Radiosurgical Treatment of Vestibular Schwannomas in Patients With Neurofibromatosis Type 2

Transcription

1 Radiosurgical Treatment of Vestibular Schwannomas in Patients With Neurofibromatosis Type 2 Tumor Control and Hearing Preservation Ji Hoon Phi, MD 1, Dong Gyu Kim, MD, PhD 1, Hyun-Tai Chung, PhD 1, Joongyub Lee, MD 2, Sun Ha Paek, MD, PhD 1, and Hee-Won Jung, MD, PhD 1 BACKGROUND: The radiosurgical treatment of vestibular schwannomas in patients with neurofibromatosis type 2 (NF2) is controversial. The authors investigated the radiologically proven tumor control rate after gamma knife radiosurgery. The factors that affect tumor control and serviceable hearing preservation were analyzed. METHODS: Thirty-six lesions in 30 patients were included. The median lengths of the clinical and radiologic follow-ups were 48.5 months and 36.5 months, respectively. The median tumor volume was 3.2 cm 3. The mean marginal dose was 12.1 grays (Gy) (range, 8-14 Gy) at an isodose line of 50% 0.6%. The Kaplan-Meier method and Cox proportional hazards model were used for the statistical analyses. RESULTS: The actuarial tumor control rate was 81%, 74%, and 66%, respectively, in the first, second, and fifth years. Five tumors required a salvage surgery because of tumor control failure. A low marginal dose and a young age at radiosurgery were associated with poor tumor control. Of the 16 tumors with which ipsilateral hearing was serviceable, the actuarial serviceable hearing preservation rates were 50%, 45%, and 33%, respectively, in the first, second, and fifth years. Better ipsilateral hearing (Gardner-Robertson grade 1, compared with grade 2) at the time of radiosurgery was associated with significantly greater serviceable hearing preservation. CON- CLUSIONS: Gamma knife radiosurgery for vestibular schwannomas in NF2 patients provided 5-year tumor control in approximately two-thirds of patients and preserved serviceable hearing in approximately one-third. The rates of other cranial nerve deficits were low, and no secondary malignancy was observed. Radiosurgery should be included in treatment options for NF2 patients. Cancer 2009;115: VC 2009 American Cancer Society. KEY WORDS: vestibular schwannoma, neurofibromatosis type 2, gamma knife radiosurgery, hearing preservation. Neurofibromatosis type 2 (NF2) is a tumor-prone genetic disorder. 1 The hallmark of NF2 is bilateral vestibular schwannomas, which impose a significant risk of becoming deaf at a young age. 1,2 Since the discovery of the gene disrupted in NF2, NF2, much has been discovered regarding the molecular pathogenesis of the disease. 1-4 However, the clinical management of NF2 patients is still a challenging task. Although the Corresponding author: Dong Gyu Kim, MD, PhD, Department of Neurosurgery, Seoul National University College of Medicine, 101 Daehangno, Jongno-gu, Seoul , Republic of Korea; Fax: (011) ; gknife@plaza.snu.ac.kr 1 Department of Neurosurgery, Seoul National University College of Medicine, Seoul, Republic of Korea; 2 Department of Preventive Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea Received: April 14, 2008; Revised: August 24, 2008; Accepted: August 26, 2008 Published online: December 24, 2008, VC 2008 American Cancer Society DOI: /cncr.24036, Cancer January 15, 2009

2 Radiosurgery for Vestibular Schwannoma/Phi et al total surgical resection of vestibular schwannomas may result in definite tumor control, it is associated with a high probability of hearing loss and other cranial nerve damage, because the tumors tend to be multilobulated and to infiltrate the cochlear nerve in NF2 patients. 2,5 Moreover, the clinical burden of other NF2-associated tumors should be considered before radical surgery is undertaken. 2 Radiosurgery is now regarded as 1 of the first-line treatment options for sporadic vestibular schwannomas. 6-8 To our knowledge to date, only a few series of radiosurgery for NF2-associated vestibular schwannomas have been reported The reported outcomes have been promising, despite the obvious inconsistency with the outcomes of sporadic vestibular schwannomas. Local tumor control rates, defined as the rate of avoiding surgical intervention after radiosurgery, are reported to be approximately 80%, with hearing preservation rates of approximately 40%. 