J Neurooncol (2014) 116:187–193 DOI 10.1007/s11060-013-1282-4
CLINICAL STUDY
Single institution experience treating 104 vestibular schwannomas with fractionated stereotactic radiation therapy or stereotactic radiosurgery Bethany M. Anderson • Deepak Khuntia • Søren M. Bentzen • Heather M. Geye Lori L. Hayes • John S. Kuo • Mustafa K. Baskaya • Behnam Badie • Amar Basavatia • G. Mark Pyle • Wolfgang A. Tome´ • Minesh P. Mehta
•
Received: 21 January 2013 / Accepted: 14 October 2013 / Published online: 20 October 2013 Ó Springer Science+Business Media New York 2013
Abstract The pupose of this study is to assess the longterm outcome and toxicity of fractionated stereotactic radiation therapy (FSRT) and stereotactic radiosurgery (SRS) for 100 vestibular schwannomas treated at a single institution. From 1993 to 2007, 104 patients underwent were treated with radiation therapy for vestibular schwannoma. Forty-eight patients received SRS, with a median prescription dose of 12.5 Gy for SRS (range 9.7–16 Gy). For FSRT, two different fraction schedules were employed: a conventional schedule (ConFSRT) of 1.8 Gy per fraction (Gy/F) for 25 or 28 fractions to a total dose of 45 or 50.4 Gy (n = 19); and a once weekly hypofractionated course (HypoFSRT) consisting of 4 Gy/F for 5 fractions to
a total dose of 20 Gy (n = 37). Patients treated with FSRT had better baseline hearing, facial, and trigeminal nerve function, and were more likely to have a diagnosis of NF2. The 5-year progression free rate (PFR) was 97.0 after SRS, 90.5 % after HypoFSRT, and 100.0 % after ConFSRT (p = NS). Univariate analysis demonstrated that NF2 and larger tumor size (greater than the median) correlated with poorer local control, but prior surgical resection did not. Serviceable hearing was preserved in 60.0 % of SRS patients, 63.2 % of HypoFSRT patients, and 44.4 % of ConFSRT patients (p = 0.6). Similarly, there were no significant differences in 5-year rates of trigeminal toxicity facial nerve toxicity, vestibular dysfunction, or tinnitus.
B. M. Anderson (&) S. M. Bentzen H. M. Geye L. L. Hayes A. Basavatia W. A. Tome´ Department of Human Oncology, University of Wisconsin, 600 Highland Avenue, Madison, WI 53792, USA e-mail: [email protected]
M. K. Baskaya e-mail: [email protected]
S. M. Bentzen e-mail: [email protected] L. L. Hayes e-mail: [email protected] A. Basavatia e-mail: [email protected] W. A. Tome´ e-mail: [email protected] D. Khuntia Western Radiation Oncology, 2500 Grant Road, Mountain View, CA 94040, USA e-mail: [email protected]
B. Badie Department of Neurosurgery, City of Hope Cancer Hospital, 1500 East Duarte Road, Duarte, CA 91010, USA e-mail: [email protected] G. M. Pyle Division of Otolaryngology, Department of Surgery, University of Wisconsin, 600 Highland Avenue, Madison, WI 53792, USA e-mail: [email protected] W. A. Tome´ Department of Medical Physics, University of Wisconsin, 600 Highland Avenue, Madison, WI 53792, USA M. P. Mehta Department of Radiation Oncology, University of Maryland, 22 South Greene Street, Baltimore, MD 21201, USA e-mail: [email protected]
J. S. Kuo M. K. Baskaya Department of Neurological Surgery, University of Wisconsin, 600 Highland Avenue, Madison, WI 53792, USA e-mail: [email protected]
123
188
Conclusions:Equivalent 5-year PFR and toxicity rates are shown for patients with vestibular schwanoma selected for SRS, HypoFSRT, and ConFSRT after multidisciplinary evaluation. Factors correlating with tumor progression included NF2 and larger tumor size. Keywords Vestibular schwannoma Acoustic neuroma Stereotactic radiosurgery Fractionated stereotactic radiation therapy Radiotherapy
