Radiosurgery of parasellar meningiomas

Gamma Knife radiosurgery for sellar and parasellar meningiomas: a multicenter study Clinical article J Neurosurg 120:1268 1277, 2014 AANS, 2014 Jason P. Sheehan, M.D., Ph.D., 1 Robert M. Starke, M.D., M.Sc., 1 Hideyuki Kano, M.D., Ph.D., 2 Anthony M. Kaufmann, M.D., 3 David Mathieu, M.D., 4 Fred A. Zeiler, M.D., 3 Michael West, M.D., Ph.D., 3 Samuel T. Chao, M.D., 5 Gandhi Varma, M.D., 5 Veronica L. S. Chiang, M.D., 6 James B. Yu, M.D., 6 Heyoung L. McBride, M.D., M.S., 7 Peter Nakaji, M.D., 7 Emad Youssef, M.D., 7 Norissa Honea, Ph.D., 7 Stephen Rush, M.D., 10 Douglas Kondziolka, M.D., 10 John Y. K. Lee, M.D., 8 Robert L. Bailey, M.D., 8 Sandeep Kunwar, M.D., 9 Paula Petti, Ph.D., 9 and L. Dade Lunsford, M.D. 2 1 University of Virginia, Charlottesville, Virginia; 2 University of Pittsburgh, Pennsylvania; 3 University of Manitoba, Winnipeg, MB; 4 University of Sherbrooke, QC, Canada; 5 Cleveland Clinic, Cleveland, Ohio; 6 Yale University, New Haven, Connecticut; 7 Barrow Neurological Institute, Phoenix, Arizona; 8 University of Pennsylvania, Philadelphia, Pennsylvania; 9 Taylor McAdam Bell Neuroscience Institute, Washington Hospital Healthcare System, Fremont, California; and 10 New York University, New York, New York Object. Parasellar and sellar meningiomas are challenging tumors owing in part to their proximity to important neurovascular and endocrine structures. Complete resection can be associated with significant morbidity, and incomplete resections are common. In this study, the authors evaluated the outcomes of parasellar and sellar meningiomas managed with Gamma Knife radiosurgery (GKRS) both as an adjunct to microsurgical removal or conventional radiation therapy and as a primary treatment modality. Methods. A multicenter study of patients with benign sellar and parasellar meningiomas was conducted through the North American Gamma Knife Consortium. For the period spanning 1988 to 2011 at 10 centers, the authors identified all patients with sellar and/or parasellar meningiomas treated with GKRS. Patients were also required to have a minimum of 6 months of imaging and clinical follow-up after GKRS. Factors predictive of new neurological deficits following GKRS were assessed via univariate and multivariate analyses. Kaplan-Meier analysis and Cox multivariate regression analysis were used to assess factors predictive of tumor progression. Results. The authors identified 763 patients with sellar and/or parasellar meningiomas treated with GKRS. Patients were assessed clinically and with neuroimaging at routine intervals following GKRS. There were 567 females (74.3%) and 196 males (25.7%) with a median age of 56 years (range 8 90 years). Three hundred fifty-five patients (50.7%) had undergone at least one resection before GKRS, and 3.8% had undergone prior radiation therapy. The median follow-up after GKRS was 66.7 months (range 6 216 months). At the last follow-up, tumor volumes remained stable or decreased in 90.2% of patients. Actuarial progression-free survival rates at 3, 5, 8, and 10 years were 98%, 95%, 88%, and 82%, respectively. More than one prior surgery, prior radiation therapy, or a tumor margin dose < 13 Gy significantly increased the likelihood of tumor progression after GKRS. At the last clinical follow-up, 86.2% of patients demonstrated no change or improvement in their neurological condition, whereas 13.8% of patients experienced symptom progression. New or worsening cranial nerve deficits were seen in 9.6% of patients, with cranial nerve (CN) V being the most adversely affected nerve. Functional improvements in CNs, especially in CNs V and VI, were observed in 34% of patients with preexisting deficits. New or worsened endocrinopathies were demonstrated in 1.6% of patients; hypothyroidism was the most frequent deficiency. Unfavorable outcome with tumor growth and accompanying neurological decline was statistically more likely in patients with larger tumor volumes (p = 0.022) and more than 1 prior surgery (p = 0.021). Conclusions. Gamma Knife radiosurgery provides a high rate of tumor control for patients with parasellar or sellar meningiomas, and tumor control is accompanied by neurological preservation or improvement in most patients. (http://thejns.org/doi/abs/10.3171/2014.2.jns13139) Key Words radiosurgery Gamma Knife meningioma sellae parasellar oncology stereotactic radiosurgery Abbreviations used in this paper: CN = cranial nerve; GKRS = Gamma Knife radiosurgery; SRS = stereotactic radiosurgery. Parasellar and sellar meningiomas are clinically challenging for neurosurgeons to manage. The difficulty posed by these tumors relates to the potential morbidity associated with tumor exposure and dissection, their frequently firm consistency and vascularity, and their proximity to critical neuroendocrine, vascular, and cranial nerve (CN) structures. Such features usually hinder complete resection, whether by endoscopic or trans cranial skull base approaches. 5,10,41 Open surgical ap- 1268 J Neurosurg / Volume 120 / June 2014

Radiosurgery of parasellar meningiomas proaches have been accompanied by widely varying rates of complete resection and procedure-related morbidity and mortality. 2,6,21,28,40,46,50,58 Initial resection has also been associated with significant risks of delayed recurrence. 21,23,35,43,47,50,53 With earlier detection of sellar and parasellar meningiomas using MRI and/or CT, these tumors are now diagnosed when they have a small to moderate volume. Such tumors often produce minimal symptoms. Stereotactic radiosurgery has been advocated for recurrent sellar and parasellar meningiomas after resection or as an upfront treatment for others. 