10,11 However, the radiologically proven tumor control rate was only 50% in 1 series. 11 Therefore, more precise information on the efficacy of radiosurgery in NF2 patients is required. In this study, we retrospectively analyzed the radiosurgical outcomes of NF2 patients who had undergone gamma knife radiosurgery for vestibular schwannomas. The radiologically proven tumor control rate was calculated, and a strategy to enhance the preservation of serviceable hearing is proposed. MATERIALS AND METHODS Patients The inclusion criteria were as follows: an NF2-associated vestibular schwannoma treated by gamma knife radiosurgery at our institution, and clinical and radiologic followups of at least 1 year. Between December 1997 and June 2006, 401 vestibular schwannomas were treated by gamma knife radiosurgery at our institution, and 45 of these were associated with NF2. Nine tumors were excluded because of insufficient follow-up. Thus, 36 tumors in 30 patients were included. Sixteen patients were male, and 14 were female. Twenty-eight patients had bilateral vestibular schwannomas. Two patients had a unilateral tumor but met the diagnostic criteria for NF2. 2 In 6 patients, radiosurgery was performed bilaterally for vestibular schwannomas. The median age at the time of radiosurgery was 32.5 years (range, years). For all patients, basal otological examinations including pure tone audiometry (PTA) and speech discrimination scores (SDS) were performed before radiosurgery. In 22 tumors, the affected ear retained hearing capacity before radiosurgery. The hearing was serviceable (ie, Gardner-Robertson [G-R] grades 1 and 2) with 16 tumors (9 grade 1 tumors and 7 grade 2 tumors). 13 Hearing was nonserviceable (G-R grades 3 and 4) with the other 6 tumors. The indications for radiosurgery were as follows. For tumors with ipsilateral hearing retained (22 tumors), radiosurgery was performed primarily when the tumor was growing continuously with ipsilateral hearing deterioration (17 tumors). However, radiosurgery was performed for 4 growing tumors without hearing deterioration. In these cases, contralateral tumors were smaller, and contralateral hearing was intact. One patient received radiosurgery to a residual tumor after intended subtotal resection for hearing preservation. In 14 tumors, the ipsilateral ear was deaf at the time of radiosurgery. Radiosurgery was performed when the tumor was growing continuously with or without neurologic symptoms (2 tumors and 4 tumors, respectively) or when the tumor was incompletely removed by microsurgery (8 tumors). In all, 27 (75%) tumors were treated primarily with radiosurgery, and 9 (25%) tumors were initially treated with microsurgery, followed by radiosurgery. Radiosurgical Techniques The Leksell Model G Stereotactic frame (Elekta Instruments AB, Stockholm, Sweden) was applied to the patient s head under local anesthesia. Spoiled gradientrecalled echo magnetic resonance image (MRI) sequences were obtained with and without contrast. Treatment planning was undertaken with Leksell GammaPlan software (Elekta Instruments AB). Multiple isocenter planning was applied to minimize the radiation exposure to adjacent structures. The median tumor volume was 3.2 cm 3 (range, cm 3 ). The mean marginal dose was 12.1 grays (Gy) (range, Gy) at an isodose line of 50% 0.6%, and the mean maximal dose was 24.4 Gy (range, Gy). At the early stage of our Cancer January 15,

3 experience, the radiation dose was selected mainly according to the tumor volume. After 2 to 3 years of experience, we excluded larger tumors that required a marginal dose <11 Gy. We also noticed that the hearing status after radiosurgery was affected by the marginal dose from the treatment experience of sporadic tumors. Thereafter, the dose was determined primarily by the hearing status. If hearing was present and serviceable in the affected ear, we applied <13 Gy. If hearing was completely lost in the affected ear, we applied 13 Gy. The gross tumor volume without expansion was used as the planning target volume. The median number of shots was 11 (range, 4-17 shots). Follow-up and Outcome Analysis Regular clinical follow-up was offered to all patients, beginning 1 month from radiosurgery. Brain MRIs were taken at 6 month and 1 