Long-term outcomes of high-dose single-fraction radiosurgery for chordomas of the spine and sacrum

OBJECTIVE The current treatment of chordomas is associated with significant morbidity, high rates of local recurrence, and the potential for metastases. Stereotactic radiosurgery (SRS) as a primary treatment could reduce the need for en bloc resection to achieve wide or marginal margins. Spinal SRS outcomes support the exploration of SRS's role in the durable control of these conventionally radioresistant tumors. The goal of the study was to evaluate outcomes of patients with primary chordomas treated with spinal SRS alone or in combination with surgery. METHODS Clinical records were reviewed for outcomes of patients with primary chordomas of the mobile spine and sacrum who underwent single-fraction SRS between 2006 and 2017. Radiographic local recurrence-free survival (LRFS), overall survival (OS), symptom response, and toxicity were assessed in relation to the extent of surgery. RESULTS In total, 35 patients with de novo chordomas of the mobile spine (n = 17) and sacrum (n = 18) received SRS and had a median post-SRS follow-up duration of 38.8 months (range 2.0-122.9 months). The median planning target volume dose was a 24-Gy single fraction (range 18-24 Gy). Overall, 12 patients (34%) underwent definitive SRS and 23 patients (66%) underwent surgery and either neoadjuvant or postoperative adjuvant SRS. Definitive SRS was selectively used to treat both sacral (n = 7) and mobile spine (n = 5) chordomas. Surgical strategies for the mobile spine were either intralesional, gross-total resection (n = 5) or separation surgery (n = 7) and for the sacrum en bloc sacrectomy (n = 11). The 3- and 5-year LRFS rates were 86.2% and 80.5%, respectively. Among 32 patients (91%) receiving 24-Gy radiation doses, the 3- and 5-year LRFS rates were 96.3% and 89.9%, respectively. The 3- and 5-year OS rates were 90.0% and 84.3%, respectively. The symptom response rate to treatment was 88% for pain and radiculopathy. The extent or type of surgery was not associated with LRFS, OS, or symptom response rates (p > 0.05), but en bloc resection was associated with higher surgical toxicity, as measured using the Common Terminology Criteria for Adverse Events (version 5.0) classification tool, than epidural decompression and curettage/intralesional resection (p = 0.03). The long-term rate of toxicity >= grade 2 was 31%, including 20% grade 3 tissue necrosis, recurrent laryngeal nerve palsy, myelopathy, fracture, and secondary malignancy. CONCLUSIONS High-dose spinal SRS offers the chance for durable radiological control and effective symptom relief with acceptable toxicity in patients with primary chordomas as either a definitive or adjuvant therapy.