermediate spinal stability (n=267, 55%) and no baseline VCF (n=315, 65%). Prior treatment was mainly conventional radiation alone (n=342, 70%), followed by SBRT alone (n=116, 24%) and a combination of the two (n=29, 6%). Median follow-up was 15.8 months vs. 14.2 months in the adherent vs. CNS-escalated cohorts, respectively. Median reirradiation CNS EQD22 Dmax was significantly higher in the CNS-escalated vs. adherent cohort (33.2 Gy [interquartile range {IQR} 32.0–33.9] vs. 22.1 Gy [IQR 18.9–24.5]; p<0.0001), as was median cumulative CNS EQD22 Dmax (68.7 Gy [IQR 63.6–77.6] vs. 60.6 Gy [IQR 50.8–71.0]; p<0.0001). The 1-year LF, OS, and VCF rates were 18%, 69%, and 8% in the adherent cohort vs. 19%, 58%, and 8% in the CNS-escalated cohort, with no significant differences. Nerve root AE occurred in 5 adherent patients (6%) and 5 CNS-escalated patients (4%), with one case of radiation myelopathy (RM) in the latter (0.7%). This patient had a reirradiation CNS EQD22 Dmax of 40.2 Gy and a cumulative CNS EQD22 Dmax of 89.1 Gy.
Conclusion: To our knowledge, this represents the largest cohort of reirradiation spine SBRT patients. Escalating CNS doses beyond HyTEC guidelines appeared safe with a low risk of RM, even after previous SBRT; however, this did not translate into improved spinal oncologic outcomes.