year, and thereafter annual scheduled MRIs were recommended. Otologic examinations, including PTA and SDS, were recommended every 6 months. If patients reported symptomatic deterioration, MRI and/or an otologic examination were promptly recommended. Tumor growth was assessed independently by an attending neuroradiologist and the authors. Tumor volumes were calculated using Osiris 3.0 (University Hospital of Geneve, Geneva, Switzerland). Volume changes (either increases or decreases) 25% were considered meaningful. Volume changes <25% were regarded as stable volumes. The median length of the clinical follow-up was 48.5 months (range, months), and the median length of the radiologic follow-up was 36.5 months (range, 4-99 months). One patient was followed radiologically for <1 year (4 months), because the patient experienced rapid tumor growth after radiosurgery and underwent a salvage surgery 4 months after radiosurgery. Three aspects of the treatment outcomes were measured: tumor control; serviceable hearing preservation; and radiosurgical complications. Survival analyses were performed on tumor control and serviceable hearing preservation. The factors affecting tumor control and serviceable hearing preservation were analyzed with the Kaplan-Meier method and the Cox proportional hazards model. For univariate analyses, the significance level was 5%. Patient factors (age at the time of radiosurgery, sex, FIGURE 1. A Kaplan-Meier plot illustrates the crude cumulative survival of tumor control after gamma knife radiosurgery. The actuarial tumor control rates were 81%, 74%, and 66%, respectively, in the first, second, and fifth years. hearing capacity, the side of the tumor, and bilateral/unilateral treatment) and treatment-related factors (tumor volume, marginal dose, cochlear dose, and number of shots) were included in the analyses. Cochlear doses (maximum, mean, and minimum values) were measured with Leksell GammaPlan software, based on the MRI scans of each radiosurgical plan. RESULTS On the basis of MRI volumetric assessments, tumor volume was found to have decreased in 14 tumors (39%), was stable in 11 tumors (31%), and increased in 11 tumors (31%) at the time of last follow-up (for tumors removed by a salvage surgery, at the time of the salvage surgery). To calculate the crude cumulative incidence of tumor control, the endpoint was defined as the time when tumor control failure was evident radiologically (an increase 25% of the tumor volume) and/or clinically (salvage surgery because of symptomatic deterioration). For 36 tumors treated, the actuarial tumor control rate was 81% in the first year, 74% in the second year, and 66% in the fifth year after radiosurgery (Fig. 1). Five tumors required a salvage surgery because of neurologic symptoms after the failure of radiosurgical tumor control. 392 Cancer January 15, 2009

4 Radiosurgery for Vestibular Schwannoma/Phi et al Table 1. Relative Risks for the Loss of Tumor Control Estimated with the Cox Proportional Hazards Model Factors Univariate Analysis Multivariate Analysis P Crude RR 95% CI P Adjusted RR 95% CI Age at the time of radiosurgery Tumor volume Shot Marginal dose RR indicates relative risk; 95% CI, 95% confidence interval. FIGURE 2. A graph represents changes in hearing status after radiosurgery in 16 patients with preradiosurgery serviceable hearing in the affected ear. G-R indicates Gardner-Robertson. The actuarial rate of salvage surgery was 6% in the first year and 19% in the fifth year. In a univariate analysis with the Cox proportion hazards model, no patient or treatment-related factors were found to be statistically significant. In a multivariate analysis with the same model, patient age at the time of radiosurgery and the marginal dose were related to tumor control failure (Table 1). The older the patient, and the greater the marginal dose given, the more chance of tumor control in the patient (relative risk of tumor control failure of 0.92/year and 0.40/Gy, respectively). Ipsilateral hearing was in the serviceable range at the time of radiosurgery for 16 tumors. When the endpoint was defined as the moment when the ipsilateral hearing became nonserviceable or permanently lost (G-R grade, 3-5), the actuarial serviceable hearing preservation rate was 50% in