Introduction Vestibular schwannoma is a benign neoplasm derived from Schwann cells insulating the vestibular portion of the eighth cranial nerve. This tumor occurs with an incidence of 0.6–0.8 per 100,000 person-years. Over 90 % are unilateral; bilateral vestibular schwannoma is a pathognomonic feature of neurofibromatosis type II (NF2). Disease manifestations include hearing loss, tinnitus, vertigo, and cranial nerve deficits, especially nerves VIII, VII, and V. For patients with unilateral disease, vertigo may be the most troublesome symptom, whereas bilateral deafness is the most common disability for patients with neurofibromatosis. Common management options include observation, surgery, or radiation therapy. Radiation techniques employed to treat these patients include stereotactic radiosurgery (SRS) using various treatment systems (i.e. GammaKnife, CyberKnife, linear-accelerator based), or fractionated stereotactic radiotherapy (FSRT) delivered as a series of several small (‘‘conventional’’) or large (‘‘hypofractionated’’) precisely targeted treatments. Recent data suggests that anti-angiogenic agents such as bevacizumab can induce tumor regression and also, in some patients, restore hearing; larger trials with this agent are therefore planned for NF-2 associated vestibular schwannoma [1]. As randomized controlled trials on this patient population are lacking, the choice of treatment modality is based on physician familiarity, bias and choice, patient characteristics, such as age, symptoms, location and size of the lesion(s), and access to different technologies. In the absence of randomized data, comparisons between SRS and FSRT remain anecdotal. To add to the published data for this intercomparison, we present the long-term tumor control and late cranial nerve toxicity rates achieved in a cohort of 103 consecutively treated patients with 104 vestibular schwannomas who underwent SRS or FSRT at a single institution by a single multidisciplinary team. All patients included in this retrospective review, which spans a 14 year period of time, were treated a minimum of 5 years prior to this report to allow the opportunity for long-term follow-up data.
123
J Neurooncol (2014) 116:187–193
Materials and methods This series includes a total of 104 consecutively treated tumors occurring in 103 patients, from 1993 to 2007. The majority of lesions arose sporadically (n = 94), whereas the remainder occurred in patients with NF-2 (n = 10). 48 tumors were treated with SRS and 56 were treated with FSRT, using either a hypofractionated regimen (HypoFSRT) of 20 Gy in 5 once-weekly fractions (n = 37) or a conventionally-fractionated regimen (ConFSRT) of 45–50.4 Gy at 1.8 Gy/F (n = 19). Our patient cohort was evaluated in multidisciplinary fashion by an otolaryngologist and/or neurosurgeon with expertise in the management of vestibular schwannoma, as well as by radiation oncology. This analysis includes only patients treated with radiation therapy, either as primary management (n = 80) or due to disease progression after prior subtotal or gross total resection (n = 24). Patients were typically referred for radiotherapy on the basis of factors making them less well-suited for microsurgical resection, such as age and medical comorbidities, as well as by patient preference. SRS was most strongly considered for patients with sporadic intracanalicular vestibular schwannomas. HypoFSRT was the preferred regimen for patients with larger sporadic lesions located in close proximity to the brainstem. ConFSRT was recommended for patients with the largest tumors (up to 4 cm in maximum dimension) and those with NF-2, based upon the potential to reduce late neurologic toxicity through radiobiological dose sparing. Data was collected by means of retrospective chart review after obtaining institutional IRB approval. Pretreatment characteristics are summarized in Table 1. Chi squared analysis was used to identify significant differences in the pretreatment characteristics of patients treated with SRS versus FSRT. Patients treated with FSRT had better baseline hearing, facial, and trigeminal nerve function, and were more likely to have a diagnosis of NF2. Hearing was defined as serviceable