17,51,52 To better define the outcomes of patients who have undergone stereotactic radiosurgery (SRS) as primary or adjuvant management, we performed a retrospective multiinstitutional analysis of patients who had undergone Gamma Knife radiosurgery (GKRS) for a sellar or parasellar meningioma. Methods Study Design and Patient Population Ten medical centers participating in the North American Gamma Knife Consortium (NAGKC) received individual institutional review board approval to submit their retrospective clinical outcome analysis of patients with sellar and parasellar meningiomas. An Excel spreadsheet database with selected variables was created and sent to all participating centers. The records of patients with meningioma who had undergone GKRS between 1988 and 2011 were assessed by each center for inclusion. Participating centers reviewed the medical records of their patients, entered data in the spreadsheet, and removed all patient identifiers from the data. Under institutional review board approval, an independent third party screened the pooled and de-identified data for errors. Ambiguities were addressed to the respective contributing center. Afterward, data were transmitted to the first author, who along with his coauthors devised this report. Patients were included in the study if they had a histologically diagnosed WHO Grade I meningioma involving the sellar and/or parasellar region. Patients were also included if they had clinical and imaging features consistent with a benign meningioma in this same neuroanatomical region. Such features would include a medical history absent of prior cancer and an intracranial tumor located in the sellar or parasellar region with MRI and/or CT features consistent with a meningioma. Such features included an extraaxial location, contrast enhancement, dural attachment, and, in some patients, tumor calcification. In addition, patients were required to have a minimum of 6 months of imaging and clinical follow-up after GKRS. Radiosurgical Technique The Models U, B, C, 4C, or Perfexion Gamma Knife units were used depending on the technology available to the various participating centers at the time of treatment. The radiosurgical procedure began with the application of the Leksell Model G stereotactic frame (Elekta Inc.) after applying a local anesthetic supplemented by additional sedation as needed. After the frame was placed, J Neurosurg / Volume 120 / June 2014 high-resolution stereotactic MRI was performed. In rare cases in which MRI was not feasible or when MRI distortion was a concern, stereotactic CT was performed. Thinsliced axial and/or coronal plane images were obtained after the administration of intravenous contrast. A neurosurgeon, radiation oncologist, and medical physicist then performed SRS dose planning. At each center, dose selection was based on an empirical algorithm that evaluated tumor volume, proximity to critical structures such as the optic apparatus and pituitary gland, preexisting neurological deficits, and history of treatment with fractionated radiation therapy. Clinical and Imaging Follow-Up Clinical and imaging evaluations were typically performed at follow-up intervals of 6 months for the first 2 years after radiosurgery. In patients demonstrating no evidence of either tumor growth or new neurological findings, follow-up intervals were increased to every 1 2 years. Whenever possible, patients underwent follow-up neurological examination and neuroimaging at the respective treating center. Endocrine testing was performed at the discretion of the treating physician and/or the patient s referring physician. Endocrine testing of patients included tests for cortisol, thyroid, gonadotropic, and growth hormone function. Since most of the participating institutions are referral centers for a broad geographic area, some patients underwent follow-up evaluations by their local physicians. For such patients, clinical notes and neuroimaging studies were received and reviewed by the clinicians who had performed the GKRS. Follow-up images were compared with the images obtained at the time of GKRS. Tumor dimensions were measured in the axial, sagittal, and coronal planes. Tumor volume was estimated by multiplying the left-right (x), anterior-posterior (y), and superior-inferior (z) dimensions and dividing this number by 2. Tumor growth within the planned treatment volume or adjacent to it was considered tumor progression. Statistical Analysis Calculations for a given parameter are based on the total number of patients with complete data for that parameter. Data are presented as the median or mean and range for continuous variables, and as the frequency and percentage for categorical variables. Statistical analyses of categorical variables were performed using the chisquare and Fisher exact test, as appropriate. Statistics of means were performed using the unpaired Student t-test both with and without equal variance (Levene s test), as necessary, and the Wilcoxon rank-sum tests when variables were not normally distributed. Kaplan-Meier plots for tumor progression free survival using the dates of the first SRS, follow-up MRI studies, and death or last follow-up. Progression-free survival and overall survival time were calculated from the day of the first SRS by using the Kaplan-Meier method. Univariate analysis was performed on the Kaplan-Meier curves using log-rank statistics. Factors predictive of tumor progression (p < 0.15) were entered into a Cox regression analysis to assess hazard ratios. 1 Patients with an unfavorable outcome were 1269

J. P. Sheehan et al. defined as those with worsening or new deficits following SRS or with tumor progression. Clinical covariates predicting an unfavorable outcome with a univariate p < 0.15 were included in multivariate logistic regression analysis. Clinically significant variables and interaction expansion covariates were further assessed in both Cox and logistic multivariate analysis as deemed relevant. A p 0.05 was considered statistically significant. Commercially available statistical processing software (Stata, version 11.0, Stata Corp.) was used. Results Patient Attributes The authors identified 763 patients with sellar and/or parasellar meningiomas treated with GKRS. The following centers contributed data: the University of Pittsburgh (281 patients), Yale University (11 patients), Cleveland Clinic (29 patients), University of Manitoba (27 patients), University of Sherbrooke (69 patients), Barrow Neurological Institute (158 patients), University of Pennsylvania (29 patients), Washington Hospital Center (2 patients), New York University (19 patients), and the University of Virginia (138 patients). The median patient age was 56 years (range 8 90 years). There was a clear sex predilection with 567 females (74.3%) and 196 males (25.7%). Three hundred fifty-five patients (50.7%) had undergone resection of a histologically confirmed WHO Grade I meningioma before GKRS; 55 patients (10.6%) had