the first year, 45% in the second year, and 33% in the fifth year after radiosurgery. Of the 9 tumors with a preradiosurgery ipsilateral hearing capacity of G-R grade 1, serviceable hearing was preserved at the last follow-up in 6 (5 at grade 1 and 1 at grade 2). The median follow-up period of the 5 tumors with preserved G-R grade 1 hearing was 44 months FIGURE 3. Survival functions for serviceable hearing preservation are plotted according to the hearing capacity at the time of radiosurgery. If the ipsilateral hearing was good (Gardner-Robertson [G-R] grade 1) at the time of radiosurgery, there was a significantly greater chance of hearing preservation, compared with patients with ipsilateral hearing of G-R grade 2 (P ¼.006, log-rank test). (range, months). Of the 7 tumors with preradiosurgery hearing capacity of G-R grade 2, none retained serviceable hearing after radiosurgery (Fig. 2). The difference in hearing preservation according to hearing capacity was significant (P ¼.006, log-rank test) (Fig. 3). Univariate analysis with a Cox proportional hazards model revealed that preradiosurgery hearing capacity (assessed with G-R grades) was the single significant factor influencing the hearing outcome (Table 2). If ipsilateral hearing was G-R grade 2 at the time of radiosurgery, there was a significantly greater risk of serviceable hearing loss, compared with those tumors associated with ipsilateral hearing G-R grade 1 (relative risk of 5.72). Higher maximal cochlear Cancer January 15,

5 Table 2. Relative Risks for the Loss of Serviceable Hearing Estimated With the Cox Proportional Hazards Model Factors Univariate Analysis Multivariate Analysis P Crude RR 95% CI P Adjusted RR 95% CI G-R grade before radiosurgery Age at the time of radiosurgery Tumor volume Maximal cochlear dose RR indicates relative risk; 95% CI, 95% confidence interval; G-R, Gardner-Robertson. FIGURE 4. (A) A magnetic resonance imaging scan of an 18-year-old patient with neurofibromatosis type 2 is shown. The tumor on the left side was larger (2.6 cm 3 ). Bilateral hearing was intact and the pure tone audiometry (PTA) threshold and speech discrimination scores (SDS) for the left side were 13 decibels (db) and 100%, respectively. (B) Six months later, the tumor on the left side had grown to 3.0 cm 3, with intact hearing (PTA threshold of 10 db and SDS of 100%). Gamma knife radiosurgery was performed. (C) Six months later, tumor expansion and central loss of enhancement were observed. The tumor volume was 3.4 cm 3, and hearing was still intact (PTA of 12 db and SDS of 100%). (D) Eleven months after radiosurgery, the patient reported hearing deterioration and tinnitus on the left side. The tumor volume had increased slightly (3.5 cm 3 ), and ipsilateral hearing had deteriorated to Gardner-Robertson (G-R) grade 2 (PTA threshold of 36 db and SDS of 88%). Short-term oral steroids were prescribed. (E) Twenty-two months after radiosurgery, the tumor volume decreased to 3.3 cm 3, and tinnitus had disappeared. (F) Thirty-four months after radiosurgery, the tumor had shrunk to 2.5 cm 3, and hearing had returned to G-R grade 1 (PTA threshold of 20 db and SDS of 100%). dose showed a tendency toward adverse hearing outcome without significance. In a multivariate analysis with the same model, preradiosurgery hearing capacity was related to the hearing preservation outcome without significance. In both univariate and multivariate analyses, the confidence intervals for preradiosurgery hearing capacity were rather broad because of the small number of subjects, and this may limit the statistical meaning of the results. In 2 tumors with ipsilateral preradiosurgery hearing of G-R grade 1, transient hearing deterioration occurred because of transient tumor expansion. The prescription of short-term oral corticosteroids was helpful in these cases for the eventual restoration of hearing to G-R grade 1 (Fig. 4). Two patients developed facial nerve palsy of House- Brackman grade 2. One was transient, and the other was permanent. Trigeminal neuropathic pain developed in 1 patient. The pain was made tolerable with the administration of oral carbamazepine. Transient vestibular dysfunction was observed in 1 patient (Table 3). No secondary malignancy developed in the follow-up period. DISCUSSION Radiosurgical treatment of vestibular schwannomas in NF2 patients is a controversial issue. In the English literature, 2 large series provide relatively long-term follow-up data on this treatment. In the study by Rowe et al, 11 the local control rate was 74% after a mean clinical follow-up of 50 months. In the recent study by Mathieu et al, 10 the actuarial local control rate was 85% in the fifth year and 81% in the 15th year. However, the local control rate, defined as the chance of avoiding surgical intervention, in these studies is not an optimal parameter with which to 394 Cancer January 15, 2009

