ePoster

#45484: Non-myelitis neurologic toxicity of stereotactic ablative radiotherapy for spine

Treatment Type: SBRT
Location of Treatment: Spine
Authors:
Soon Woo Hong (1), Jin Ho Kim (2)
Institution & Country:
1. , Seoul National University Bundang Hospital , Seoul, Republic of Korea
2. , , ,

Purpose: Radiation myelitis (RM) is the most dreaded complication of stereotactic ablative radiotherapy (SABR) for the spine. We investigated radiation-induced radiculo-plexopathy (RIRP) in patients who underwent SABR for spinal metastases.

Materials and Methods: Patients who received SABR for metastatic spine tumors arising in C3 to T1 and T12 to S2 vertebrae from 2013 to 2023 were analyzed. Medical records and images were retrospectively reviewed. RIRP was defined as a motor power decrease by MRC grade 2 or greater in the myotome of the spine treated by SABR that was not caused by local progression or RM. In addition to the spinal cord and cauda equina, neural foramina that is either included within or adjacent to target volumes were designated as organs at risk. While 407 neural foramina were contoured, only those with a maximum EQD2 (a/b ration=2) greater than 45 Gy were included for dosimetric analysis. The Cox proportional hazard model was used to identify risk factors for RIRP.

Results: Of 132 patients with spinal metastases, 168 C1-T1 and T12 to S3 spine segments were treated with SABR. The median follow-up was 9.5 months. Twenty-four local recurrences occurred (14.3%, 24/168). No cases of radiation myelitis were observed, and one case of radiation-induced cauda equina syndrome was identified. Overall, 20 SABR cases in 12 patients were identified to have decreased motor power by 2+ grades. Of these, 7 spine segments in 6 patients showed bilateral limb weakness. Detailed analysis revealed that none of these motor weaknesses were caused by local progression or RM. The crude incidence of RIRP was 11.9% (20/168) in 12 patients. The incidence of bilateral RIRP was 4.2% (7/168). The planning target volume and the maximum dose to neural foramina were significantly associated with the occurrence of RIRP (p<0.001).

Conclusions: In contrast to RM or radiation-induced cauda equina syndrome, RIRP occurs more frequently after spinal SABR. Neural foramina may serve as provisional surrogate organs at risk for RIRP. Further studies are necessary to characterize RIRP and relevant dosimetric constraints.