Abstract
INTRODUCTION
The effect of stereotactic radiosurgery (SRS) dose-rate on tumor control for acoustic neuroma (AN, or vestibular schwannoma) is unclear.
METHODS
This was a retrospective study of all patients treated for AN with frame-based cobalt-60 SRS at the Toronto Western Hospital between 2005-2019. Dose rates on the day of SRS were calculated from the calibration dose-rate while accounting for the cobalt-60 half-life of 5.2713 years. Local failure was defined as continued tumor growth >36 months post-SRS, tumor resection for LF, or use of any repeat SRS for LF. Cumulative incidence of LF was reported after accounting for competing risks of death, on a per-lesion basis. Comparisons of actuarial LF were made using Gray’s test. Multivariable analysis of LF was performed using a proportional hazards model.
RESULTS
A total of 607 patients were treated for 617 acoustic neuromas. Median follow-up was 5.0 years. 158 tumors (26%) were cystic. 71 tumors (12%) had previous resection. Nine patients received 10-11 Gy due to large tumor size; all remaining patients received 12 Gy to approximately the 50% isodose line. Median dose rate was 2.4 Gy/min (range, 1.3-3.7). There was no association between dose rate and LF (≥ 2.4 Gy/min vs. < 2.4 Gy/min, 6.07% vs. 6.12% at 5-year follow-up, p = 0.75). The adjusted local failure-specific hazard ratio (HR) for dose rate (per Gy/min) was 1.2 (95% CI 0.69-2.1, p = 0.52). Patients with previous surgery had higher LF, with a HR of 3.6 (95% CI 1.7-7.8, p = 0.0012), after adjusting for presence of cysts (HR 0.27, p = 0.034) and maximum tumor diameter (HR 1.055 per cm, p = 0.071).
CONCLUSIONS
In a large cohort of patients with acoustic neuromas, radiosurgery dose-rate was not associated with tumor control. Previous resection was a strong risk factor for local failure after SRS.