6 Radiosurgery for Vestibular Schwannoma/Phi et al Table 3. Radiosurgical Complications Reported in This Study and Other Series Current Study Rowe et al 11 * Mathieu et al 10 Low-dose Group High-dose Group Facial neuropathyy 5.6% 7.5% 33.3% 16.2% Trigeminal neuropathyy 2.8% 3.2% 23.3% 12.2% Mean margin dose 12.1 Gy 13.4 Gy Gyz 14 Gy Gy indicates grays. * Rowe et al reported that their treatment policy changed from high-dose radiation to low-dose radiation, and they analyzed the outcomes of the groups separately. y All figures are crude percentages, and they include both transient and permanent neuropathies. z Not a mean but a range. evaluate outcomes. Instead, tumor volumetry has generally been used for outcome evaluation for benign intracranial tumors. In this study, the actuarial tumor control rate, measured by serial MRI volumetry, was 74% in the second year and 66% in the fifth year. These values are obviously lower than those for sporadic vestibular schwannomas and the local control rates presented by the series mentioned above. 7,10,11 Why vestibular schwannomas of NF2 patients respond less well than sporadic tumors is not known. Genetic background, multilobulated tumor stereology, and younger patient age may play a role in the radiobiology of NF2-associated vestibular schwannomas. In particular, younger patient age was found to be a predictor of poor tumor control in this and other studies. 10 It is known that specific mutations in the NF2 gene are associated with younger age at onset and the overall severity of the disease. 14,15 These genotype-phenotype correlations may also extend to treatment responses. Surgical resection is the standard of care for benign intracranial tumors. However, hearing preservation rates after radical surgery in NF2-associated vestibular schwannomas are lower and the risks of postoperative cranial nerve deficits are higher than those of sporadic vestibular schwannomas. 5,16-18 Furthermore, the bilateral nature of vestibular schwannomas and the presence of other tumors in the central nervous system can increase the likelihood of other interventions and neurologic signs and symptoms in these patients. In this and other studies, radiosurgery has been reported to result in relatively long-term hearing preservation in 33% to 53% of NF2 patients. 10,11 In the current study, patients with better ipsilateral hearing function appear to have had better hearing preservation. The benefits of early intervention are noteworthy, because conservative approaches consisting of observation, with surgical intervention as the last resort when deafness is evident, have predominated in the management of NF2 patients. Early proactive radiosurgical treatment when hearing begins to deteriorate or when the tumor grows steadily without hearing impairment can be a strategy to increase the chance of hearing preservation, but only if contralateral hearing of the patient is intact. However, some points should be considered in the interpretation of this result. First, this finding is not specific to radiosurgery of NF2 patients, because earlier treatment of a smaller vestibular schwannoma, whether it is sporadic or NF2-associated, with any modalities such as radiosurgery, radiotherapy, and microsurgery generally leads to a better hearing outcome. 5,19-21 Second, more data regarding this issue must be collected because the statistical power of the present study was not convincingly high. Last, the apparent benefits of early treatment may reflect a well-known bias, that is, the patients diagnosed and treated earlier only appear to do better, although the long-term outcome is in fact unchanged. A longitudinal study demonstrated that the hearing of patients with a vestibular schwannoma declined more rapidly if the initial hearing was poorer. 