undergone more than one resection. The remaining patients displayed neuroimaging and clinical features consistent with a benign meningioma. Fractionated radiation therapy had been performed before GKRS in 3.9% of patients. Preoperative patient characteristics and presentations are detailed in Table 1. Tumor Attributes and Response One hundred fifty-six patients (32.4%) underwent treatment before 2000, whereas 325 patients (67.6%) received treatment after 2000; treatment dates were unknown in the remaining 282 patients. The mean tumor volume was 8.8 cm 3 (range 0.05 54.8 cm 3 ). The median prescription dose delivered to the tumor margin was 13 Gy (range 5 30 Gy). The maximum dose varied from 10 to 60 Gy (median 28 Gy; Table 2). Many centers limited the maximum dose to the optic nerve and chiasm to between 8 and 12 Gy. The median follow-up after SRS was 66.7 months (range 6 216 months). At the last follow-up, 74 (9.8%) of 758 patients with available imaging demonstrated tumor progression, 334 (44.1%) had a decrease in tumor volume, and 350 (46.2%) had unchanged tumor volumes (Table 3), which makes for a 90.2% overall tumor control rate. Actuarial tumor progression free survival rates at 3, 5, 8, and 10 years following SRS were 98%, 95%, 88%, and 82%, respectively (Fig. 1). Further surgery as planned treatment, shunting for hydrocephalus, or treatment due to tumor progression was performed in 23 (3.3%) of 690 patients. Additionally, 22 (3.2%) of 687 patients subsequently underwent fractionated radiation therapy because of tumor progression. TABLE 1: Attributes of patients with a sellar and/or parasellar meningioma before GKRS* Variable No. (%) total patients 763 females 567 (74.3) age in yrs mean 55.8 median 56 range 8 90 patients w/ resection pre-gkrs 355 (50.7) of 700 patients w/ >1 resection pre-gkrs 74 (10.6) of 700 patients w/ radiation therapy pre-gkrs 23 (3.8) of 598 patients w/ symptom other than headache 584 (78.7) of 742 patients w/ CN deficits prior to GKRS per no. CN deficits 0 162 (23.5) of 688 1 280 (40.7) of 688 >1 246 (35.8) of 688 patients w/ specific CN deficit prior to GKRS II 99 (14.5) of 684 III 167 (24.4) of 684 IV 56 (8.2) of 684 V 245 (35.8) of 684 VI 225 (32.9) of 684 time from diagnosis to GKRS in yrs mean 2.1 median 1 range 0 30 * Total number reflects the number of patients for whom evaluable data were available. There was no difference in the tumor progression free survival between patients with no prior resection (upfront radiosurgery) and those who had one prior surgical procedure (p = 0.996; Fig. 2). Patients who underwent more than 1 prior surgery were significantly more likely to have tumor progression despite radiosurgery (p = 0.004). Patients who underwent prior fractionated radiation therapy were 7.7 times more likely to have tumor progression (p < 0.001; Fig. 3). The pre-srs factors predictive of tumor progression in univariate analysis are shown in Table 4. TABLE 2: Gamma Knife radiosurgery parameters, tumor volume, and duration of follow-up* Characteristic Mean Median Range SD tumor margin dose (Gy) 13.2 13 5 30 2.4 max dose (Gy) 27.7 28 10 60 5.9 tumor vol (cm 3 ) 8.8 6.7 0.05 54.8 7.3 follow-up after GKRS (mos) 66.7 55 6 216 45.8 * SD = standard deviation. 1270 J Neurosurg / Volume 120 / June 2014

Radiosurgery of parasellar meningiomas TABLE 3: Clinical and radiographic outcomes of GKRS Outcome (no. evaluable patients) Effect of Tumor Margin Dose The median tumor margin dose was 13 Gy (range 5 30 Gy) to a median treatment volume of 6.7 cm 3 (range 0.05 54.8; Table 2). A decreasing tumor margin dose (HR = 1.09, 95% CI 1.01 1.18, p = 0.017) and decreasing maximum dose (HR = 1.03, 95% CI 1.00 1.17, p = 0.087) were predictive of tumor progression on univariate analysis. Patients receiving a margin dose 13 Gy were 1.6 times less likely to have tumor progression (95% CI 1.00 2.63, p = 0.50; Fig. 4). J Neurosurg / Volume 120 / June 2014 % Evaluable Patients neurological condition (737) improved 31.10 worse 13.80 no change 55.10 improvement in preexisting CN deficit (737)* 34 II 5.50 III 5.50 IV 1.10 V 22.30 VI 10.06 worse or new CN deficit (737) 9.60 II 3.00 III 2.20 IV 0.40 V 4.70 VI 1.10 any new or worsened hypopituitarism (719) 1.80 cortisol 0.30 thyroid hormone 0.70 gonadotropin 0.40 growth hormone 0.10 diabetes insipidus 0.40 panhypopituitarism 0.10 tumor size (758) increase 9.8 decrease 44.1 no change 46.2 further surgery (690) 3.30 further radiation therapy (687) 3.20 overall outcome (732) favorable 79.60 unfavorable 20.40 * Some patients demonstrated improvement of more than one CN. Seventy patients had 96 deficits, and some patients had more than one CN deficit. Twelve patients had 15 cases of new or worsening pituitary function, and 1 patient had panhypopituitarism. Actuarial tumor control was 99%, 98%, 95%, 88%, 82%, and 74% at 2, 3, 5, 8, 10, and 12 years, respectively. Favorable outcome = no new or worsening neurological function and tumor control. Fig. 1. Tumor-free progression after GKRS. Censored data are indicated by the circles. Integer values below the x-axis reflect the number of patients at each interval of the Kaplan-Meier analysis. On multivariate analysis, both a history of more than one surgery a history of fractionated radiation therapy were significant predictors of tumor progression (Table 4). These remained significant predictors of tumor progression even after controlling for increasing tumor margin dose (HR = 0.88, 95% CI 0.80 0.97, p = 0.009). Clinical Response At the last clinical follow-up, overall neurological function worsened in 102 (13.8%) of 737 patients, improved in 229 (31.1%), and remained stable in 406 (55.1%; Table 3). New or worsening CN deficits were observed in 9.6% of patients. Cranial nerves II (3%) and V (4.7%) were the nerves that most often demonstrated new or worsening dysfunction following GKRS. Additional CN deficits following GKRS are detailed in Table 3. Among the patients with CN deficits, 25 (24.5%) had tumor progression, and 77 (75.5%) demonstrated no change or a decrease in tumor volume. In those patients with CN II dysfunction and no evidence of tumor growth, the maximum dose to the optic apparatus was possible in 45% of cases, and in those cases, the maximum dose to the optic apparatus ranged from 0.75 to 8 Gy. One hundred seventy-nine (34%) of 529 patients with prior CN deficits showed improvement in at least one deficit following GKRS. Some patients demonstrated improvement in more than one CN following radiosurgery (Table 3). Among those with improvement, the distribution by CN was as follows: II, 12.5%; III, 12.5%; IV, 2.5%; V, 50%; and VI, 22.5%. 1271

