Cyberknife Radiosurgery for Acoustic Neuromas: Tumor Control and Clinical Outcomes

Background CyberKnife radiosurgery (CKRS) is increasingly utilized as a definitive or adjuvant treatment for acoustic neuromas. Fractionated CKRS may reduce the risk of long-term radiation toxicity to nearby critical structures compared with single fraction radiosurgery. However, tumor control rates and clinical outcomes following CKRS for acoustic neuromas have not been thoroughly investigated.

Methods An IRB-approved database of all patients treated with CKRS for acoustic neuromas from 2004 to 2011 at our institution was queried. Clinical and radiographic data were collected via retrospective chart and imaging review. Treatment failure was defined as the need for additional surgical intervention. Kaplan–Meier analysis was performed to analyze treatment failure rates.

Results Of 119 patients with evaluable clinical and radiographic data, the median age was 55 (range: 18–83) years, and 67 patients (56%) were female. The median follow-up period was 49 (range: 6–133) months. Twenty-five patients (21%) had a history of prior surgery, one patient (1%) had a history of prior radiosurgery, and four patients (3%) had a history of neurofibromatosis II. The median tumor volume was 1.6 (range: 0.02–17) cm³. Ninety-seven tumors (81%) were KOOS Grades I to III, and 22 (19%) were Grade IV. The median prescription dose was 18 (range: 13–25) Gy delivered in a median of three fractions (range: 1–5). The median maximum cochlear dose was 18 (range: 7–27) Gy. Of 59 patients with pre-radiosurgery serviceable hearing (AAO HNS class A or B), 35 (59%) maintained serviceable hearing at last audio follow-up (median: 21 months). Two of 111 patients (2%) with facial nerve function House Brackmann (HB) ≤ 3 progressed to HB > 3 following radiosurgery, while 3 of 8 patients (37.5%) with facial nerve function HB > 3 improved to HB ≤ 3 following radiosurgery. New facial pain was reported by five patients following radiosurgery (5/106; 5%). KOOS Grade IV was predictive of radiographic tumor growth following radiosurgery compared with Grades I to III (p = 0.02). Treatment failure occurred in nine patients (8%) at a median time to failure of 29 (range: 4–70) months. Five patients with treatment failure had progressive neurologic symptoms. The actuarial rates of tumor control (no additional surgical intervention required following radiosurgery) at 1, 3, 5, and 7 years were 96, 94, 88, and 88%, respectively.

Conclusion CKRS affords effective tumor control for acoustic neuromas with an acceptable rate of hearing preservation. KOOS Grade IV tumors were significantly less likely to respond to CKRS than Grade I to III tumors and should be strongly considered for surgical resection when possible. Further studies are needed to compare the clinical and radiographic outcomes of CKRS to single fraction radiosurgery for acoustic neuromas.