22 Furthermore, the same study showed that diagnosis of NF2 had a protective effect on hearing deterioration. Therefore, it is plausible that there is a sufficient time lag before hearing of G-R grade 1 deteriorates to grade 2 in NF2 patients. However, the study also likely reflects the same bias, because vestibular schwannomas are diagnosed earlier in NF2 patients than in sporadic patients with screening examinations, and an apparent protective effect of NF2 may be observed. 22,23 An audiometric follow-up study in a large group of NF2 Cancer January 15,

7 patients demonstrated that hearing function deteriorated after 2 years of observation, and that initial normal mean PTA value increased to over 30 decibels after 2 years (this PTA value corresponds to G-R grade 2 or higher). 23 In the current study, the follow-up periods of the tumors with preserved G-R grade 1 hearing were all >2 years (median, 44 months; range, months). This fact indirectly supports that the results were not severely biased. However, a long-term follow-up study is required to control the bias, although a controlled trial on this uncommon genetic disease is hardly expected to be available in the near future. In the study by Mathieu et al, 10 the mean marginal dose given was 14 Gy, and marginal doses <14 Gy were associated with better serviceable hearing preservation. Because a relatively low marginal dose (mean, 12.1 Gy) was given, and no tumor with serviceable hearing in the affected ear received >13 Gy, the marginal dose was not believed to be a significant factor affecting the hearing outcome in the present study. It is well known that lowdose (12-13 Gy) radiosurgery results in better hearing preservation for sporadic vestibular schwannomas. 24,25 Therefore, it could be a reasonable radiosurgical strategy in NF2 patients to apply low doses (12-13 Gy) for hearing preservation and slightly higher doses to tumors without ipsilateral hearing for better tumor control. The usefulness of cochlear doses as a predicator of hearing preservation after radiosurgery and radiotherapy has been reported by many authors Although we observed no significant effect of cochlear doses in the current study, this issue deserves further investigation in a larger series. We experienced 2 patients whose serviceable hearing deteriorated during transient tumor expansion, but was eventually preserved with short-term prescription of corticosteroid. Transient tumor expansion with loss of contrast enhancement is a common finding after radiosurgery and is frequently followed by tumor shrinkage Therefore, it is regarded as evidence of radiosurgical effect. 29,30 However, cranial nerve damage can develop during the period of transient tumor expansion in some patients. 28,29 Because the cochlear nerves of NF2 patients appear to be quite vulnerable to injury, close follow-up with serial MRI and audiometry and timely intervention are required after radiosurgery. Low-dose treatment appears to lower the cranial nerve complication rate, as it does for sporadic vestibular schwannomas. 25 Facial neuropathy, both transient and permanent, occurred in 2.8% of tumor-affected sides in this study. This figure is similar to that of the low-dose treatment group in the study by Rowe et al 11 (Table 3). Fractionated stereotactic radiotherapy (FSRT) is an alternative therapeutic modality for vestibular schwannomas that yields local tumor control rates comparable to those achieved by radiosurgery and shows lower cranial nerve toxicity rates, especially of cochlear nerves. 6,21,32-34 Combs et al 33 reported that an actuarial 5-year local tumor control rate was 93% after FSRT. Interestingly, NF2 was significantly associated with adverse outcomes in this study, with an actuarial 5-year hearing preservation rate of 64% in NF2 patients, in contrast to 98% in non-nf2 patients. In the study of Andrews et al, 21 the tumor control rate after FSRT was lower in NF2 patients than in non-nf2 patients (67% and 97%, respectively). In contrast, Chan et al 6 observed no difference in tumor control and hearing preservation between NF2 and non-nf2 patients. Because the numbers of NF2 patients in these studies were relatively small, a larger series focusing on NF2 patients would be helpful to define the role of FSRT in NF2 patients. Although we did not observe any occurrence of radiation-induced secondary malignancy, it is still a great concern for many neuro-oncologists. There are several case reports of malignant transformation of NF2-associated vestibular schwannomas after radiosurgery. 