P11.73.B COMPARING GAMMA KNIFE RADIOSURGERY WITH MICROSURGERY AND THEIR RISK-BENEFIT-RATIO OF TUMOR CONTROL AND FACIAL DETERIORATION IN LARGE VESTIBULAR SCHWANNOMAS (KOOS III-IV)

Abstract

Abstract BACKGROUND The level of evidence to provide treatment recommendations for vestibular schwannoma (VS) is low. Both stereotactic radiosurgery (SRS) and microsurgical resection (SURGERY) are available as treatment options for VS. If treated with monotherapy alone, SRS has shown to be inferior in tumor control, but superior in facial nerve preservation compared to SURGERY. The aim of this study was to analyze the risk-benefit-ratio in the treatment of large VS (Koos III-IV) comparing SRS and SURGERY. MATERIAL AND METHODS This is a retrospective dual-center cohort study enrolling consecutive patients with solitaryVS between 2005 und 2011 to enable follow-up of up to 10 years. Tumor size was classified by Koos Classification. Clinical state was reported by House and Brackmann (H&B) and Gardner-Robertson (G&R) scale (with H&B and G&R 1-2 considered as good outcome), and Recurrence-free-survival (RFS) was assessed radiographically by contrast-enhanced MR imaging. As an expression of success/harm, the absolute risk reduction (ARR) and risk increase (ARI) were calculated as the difference between incidence of recurrences and facial deterioration rates in patients treated with SRS and SURGERY, respectively. ARR and ARI were used to derive additional estimates of treatment effectiveness (in SURGERY), i.e. the number-to-operate (NNO) and the number-to-harm (NNH), respectively. RESULTS The study population included N=901 patients, who were treated in both centers. N=492 (55%) were large VS (Koos III-IV). Of them N=260 (53%) were treated with SRS and N=232 (47%) with SURG. The SRS treatment arm presented with a recurrence rate of 14.2%, and SURGERY with 3.0%. Therefore, ARR for recurrence was 11.21% (95% CI: 6.4%-16.0%) with a NNO of 9 (95% CI: 6.3-15.6) in SURGERY meaning that one in every 9 patients will benefit from SURGERY treatment. When analyzing facial nerve deterioration (facial paresis HB>2 and facial spasm) in large VS, SRS presented with 6.9%, and SURGERY with 12.5% adverse facial nerve outcome resulting in an ARI of 5.6% (95%CI: 0.3%-10.8%) and a NNH of 18 (95%CI:9.2-312.1). CONCLUSION The analysis of the different proportion in success/harm in the management of large VS (SRS vs. SURGERY) in this cohort yielded in a low NNO, combined with a high NNH. This shows that SURGERY was superior to SRS considering tumor control in large VS (Koos III-IV). Still, the direct translation of these results into treatment recommendations is not self-evident, but multi-faceted: This analysis is a vivid representation of a dilemma well-known to neuro-oncological surgeons - the negotiation between tumor control and functional outcome. Large VS should be treated in specialized centers, which have enough experience to ensure a high-rate of facial preservation in large VS (Koos III-IV).