J. P. Sheehan et al. Fig. 2. Tumor-free progression in patients with and without a history of surgery prior to GKRS. Censored data are indicated by circles and triangles. Integer values below the x-axis reflect the number of patients at each interval of the Kaplan-Meier analysis. Endocrine Response Endocrine function after GKRS was recorded in 719 patients. After GKRS, new or worsened endocrine deficiencies were demonstrated in 1.8% of patients at the last follow-up (Table 3). Thyroid hormone dysfunction was the most frequent endocrinopathy post-gkrs. Panhypopituitarism was rare and occurred in only one patient who also demonstrated tumor progression on follow-up imaging. Other Serious Complications In the current series, no evidence of radiosurgeryrelated neoplasia, symptomatic carotid artery occlusion, or treatment-related stroke was observed. Overall Outcome An overall favorable outcome was demonstrated in 583 patients (79.6%) who had no tumor progression, no worsening of a preexisting CN deficit, and no new deficit. An unfavorable outcome occurred in 149 (20.4%) of 732 patients. Preradiosurgical patient and tumor characteristics predictive of an unfavorable outcome on univariate analysis are demonstrated in Table 5. There was no difference in overall outcome between patients who underwent resection prior to radiosurgery and those who underwent primary radiosurgery (OR = 1.24, 95% CI 0.86 1.80, p = Fig. 3. Tumor-free progression in patients with and without prior radiation therapy. Censored data are indicated by the circles and triangles. Integer values below the x-axis reflect the number of patients at each interval of the Kaplan-Meier analysis. RT = radiation therapy. 0.240). However, those with more than one surgery were 1.5 times more likely to have an unfavorable outcome (95% CI 1.11 1.89, p = 0.007). Increasing tumor volume was associated with an unfavorable outcome (1.03, 95% CI 1.01 1.06, p < 0.008); an unfavorable outcome occurred in 16.9% of patients with a tumor volume < 5 cm 3, 20.2% of patients with a tumor volume between 5 and 15 cm 3, and 24.2% of patients with a tumor volume > 15 cm 3. Analysis of SRS-associated variables demonstrated that a decreasing tumor margin dose (OR = 1.10, 95% CI 1.02 1.19, p = 0.016) and maximal dose (1.04, 95% CI 1.01 1.08, p = 0.019) were associated with unfavorable outcomes on univariate analysis. Unfavorable outcomes occurred in 22.8% of patients receiving a peripheral dose < 14 Gy, 16.9% of patients receiving 14 18 Gy, and 17.7% of patients receiving more than 18 Gy. On multivariate analysis, patients with a history of more than one surgery (OR = 1.37, 95% CI 1.05 1.80, p = 0.021) and a larger tumor volume (OR = 1.02, 95% CI 1.00 1.05, p = 0.022) were more likely to have unfavorable outcomes. When adding SRS-associated variables to the multivariate models, a history of more than 1 surgery was the most significant predictor of an unfavorable outcome (OR = 1.4, 95% CI 1.07 1.184, p = 0.016). We noted a trend toward an unfavorable outcome with increasing tumor volume (OR = 1.03, 95% CI 1.00 1.05, p = 0.076) and decreasing tumor margin dose (OR = 1.08, 95% CI 1.00 1.17, p = 0.069). 1272 J Neurosurg / Volume 120 / June 2014

Radiosurgery of parasellar meningiomas TABLE 4: Preradiosurgical factors predictive of tumor progression* Covariate Univariate Analysis Multivariate Analysis HR (95% CI) p Value HR (95% CI) p Value history of radiation therapy 7.71 (0.3.61 16.47) <0.001 4.89 (0.2.09 11.41) <0.001 increasing volume 1.02 (1.00 1.05) 0.103 history of > 1 surgery 1.65 (1.18 2.33) 0.004 1.53 (1.03 2.283) 0.032 * Factors predictive of tumor recurrence have p < 0.15. Discussion Natural History of Sellar and Parasellar Meningiomas The natural history of sellar and parasellar meningiomas is not well defined. In a recent systematic review of 22 studies covering the natural history of meningiomas, cavernous sinus lesions produced new or progressive symptoms 61% of the time over a median follow-up of 4.6 years. 55 Additionally, the authors found specific patterns of tumor growth and size that were more likely to generate symptoms. For example, lesions 2 2.5 cm in size and growing more than 10% per year had a 42.3% rate of symptom development. Compare this with lesions smaller than 2 cm with less than 10% annual growth having a 0% rate of symptom development. In contrast, Bindal et al. described the natural history of 40 untreated meningiomas located in the cavernous sinus, petroclival region, or anterior clinoidal region. 3 Fig. 4. Tumor-free progression in patients as a function of radiosurgical dose to the tumor margin. Censored data are indicated by circles. Integer values below the x-axis reflect the number of patients at each interval of the Kaplan-Meier analysis. J Neurosurg / Volume 120 / June 2014 Clinically, 28 patients (70%) presented with a CN deficit and 11 patients (27.5%) were asymptomatic. Over a mean clinical follow-up of 83 months, 11 patients (27.5%) experienced new or worsening CN deficits. During a mean radiographic follow-up of 76 months, 7 patients (17.5%) demonstrated tumor growth. The authors noted that there was a poor correlation between clinical and radiographic changes. In particular, 54.5% of patients with neurological progression did not have any changes on neuroimaging, and 27.2% had simultaneous neurological and radiographic progression. 3 Resection and Radiation Therapy Patients with WHO Grade I meningiomas of the parasellar and sellar region are typically expected to live for a long period and to do so with a reasonably high neurological performance status. Surgical removal of meningiomas can be curative but difficult to achieve in this region. Resection via transcranial or transsphenoidal routes has historically been the primary treatment option for patients with sellar and parasellar meningiomas. However, a neurosurgeon s enthusiasm for achieving complete resection must be tempered by the desire to preserve or improve a patient s neurological function. During resection, damage to venous channels or the cavernous segment of the internal carotid artery and traversing the CNs can have serious long-term ramifications for the patient. Complication rates for gross-total resection of meningiomas in the parasellar region vary widely, and adverse events have been reported to occur in between 0% and 60% of patients. 