35,36 It has also been reported that radiation-induced tumors can be easily mistaken for de novo tumors in the context of NF2. 37 Conversely, Rowe et al 38 reported that the risk of secondary malignancy in NF2 patients is not as formidable as previously considered. Because the follow-up period in this study was not long, and many young patients were included, careful observation and reevaluation of the risks are required. Conclusions Gamma knife radiosurgery for vestibular schwannoma in NF2 patients provided 5-year tumor control in approximately two-thirds of patients and preserved serviceable hearing in approximately one-third. The rates of facial and trigeminal nerve deficits were low with application of low marginal doses, and no secondary malignancy was observed in these patients. Early intervention before 396 Cancer January 15, 2009

8 Radiosurgery for Vestibular Schwannoma/Phi et al profound hearing impairment can be a strategy to increase the chance of hearing preservation. Transient tumor expansion after radiosurgery causes hearing deterioration in some patients, and vigilant follow-up and timely intervention can be helpful for hearing preservation. Radiosurgery should be included in the treatment options for NF2 patients as a less invasive treatment modality. Conflict of Interest Disclosures Supported in part by the Korea Ministry of Education, Science and Technology, with study number M M References 1. Baser ME, DG RE, Gutmann DH. Neurofibromatosis 2. Curr Opin Neurol. 2003;16: Ferner RE. Neurofibromatosis 1 and neurofibromatosis 2: a twenty first century perspective. Lancet Neurol. 2007;6: Trofatter JA, MacCollin MM, Rutter JL, et al. A novel moesin-, ezrin-, radixin-like gene is a candidate for the neurofibromatosis 2 tumor suppressor. Cell. 1993;72: Xiao GH, Chernoff J, Testa JR. NF2: the wizardry of merlin. Genes Chromosomes Cancer. 2003;38: Samii M, Matthies C, Tatagiba M. Management of vestibular schwannomas (acoustic neuromas): auditory and facial nerve function after resection of 120 vestibular schwannomas in patients with neurofibromatosis 2. Neurosurgery. 1997;40: Chan AW, Black P, Ojemann RG, et al. Stereotactic radiotherapy for vestibular schwannomas: favorable outcome with minimal toxicity. Neurosurgery. 2005;57: Kondziolka D, Lunsford LD, McLaughlin MR, Flickinger JC. Long-term outcomes after radiosurgery for acoustic neuromas. N Engl J Med. 1998;339: Pollock BE, Driscoll CL, Foote RL, et al. Patient outcomes after vestibular schwannoma management: a prospective comparison of microsurgical resection and stereotactic radiosurgery. Neurosurgery. 2006;59: Kida Y, Kobayashi T, Tanaka T, Mori Y. Radiosurgery for bilateral neurinomas associated with neurofibromatosis type 2. Surg Neurol. 2000;53: Mathieu D, Kondziolka D, Flickinger JC, et al. Stereotactic radiosurgery for vestibular schwannomas in patients with neurofibromatosis type 2: an analysis of tumor control, complications, and hearing preservation rates. Neurosurgery. 2007;60: Rowe JG, Radatz MW, Walton L, Soanes T, Rodgers J, Kemeny AA. Clinical experience with gamma knife stereotactic radiosurgery in the management of vestibular schwannomas secondary to type 2 neurofibromatosis. J Neurol Neurosurg Psychiatry. 2003;74: Subach BR, Kondziolka D, Lunsford LD, Bissonette DJ, Flickinger JC, Maitz AH. Stereotactic radiosurgery in the management of acoustic neuromas associated with neurofibromatosis Type 2. J Neurosurg. 1999;90: Gardner G, Robertson JH. Hearing preservation in unilateral acoustic neuroma surgery. Ann Otol Rhinol Laryngol. 1988;97: Baser ME, Kuramoto L, Woods R, et al. The location of constitutional neurofibromatosis 2 (NF2) splice site mutations is associated with the severity of NF2. J Med Genet. 2005;42: Evans DG, Trueman L, Wallace A, Collins S, Strachan T. Genotype/phenotype correlations in type 2 neurofibromatosis (NF2): evidence for more severe disease associated with truncating mutations. J Med Genet. 