or non-serviceable on the basis of the Gardner–Robertson classification, or subjectively by ability to use the telephone on the affected side if formal audiologic testing was unavailable. Facial nerve function was scored using the House–Brackmann scale. Trigeminal nerve function was assessed in all three divisions (V1–V3) via cranial nerve examination (facial sensation and corneal reflex), with patients scored as ‘‘normal’’ at baseline if no sensory deficits were detected. Vestibular function was scored as ‘‘normal’’ if no deficits were detected on cranial nerve examination and the patient denied symptoms of imbalance or vertigo. Tinnitus was assessed as present or absent, by patient report. SRS was delivered with a modified linear accelerator, with the patient immobilized in a modified Brown–
J Neurooncol (2014) 116:187–193
Roberts–Wells stereotactic frame. Each patient underwent a treatment-planning CT scan after frame placement, which was fused with a gadolinium-enhanced high resolution MRI scan of the skull base obtained no more than 30 days prior to the procedure. The target volume was delineated and prescribed a median peripheral dose of 12.5 Gy (range 9.7–16 Gy). FSRT was delivered on a linear accelerator with the patient immobilized in an aquaplast head mask. To ensure treatment delivery accuracy, pre- and intra-treatment real time patient localization was carried out using optical guidance. A patient-specific maxillary bite block was manufactured and attached to a fiducial array with passive optical markers prior to CT simulation. Prior to the planning CT and prior to each treatment fraction the fiducial array bite block complex was inserted and patient setup was continuously monitored with optical guidance. If the patient moved out of the predetermined error tolerance band of 0.3 mm and/or 0.3 degrees of rotation, treatment was interrupted and continued only after alignment correction. The interested reader is referred to references [2] and [3] for a more detailed description of the technique. A treatment-planning CT scan was performed and fused with a recent gadolinium-enhanced MRI scan. The target volume was delineated and prescribed either 20 Gy in 5 once-weekly fractions (37 tumors) or 45–50.4 Gy in 25–28 daily fractions (19 tumors). The fractionation regimen was individualized based on tumor size, patient preference, and the judgment of the prescribing radiation oncologist. Following treatment, patients were evaluated clinically for toxicity after 1 month, then with MRI scans at least annually through the first 5 years after treatment, at which point the interval of follow-up was typically extended to every other year. During each follow-up visit, a neurological examination was conducted with particular attention to cranial nerve function and balance. Formal audiologic testing was performed as clinically indicated. The Kaplan–Meier method was used to estimate the 5and 10-year progression free rate (PFR) and 5-year hearing preservation rates. Because some tumors enlarged slightly on MRI but then stabilized or decreased in size, disease progression was defined as the need for further therapy, or tumor growth in excess of 5 mm in any axis, or 25 % compared to any baseline dimension. Loss of functional hearing was defined using the most objective available data for each patient, preferably as increase in Gardner–Robertson score to [2 on post-treatment audiograms, or alternatively as loss of the ability to use the telephone on the affected side. Facial function was scored on the House– Brackmann scale. Other acute and late toxicities were scored using the Common Terminology Criteria for Adverse Events (Version 3.0).
189 Table 1 Pre-treatment characteristics of patients treated with SRS, HypoFSRT, and ConFSRT for vestibular schwannoma SRS (n = 48)
FSRT (n = 56)
p value
1.35 cm3
0.150
Median tumor volume 0.66 cm3
HypoFSRT = 0.89 cm3 ConFSRT = 2.94 cm3 Median tumor maximum dimension 1.5 cm
