(P 308) OPTIMAL BIOLOGIC EFFECTIVE DOSE OF STEREOTACTIC BODY RADIOTHERAPY FOR METASTATIC AND RECURRENT SARCOMA

Objective: Patients with metastatic sarcoma can be cured if aggressive local tools are used in addition to systemic therapy. Stereotactic body radiation therapy (SBRT) affords the potential to overcome the relative radioresistance of sarcomas through delivery of a focused higher biological effective radiation dose (BED) as an alternative to invasive surgery or conventional radiotherapy. Our current SBRT regimen was based on a small institutional series and dose not vary by sarcoma histology. We sought to understand the local control of sarcoma SBRT as it relates to histology and to identify an optimal SBRT regimen.

Methods: From our IRB-approved single-institution registry, all patients treated with SBRT for metastatic sarcoma between 2014 and 2020 were identified. A receiver operating characteristic (ROC) curve was generated to determine optimal BED. Local control was compared by SBRT dose using the BED cut point and evaluated by histology. The Kaplan-Meier method was used to compare local control rates first by radiosensitivity and then by dose treatment groups. A stratified Kaplan-Meier analysis then analyzed the interaction of radiosensitivity and dose on local control. Log-rank statistics were used to compare the groups.

Results: Forty-two patients with a total of 138 lesions met inclusion criteria. Median SBRT prescription was 116.70 Gy BED (range 66.70 - 419.30). Desmoplastic small round cell tumor, Ewing sarcoma, rhabdomyosarcoma, and small round blue cell sarcomas were classified as radiosensitive (n= 63), and all other histologies were classified as radioresistant (n=75). Local control for all lesions was 66.7% (95% CI, 56.6-78.5) at 1 year and 50.2% (95% CI, 38.2-66.1) at 2 years. Stratifying by histology, 1- and 2-year local control rates were 65.3% and 55.0%, respectively, for radiosensitive, and 68.6% and 44.5%, respectively, for radioresistant histologies (p=0.49 ). The ROC cut point for BED was 95 Gy. Local control rates at 1- and 2-years were 75% and 61.6%, respectively, for lesions receiving >95 Gy BED, and 46.2% and 0%, respectively, for lesions receiving < 95 Gy BED (p=0.01). On subgroup analysis, local control by BED >95 Gy was significant for radiosensitive histologies (p=0.013), and trended toward significance for radioresistant histologies (p=0.25).

A separate analysis was performed specifically for osteosarcoma patients (23) and lesions (66). The 6- and 12-month local control rates were 89% (95% CI 0.8-0.99) and 78% (CI 66%-92%) respectively. A dose cutoff for improved local control was found at a BED of 120 Gy. Local control at one year was 89% (CI 77%-100%) for lesions that received ≥120 Gy BED versus 61% (CI 40%-91%) for those that received < 120 Gy BED (p=0.034).

No acute grade ≥3 toxicity was observed for any of the patients.

Conclusion: Our large sarcoma SBRT series demonstrates that local control for recurrent and metastatic disease is possible with minimal toxicity. Local control was significantly improved with a biologically effective dose of at least 95 Gy for radiosensitive histologies and 120 Gy for osteosarcoma. These SBRT regimens should be further investigated on prospective trials