8,36,39,54 Radiation therapy has also been used in selected patients with recurrent meningiomas or in those with significant medical comorbidities in whom resection was considered excessively risky. Control rates for skull base meningiomas after conventional radiation therapy delivering 50 55 Gy in 30 33 fractions vary from 75% to 95% at both 5- and 10-year intervals. 7,11,17,32,38 Neurological deficits associated with skull base meningiomas have been noted to improve or remain stable in 69% 100% of patients who undergo conventional fractionated radiation therapy. 11,31,32,38 Complication rates associated with radiation therapy range from 0% to 24%. 34 Injury to adjacent CNs is one of the most commonly reported complications. The risk of radiation-associated injury to the optic apparatus and other CNs varies between 0% and 3%. 34 Recent advances in the use of both photon and proton radiation techniques, intensity-modulated radiation therapy (IMRT), or image-guided radiation therapy (IGRT) have demonstrated no substantive improvement in tumor con- 1273

J. P. Sheehan et al. TABLE 5: Preradiosurgical factors predictive of an unfavorable outcome* Factor Univariate Analysis Multivariate Analysis OR 95% CI p Value OR 95% CI p Value history of > 1 surgery 1.45 1.11 1.89 0.007 1.37 1.05 1.80 0.021 history of prior radiation therapy 2.46 1.04 12.35 <0.040 increasing tumor vol 1.03 1.01 1.06 <0.008 1.02 1.00 1.05 0.022 increasing follow-up 1.01 1.00 1.01 0.111 * Favorable outcome = no new or worsening neurological function and tumor control. Factors predictive of tumor recurrence have p < 0.15. trol or any reduction in complications in the management of meningiomas. 29,33,34,57,59,60 Stereotactic Radiosurgery With the addition of SRS to the neurosurgical and radiation oncology armamentarium, the pendulum has swung away from both aggressive resection and fractionated radiation therapy for sellar and parasellar meningiomas. Instead, management options have more commonly switched to early radiosurgery for smaller-volume tumors or subtotal resection followed by radiosurgery for larger-volume tumors with symptomatic mass effect. Both strategies place a premium on the preservation of neurological function. In a recent meta-analysis of patients undergoing treatment for cavernous sinus meningioma, SRS together with resection led to improved rates of tumor control compared with surgery alone, and this benefit was afforded regardless of the extent of tumor resection. 54 Moreover, the rate of cranial neuropathy in patients undergoing resection was significantly higher than that in patients undergoing radiosurgery alone. The widespread availability of CT and MRI has led to the detection of parasellar and sellar meningiomas in patients who demonstrate few or no symptoms. Stereotactic radiosurgery has become even more appealing as an upfront treatment in patients unwilling to accept the risks of resection. Selecting SRS as the initial method of management is based on clinical and imaging evidence that the lesion is in fact a meningioma. Magnetic resonance imaging and CT cannot provide confirmation of a histologically benign meningioma. However, in a cohort of 219 patients with image-diagnosed meningiomas, a detailed history, clinical examination, and careful review of the neuroimaging studies yielded a 10-year actuarial misdiagnosis rate of 2.3% ± 1.4%. 14 The actuarial tumor control rate in this same study was noted to be 93.2% ± 2.7% at 10 years postradiosurgery. In the current study, there was no statistical difference in tumor control or neurological outcome in patients with histological confirmation of a WHO Grade I meningioma as compared with that in patients receiving a diagnosis via neuroimaging and clinical features alone. Our study indicated that an appropriate clinical history coupled with modern neuroimaging studies yields an accurate rate of diagnosis for parasellar and sellar meningiomas. Large series and long-term results for radiosurgical treatment of meningiomas are available. In a series of 972 patients with meningiomas in diverse locations and treated with GKRS at the University of Pittsburgh, the reported actuarial tumor control rates were 93% at 5 years and 87% at both 10 and 15 years. 25 Patients in this series received a median tumor margin dose of 13 Gy. These results confirmed an earlier study of 159 patients from the same institution in which tumor volumes decreased in 3%, remained stable in 60%, and eventually increased in 6% of patients. 27 In a review, Minniti et al. compiled 18 studies with 2919 skull base meningiomas treated with GKRS. 34 The 5-year actuarial control rate was 91%. Seven of the studies (1626 skull base meningiomas) reported an averaged 10-year actuarial control rate of 87.6%. In summarizing all recent large series of more than 100 patients, we found that 5- and 10-year local control rates ranged from 86.2% to 98.5% and 73% to 97%, respectively. 9,19,22,24,26,27,37,51,52,56 In the current study, we observed an overall tumor control rate of 90.2% and 5- and 10-year actuarial control rates of 95% and 82%, respectively. The available literature provides less clarity regarding long-term clinical outcomes after radiosurgery for sellar and parasellar meningiomas. Neurological improvement rates have varied between 8% and 66% in published series. 12,13,19,20,22,24 27,37,51,56 Reported postradiosurgical complication rates have varied between 3% and 40%. Most reports have indicated an average of 8% of patients who experience neurological complications (transient complications in 3% and permanent in 5%). 49 In a series of 159 patients with cavernous sinus meningiomas treated using radiosurgery and with a mean follow-up of 35 months (range 2 138 months), neurological status improved in 29%, remained stable in 62%, and worsened in 9%. 27 Among those with neurological decline, 1.9% had visual deterioration, 3.1% had new trigeminal nerve dysfunction, 1.3% had oculomotor neuropathy, 0.62% had trochlear neuropathy, 1.3% experienced partial complex seizures, and 0.62% had cognitive deterioration. 27 Long-term complications include CN dysfunction and neurological deficits associated with adverse radiation effects, such as edema and necrosis. Among published series of 100 or more patients, rates of significant long-term complications ranged from 0% to 16% in centers utilizing median doses of 12 15 Gy. 9,12,13,19,20,24 27,37,56 An analysis by a group from the University of California, San Francisco, revealed a pooled cranial neuropathy rate of 59.6% (95% CI 50.3 67.5%) for patients undergoing resection, as compared with a rate of 25.7% (95% CI 11.5 38.9%) for those undergoing radiosurgery alone (p < 0.05). 54 In the current study, we observed a 9.6% risk of any new or wors- 1274 J Neurosurg / Volume 120 / June 2014