1998;35: Baldwin D, King TT, Chevretton E, Morrison AW. Bilateral cerebellopontine angle tumors in neurofibromatosis type 2. J Neurosurg. 1991;74: Evans DG, Baser ME, O Reilly B, et al. Management of the patient and family with neurofibromatosis 2: a consensus conference statement. Br J Neurosurg. 2005;19: Moffat DA, Quaranta N, Baguley DM, Hardy DG, Chang P. Management strategies in neurofibromatosis type 2. Eur Arch Otorhinolaryngol. 2003;260: Brackmann DE, Fayad JN, Slattery WH, et al. Early proactive management of vestibular schwannomas in neurofibromatosis type 2. Neurosurgery. 2001;49: Doherty JK, Friedman RA. Controversies in building a management algorithm for vestibular schwannomas. Curr Opin Otolaryngol Head Neck Surg. 2006;14: Andrews DW, Suarez O, Goldman HW, et al. Stereotactic radiosurgery and fractionated stereotactic radiotherapy for the treatment of acoustic schwannomas: comparative observations of 125 patients treated at 1 institution. Int J Radiat Oncol Biol Phys. 2001;50: Massick DD, Welling DB, Dodson EE, et al. Tumor growth and audiometric change in vestibular schwannomas managed conservatively. Laryngoscope. 2000;110: Masuda A, Fisher LM, Oppenheimer ML, Iqbal Z, Slattery WH. Hearing changes after diagnosis in neurofibromatosis type 2. Otol Neurotol. 2004;25: Chopra R, Kondziolka D, Niranjan A, Lunsford LD, Flickinger JC. Long-term follow-up of acoustic schwannoma radiosurgery with marginal tumor doses of 12 to 13 Gy. Int J Radiat Oncol Biol Phys. 2007;68: Paek SH, Chung HT, Jeong SS, et al. Hearing preservation after gamma knife stereotactic radiosurgery of vestibular schwannoma. Cancer. 2005;104: Massager N, Nissim O, Delbrouck C, et al. Irradiation of cochlear structures during vestibular schwannoma radiosurgery and associated hearing outcome. J Neurosurg. 2007; 107: Cancer January 15,

9 27. Thomas C, Di Maio S, Ma R, et al. Hearing preservation following fractionated stereotactic radiotherapy for vestibular schwannomas: prognostic implications of cochlear dose. J Neurosurg. 2007;107: Phi JH, Paek SH, Chung HT, et al. Gamma knife surgery and trigeminal schwannoma: is it possible to preserve cranial nerve function? J Neurosurg. 2007;107: Pollock BE. Management of vestibular schwannomas that enlarge after stereotactic radiosurgery: treatment recommendations based on a 15 year experience. Neurosurgery. 2006;58: Regis J, Pellet W, Delsanti C, et al. Functional outcome after gamma knife surgery or microsurgery for vestibular schwannomas. J Neurosurg. 2002;97: Yu CP, Cheung JY, Leung S, Ho R. Sequential volume mapping for confirmation of negative growth in vestibular schwannomas treated by gamma knife radiosurgery. J Neurosurg. 2000;93: Chung HT, Ma R, Toyota B, Clark B, Robar J, McKenzie M. Audiologic and treatment outcomes after linear accelerator-based stereotactic irradiation for acoustic neuroma. Int J Radiat Oncol Biol Phys. 2004;59: Combs SE, Volk S, Schulz-Ertner D, Huber PE, Thilmann C, Debus J. Management of acoustic neuromas with fractionated stereotactic radiotherapy (FSRT): long-term results in 106 patients treated in a single institution. Int J Radiat Oncol Biol Phys. 2005;63: Fuss M, Debus J, Lohr F, et al. Conventionally fractionated stereotactic radiotherapy (FSRT) for acoustic neuromas. Int J Radiat Oncol Biol Phys. 2000;48: Bari ME, Forster DM, Kemeny AA, Walton L, Hardy D, Anderson JR. Malignancy in a vestibular schwannoma. Report of a case with central neurofibromatosis, treated by both stereotactic radiosurgery and surgical excision, with a review of the literature. Br J Neurosurg. 2002;16: Thomsen J, Mirz F, Wetke R, Astrup J, Bojsen-Moller M, Nielsen E. Intracranial sarcoma in a patient with neurofibromatosis type 2 treated with gamma knife radiosurgery for vestibular schwannoma. Am J Otol. 2000;21: Evans DG, Birch JM, Ramsden RT, Sharif S, Baser ME. Malignant transformation and new primary tumours after therapeutic radiation for benign disease: substantial risks in certain tumour prone syndromes. J Med Genet. 2006;43: Rowe J, Grainger A, Walton L, Radatz M, Kemeny A. Safety of radiosurgery applied to conditions with abnormal tumor suppressor genes. Neurosurgery. 2007;60: Cancer January 15, 2009