Radiosurgery of parasellar meningiomas ening cranial neuropathy after GKRS and a 34% chance of improvement in preexisting CN dysfunction following GKRS. Cranial nerves II and V were most likely to demonstrate new or progressive dysfunction following GKRS. Note that we observed radiosurgically induced CN II dysfunction in some patients who had received a maximum dose of 8 Gy or less to the optic apparatus. Thus, despite the 8-Gy dogma, some patients remain susceptible to visual decline even with low doses delivered to the optic apparatus. Overall, our neurological results compare very favorably with those for other treatment options (for example, resection and radiation therapy) as well as for the natural history of a sellar or parasellar meningioma. Overall, a favorable outcome in the current study was defined as tumor regression or stability accompanied by preserved or improved neurological function. A favorable outcome was most commonly seen in patients with a smaller tumor volume and no or only one prior resection. As with many neurosurgical procedures, the failure of one prior procedure often predicts a less successful subsequent intervention. Since radiosurgery is an imageguided surgery, multiple unsuccessful prior resections can make defining the radiosurgical target as well as delineating the critical structures more difficult. In patients with symptomatic larger-volume tumors that produce regional mass effect, we favor initial surgery to reduce the tumor volume followed by early radiosurgery. We did not observe a difference in overall outcomes between patients with upfront radiosurgery and those with 1 prior surgery. Rare Complications After Radiosurgery Delayed endocrinopathy after GKRS was rare in the current study, although thyroid deficiency was the most common. Periodic evaluation of thyroid function seems prudent in patients who demonstrate symptoms of hypothyroidism, and hormonal replacement for endocrinopathy is typically advisable. The actual incidence of delayed endocrine dysfunction in the present series may be underestimated since testing generally occurred at the discretion of the referring physician. The delayed development of radiation-related tumors or stroke was not observed in the current study. Radiosurgery-related tumor development has a reported risk between 0 and 3 cases per 200,000 patients, a risk that does not exceed the incidence of cancer development in the general population. 15,44 Several reports on the development of glioblastomas after radiosurgery have been published. 16,30,45,48,61 Such case reports cannot define the actual risk, but the selection of a treatment recommendation must weigh many factors. Even low doses of radiation (for example, 50 75 mgy from CT scans) or frequent dental radiography has been linked to an increased risk of brain tumor formation. 18,42 Brada et al. reported that the 5- and 10-year actuarial incidence of stroke after radiation therapy ranges from 4% to 11% of patients. 4 In the current study, we did not observe a single case of treatment-related ischemic cerebrovascular disease. The low risk of either radiation-related neoplasms or stroke provides further evidence of the favorable risk/benefit ratio of SRS as a treatment strategy for parasellar and sellar meningiomas. J Neurosurg / Volume 120 / June 2014 Study Limitations and Biases As the data presented are the result of a retrospective multicenter experience, limitations include patient selection bias and treatment bias at each center. Data were not contributed equally by all centers, and thus biases introduced from larger contributing centers may outweigh those of smaller ones. The parasellar location is sizeable, and the location of meningiomas within this space could include tumors with different natural histories and responses to radiosurgery. Selection bias may have affected the use of upfront radiosurgery, resection, radiation therapy, and salvage radiosurgery at each center. Independent auditing of the data was not performed. Data auditing could be added to future studies and prospective radiosurgical registries. Follow-up patterns varied somewhat from center to center, and this could have contributed to some bias in the study, particularly regarding endocrine deficiencies. In addition, the retrospective nature of the study and the lack of an untreated control group restrict our ability to assess the full benefits of and the complications arising from radiosurgery. Finally, it is not clear to what extent our results are generalizable to other centers. All of the contributing centers used a consistent radiosurgical platform (that is, the Gamma Knife) and dose planning software. Moreover, many of the clinicians at the contributing centers have had a decade or more of experience in radiosurgery, and the centers have a reputation for being high-volume centers with an extensive commitment to outcomes research. The clinicians experience and the radiosurgical techniques at the centers have been refined in a way that may have favorably affected patient selection and outcomes. Conclusions Stereotactic radiosurgery affords a high rate of tumor control for sellar and parasellar meningiomas. After radiosurgery, neurological preservation or improvement was observed in the majority of patients. Favorable outcomes of tumor control and neurological preservation were more likely in patients without multiple prior surgeries, those with smaller tumors, and those in whom a higher tumor margin dose was delivered. Acknowledgments Dr. Kyun-Jae Park and Aditya Iyer, M.Eng., assisted with the project at the University of Pittsburgh. Disclosure The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Drs. Lunsford, Petti, and Kondziolka are consultants for Elekta AB. Dr. Lunsford is also a shareholder in Elekta. Author contributions to the study and manuscript preparation include the following. Conception and design: Sheehan, Kondziolka, Lunsford. Acquisition of data: Sheehan, Kano, Kaufmann, Mathieu, Zeiler, West, Chao, Varma, Chiang, Yu, McBride, Nakaji, Youssef, Honea, Rush, Kondziolka, Lee, Bailey, Kunwar, Petti. Analysis and interpretation of data: Sheehan, Starke. Drafting the article: Sheehan. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Sheehan. 1275

J. P. Sheehan et al. References 1. Altman DG: Practical Statistics for Medical Research. Boca Raton, FL: Chapman & Hall/CRC, 1990 2. Bassiouni H, Asgari S, Stolke D: Tuberculum sellae meningiomas: functional outcome in a consecutive series treated microsurgically. Surg Neurol 66:37 45, 2006 3. Bindal R, Goodman JM, Kawasaki A, Purvin V, Kuzma B: The natural history of untreated skull base meningiomas. Surg Neurol 59:87 92, 2003 4. Brada M, Burchell L, Ashley S, Traish D: The incidence of cerebrovascular accidents in patients with pituitary adenoma. Int J Radiat Oncol Biol Phys 45:693 698, 1999 5. Cavallo LM, Cappabianca P, Galzio R, Iaconetta G, de Divitiis E, Tschabitscher M: Endoscopic transnasal approach to the cavernous sinus versus transcranial route: anatomic study. Neurosurgery 56 (2 Suppl):379 389, 2005 6. Chi JH, McDermott MW: Tuberculum sellae meningiomas. Neurosurg Focus 14(6):E6, 2003 7. Condra KS, Buatti JM, Mendenhall WM, Friedman WA, Marcus RB Jr, Rhoton AL: Benign meningiomas: primary treatment selection affects survival. Int J Radiat Oncol Biol Phys 39:427 436, 1997 8. DeMonte F, Smith HK, al-mefty O: Outcome of aggressive removal of cavernous sinus meningiomas. J Neurosurg 81: 245 251, 1994 9. DiBiase SJ, Kwok Y, Yovino S, Arena C, Naqvi S, Temple R, et al: Factors predicting local tumor control after gamma knife stereotactic radiosurgery for benign intracranial meningiomas. Int J Radiat Oncol Biol Phys 60:1515 1519, 2004 10. Dolenc VV: Transcranial epidural approach to pituitary tumors extending beyond the sella. Neurosurgery 41:542 552, 1997 11. Dufour H, Muracciole X, Métellus P, Régis J, Chinot O, Grisoli F: Long-term tumor control and functional outcome in patients with cavernous sinus meningiomas treated by radiotherapy with or without previous surgery: is there an alternative to aggressive tumor removal? Neurosurgery 48:285 296, 2001 12. Eustacchio S, Trummer M, Fuchs I, Schröttner O, Sutter B, Pendl G: Preservation of cranial nerve function following Gamma Knife radiosurgery for benign skull base meningiomas: experience in 121 patients with follow-up of 5 to 9.8 years. Acta Neurochir Suppl 84:71 76, 2002 13. Feigl GC, Samii M, Horstmann GA: Volumetric follow-up of meningiomas: a quantitative method to evaluate treatment outcome of gamma knife radiosurgery. Neurosurgery 61: 281 287, 2007 14. Flickinger JC, Kondziolka D, Maitz AH, Lunsford LD: Gamma knife radiosurgery of imaging-diagnosed intracranial meningioma. Int J Radiat Oncol Biol Phys 56:801 806, 2003 15. Ganz JC: Gamma knife radiosurgery and its possible relationship to malignancy: a review. J Neurosurg 97 (5 Suppl): 644 652, 2002 16. Ganz JC, Reda WA, Abdelkarim K: Gamma Knife surgery of large meningiomas: early response to treatment. Acta Neurochir (Wien) 151:1 8, 2009 17. Goldsmith BJ, Wara WM, Wilson CB, Larson DA: Postoperative irradiation for subtotally resected meningiomas. A retrospective analysis of 140 patients treated from 1967 to 1990. J Neurosurg 80:195 201, 1994 (Erratum in J Neurosurg 80:777, 1994) 18. Han YY, Berkowitz O, Talbott E, Kondziolka D, Donovan M, Lunsford LD: Are frequent dental x-ray examinations associated with increased risk of vestibular schwannoma? Clinical article. J Neurosurg 117 Suppl:78 83, 2012 19. Hasegawa T, Kida Y, Yoshimoto M, Koike J, Iizuka H, Ishii D: Long-term outcomes of Gamma Knife surgery for cavernous sinus meningioma. J Neurosurg 107:745 751, 2007 20. Hudgins WR, Barker JL, Schwartz DE, Nichols TD: Gamma Knife treatment of 100 consecutive meningiomas. Stereotact Funct Neurosurg 66 (Suppl 1):121 128, 1996 21. Ichinose T, Goto T, Ishibashi K, Takami T, Ohata K: The role of radical microsurgical resection in multimodal treatment for skull base meningioma. Clinical article. J Neurosurg 113: 1072 1078, 2010 (Erratum in J Neurosurg 113:1123, 2010) 22. Iwai Y, Yamanaka K, Ikeda H: Gamma Knife radiosurgery for skull base meningioma: long-term results of low-dose treatment. Clinical article. J Neurosurg 109:804 810, 2008 23. Jääskeläinen J: Seemingly complete removal of histologically benign intracranial meningioma: late recurrence rate and factors predicting recurrence in 657 patients. A multivariate analysis. Surg Neurol 26:461 469, 1986 24. Kollová A, Liscák R, Novotný J Jr, Vladyka V, Simonová G, Janousková L: Gamma Knife surgery for benign meningioma. J Neurosurg 107:325 336, 2007 25. Kondziolka D, Mathieu D, Lunsford LD, Martin JJ, Madhok R, Niranjan A, et al: Radiosurgery as definitive management of intracranial meningiomas. Neurosurgery 62:53 60, 2008 26. Kreil W, Luggin J, Fuchs I, Weigl V, Eustacchio S, Papaefthymiou G: Long term experience of gamma knife radiosurgery for benign skull base meningiomas. J Neurol Neurosurg Psychiatry 76:1425 1430, 2005 27. Lee JY, Niranjan A, McInerney J, Kondziolka D, Flickinger JC, Lunsford LD: Stereotactic radiosurgery providing longterm tumor control of cavernous sinus meningiomas. J Neurosurg 97:65 72, 2002 28. Linskey ME, Davis SA, Ratanatharathorn V: Relative roles of microsurgery and stereotactic radiosurgery for the treatment of patients with cranial meningiomas: a single-surgeon 4-year integrated experience with both modalities. J Neurosurg 102 Suppl:59 70, 2005 29. Lo SS, Cho KH, Hall WA, Kossow RJ, Hernandez WL, Mc- Collow KK, et al: Single dose versus fractionated stereotactic radiotherapy for meningiomas. Can J Neurol Sci 29:240 248, 2002 30. Loeffler JS, Niemierko A, Chapman PH: Second tumors after radiosurgery: tip of the iceberg or a bump in the road? Neurosurgery 52:1436 1442, 2003 31. Maire JP, Caudry M, Guérin J, Célérier D, San Galli F, Causse N, et al: Fractionated radiation therapy in the treatment of intracranial meningiomas: local control, functional efficacy, and tolerance in 91 patients. Int J Radiat Oncol Biol Phys 33:315 321, 1995 32. Mendenhall WM, Morris CG, Amdur RJ, Foote KD, Friedman WA: Radiotherapy alone or after subtotal resection for benign skull base meningiomas. Cancer 98:1473 1482, 2003 33. Metellus P, Regis J, Muracciole X, Fuentes S, Dufour H, Nanni I, et al: Evaluation of fractionated radiotherapy and gamma knife radiosurgery in cavernous sinus meningiomas: treatment strategy. Neurosurgery 57:873 886, 2005 34. Minniti G, Amichetti M, Enrici RM: Radiotherapy and radiosurgery for benign skull base meningiomas. Radiat Oncol 4:42, 2009 35. Natarajan SK, Sekhar LN, Schessel D, Morita A: Petroclival meningiomas: multimodality treatment and outcomes at longterm follow-up. Neurosurgery 60:965 981, 2007 36. Newman S: A prospective study of cavernous sinus surgery for meningiomas and resultant common ophthalmic complications (an American Ophthalmological Society thesis). Trans Am Ophthalmol Soc 105:392 447, 2007 37. Nicolato A, Foroni R, Alessandrini F, Maluta S, Bricolo A, Gerosa M: The role of Gamma Knife radiosurgery in the management of cavernous sinus meningiomas. Int J Radiat Oncol Biol Phys 53:992 1000, 2002 38. Nutting C, Brada M, Brazil L, Sibtain A, Saran F, Westbury C, et al: Radiotherapy in the treatment of benign meningioma of the skull base. J Neurosurg 90:823 827, 1999 39. O Sullivan MG, van Loveren HR, Tew JM Jr: The surgical 1276 J Neurosurg / Volume 120 / June 2014

Radiosurgery of parasellar meningiomas resectability of meningiomas of the cavernous sinus. Neurosurgery 40:238 247, 1997 40. Otani N, Muroi C, Yano H, Khan N, Pangalu A, Yonekawa Y: Surgical management of tuberculum sellae meningioma: role of selective extradural anterior clinoidectomy. Br J Neurosurg 20:129 138, 2006 41. Parkinson D: Lateral sellar compartment: history and anatomy. J Craniofac Surg 6:55 68, 1995 42. Pearce MS, Salotti JA, Little MP, McHugh K, Lee C, Kim KP, et al: Radiation exposure from CT scans in childhood and subsequent risk of leukaemia and brain tumours: a retrospective cohort study. Lancet 380:499 505, 2012 43. Roberti F, Sekhar LN, Kalavakonda C, Wright DC: Posterior fossa meningiomas: surgical experience in 161 cases. Surg Neurol 56:8 21, 2001 44. Rowe J, Grainger A, Walton L, Silcocks P, Radatz M, Kemeny A: Risk of malignancy after gamma knife stereotactic radiosurgery. Neurosurgery 60:60 66, 2007 45. Salvati M, Frati A, Russo N, Caroli E, Polli FM, Minniti G, et al: Radiation-induced gliomas: report of 10 cases and review of the literature. Surg Neurol 60:60 67, 2003 46. Sanna M, Bacciu A, Pasanisi E, Taibah A, Piazza P: Posterior petrous face meningiomas: an algorithm for surgical management. Otol Neurotol 28:942 950, 2007 47. Sekhar LN, Swamy NK, Jaiswal V, Rubinstein E, Hirsch WE Jr, Wright DC: Surgical excision of meningiomas involving the clivus: preoperative and intraoperative features as predictors of postoperative functional deterioration. J Neurosurg 81:860 868, 1994 48. Shamisa A, Bance M, Nag S, Tator C, Wong S, Norén G, et al: Glioblastoma multiforme occurring in a patient treated with gamma knife surgery. Case report and review of the literature. J Neurosurg 94:816 821, 2001 49. Sheehan JP, Williams BJ, Yen CP: Stereotactic radiosurgery for WHO grade I meningiomas. J Neurooncol 99:407 416, 2010 50. Stafford SL, Perry A, Suman VJ, Meyer FB, Scheithauer BW, Lohse CM, et al: Primarily resected meningiomas: outcome and prognostic factors in 581 Mayo Clinic patients, 1978 through 1988. Mayo Clin Proc 73:936 942, 1998 51. Stafford SL, Pollock BE, Foote RL, Link MJ, Gorman DA, Schomberg PJ, et al: Meningioma radiosurgery: tumor control, outcomes, and complications among 190 consecutive patients. Neurosurgery 49:1029 1038, 2001 52. Starke RM, Williams BJ, Hiles C, Nguyen JH, Elsharkawy MY, Sheehan JP: Gamma Knife surgery for skull base meningiomas. Clinical article. J Neurosurg 116:588 597, 2012 53. Sughrue ME, Kane AJ, Shangari G, Rutkowski MJ, McDermott MW, Berger MS, et al: The relevance of Simpson Grade I and II resection in modern neurosurgical treatment of World Health Organization Grade I meningiomas. Clinical article. J Neurosurg 113:1029 1035, 2010 54. Sughrue ME, Rutkowski MJ, Aranda D, Barani IJ, McDermott MW, Parsa AT: Factors affecting outcome following treatment of patients with cavernous sinus meningiomas. Clin ical article. J Neurosurg 113:1087 1092, 2010 55. Sughrue ME, Rutkowski MJ, Aranda D, Barani IJ, McDermott MW, Parsa AT: Treatment decision making based on the published natural history and growth rate of small meningiomas. A review and meta-analysis. J Neurosurg 113:1036 1042, 2010 56. Takanashi M, Fukuoka S, Hojyo A, Sasaki T, Nakagawara J, Nakamura H: Gamma knife radiosurgery for skull-base meningiomas. Prog Neurol Surg 22:96 111, 2009 57. Torres RC, Frighetto L, De Salles AA, Goss B, Medin P, Solberg T, et al: Radiosurgery and stereotactic radiotherapy for intracranial meningiomas. Neurosurg Focus 14(5):E5, 2003 58. Voss NF, Vrionis FD, Heilman CB, Robertson JH: Meningiomas of the cerebellopontine angle. Surg Neurol 53:439 447, 2000 59. Weber DC, Lomax AJ, Rutz HP, Stadelmann O, Egger E, Timmermann B, et al: Spot-scanning proton radiation therapy for recurrent, residual or untreated intracranial meningiomas. Radiother Oncol 71:251 258, 2004 60. Weber DC, Schneider R, Goitein G, Koch T, Ares C, Geismar JH, et al: Spot scanning-based proton therapy for intracranial meningioma: long-term results from the Paul Scherrer Institute. Int J Radiat Oncol Biol Phys 83:865 871, 2012 61. Zachenhofer I, Wolfsberger S, Aichholzer M, Bertalanffy A, Roessler K, Kitz K, et al: Gamma-knife radiosurgery for cranial base meningiomas: experience of tumor control, clinical course, and morbidity in a follow-up of more than 8 years. Neu rosurgery 58:28 36, 2006 Manuscript submitted January 18, 2013. Accepted February 4, 2014. Please include this information when citing this paper: published online March 28, 2014; DOI: 10.3171/2014.2.JNS13139. Address correspondence to: Jason P. Sheehan, M.D., Ph.D., Department of Neurological Surgery, University of Virginia, Box 800212, Charlottesville, VA 22908. email: jsheehan@virginia.edu. J Neurosurg / Volume 120 / June 2014 1277