1.7 cm
0.575
HypoFSRT = 1.5 cm ConFSRT = 1.8 cm Serviceable hearing 25.0 %
53.6 %
0.003
HypoFSRT = 56.8 % ConFSRT = 47.4 % House–Brackmann Grade 1 72.9 %
89.3 %
0.031
HypoFSRT = 97.3 % ConFSRT = 73.7 % Normal trigeminal nerve function 62.5 %
82.1 %
0.024
HypoFSRT = 91.1 % ConFSRT = 73.7 % Normal vestibular function 37.5 %
35.7 % HypoFSRT = 32.4 %
0.704
ConFSRT = 36.8 % Without tinnitus 62.5 %
32.1 %
0.002
HypoFSRT = 35.1 % ConFSRT = 26.3 % Prior surgical resection 22.9 %
21.4 %
0.971
HypoFSRT = 13.5 % ConFSRT = 36.8 % NF2 2.1 %
16.1 %
0.016
HypoFSRT = 5.4 % ConFSRT = 36.8 % Tumor size was estimated both by target volume (cm3) from radiation treatment planning software and as maximum dimension (cm) from cross-sectional imaging. Chi squared analysis was used to compare the SRS and overall FSRT patient populations
Results Local control The median follow-up time was 52.1 months for the entire patient cohort; 83.6 months after SRS, 43.1 months after HypoFSRT, and 53.6 months after ConFSRT. A total of 7 patients experienced disease progression at a median of
123
190
J Neurooncol (2014) 116:187–193
53.6 months (range 7.4–84.4 months), as shown in Fig. 1. The actuarial 5-year PFR was 97.0 ± 3.0 % for patients treated with SRS, 90.5 ± 5.2 % after HypoFSRT, and 100 % after ConFSRT, with no statistically significant difference between the treatment groups (p = 0.09). The 10-year PFR for the SRS group was 85.2 ± 6.9 %. The majority of the progression-events occurred within 5 years, but a small number were delayed beyond 5 years. Univariate analysis demonstrated that NF2 correlated with poorer local control (p = 0.053), as did tumor size larger than the median (p = 0.026), but prior surgical resection did not. Hearing At the time of treatment, a total of 42 patients had serviceable hearing; 12 patients in the SRS group, 21 patients in the HypoFSRT group, and 9 in the ConFSRT group. Serviceable hearing was defined as Gardner–Robertson score B2, or ability to use the telephone in the affected ear if a baseline audiogram was unavailable. At 5 years, serviceable hearing was preserved in 60.0 ± 15.5 % of SRS patients, 63.2 ± 11.4 % of HypoFSRT patients, and 44.4 ± 16.6 % of ConFSRT patients (p = 0.6), as depicted in Fig. 2. Three patients treated with HypoFSRT reported regaining serviceable hearing after radiation therapy. The majority of changes in hearing status occurred within the first 3 years, irrespective of treatment modality.
Fig. 2 Hearing preservation after SRS, or HypoFSRT, and ConFSRT amongst patients who had serviceable hearing at the time of treatment for vestibular schwannoma
Table 2 Trigeminal nerve, facial nerve, vestibular function and tinnitus after SRS, HypoFSRT or ConFSRT for vestibular schwannoma Improved no. pts (crude %)
Worsened no. pts (crude %)
SRS
3 (6.3)
4 (8.3)
HypoFSRT ConFSRT
1 (2.7) 0 (0.0)
0 (0.0) 1 (5.3)
Facial sensation (V)
Trigeminal neuralgia (V)
Trigeminal nerve, facial nerve, vestibular function and tinnitus Overall, SRS and FSRT had little impact on other cranial nerve function, as shown in Table 2. Grade 2 trigeminal
SRS
0 (0.0)
2 (4.2)
HypoFSRT
0 (0.0)
1 (2.7)
ConFSRT
0 (0.0)
0 (0.0)
SRS
1 (2.1)
1 (2.1)
HypoFSRT
0 (0.0)
0 (0.0)
ConFSRT
1 (5.3)
0 (0.0)
Facial weakness (VII)
Vestibular function (VIII) SRS
3 (6.3)
0 (0.0)
HypoFSRT
3 (8.1)
1 (2.7)
ConFSRT
1 (2.7)
1 (2.7)
SRS HypoFSRT
4 (8.3) 4 (10.8)
1 (2.1) 1 (2.7)
ConFSRT
3 (15.8)
0 (0.0)
Tinnitus (VIII)
Fig. 1 Local control (progression free rate) after SRS, HypoFSRT, or ConFSRT for vestibular schwannoma
123
nerve symptoms developed in 7 patients (6.7 %) after treatment, including lack of sensation (3 patients SRS, 1 patient ConFSRT), neuralgia (1 patient SRS, 1 patient HypoFSRT), or both (1 patient SRS). The actuarial 5-year rate of new CN V toxicity was 10.5 ± 5.1 % for SRS, 25.0 % ± 21.7 % for HypoFSRT and 5.3 ± 5.1 % for
J Neurooncol (2014) 116:187–193
ConFSRT (p = 0.8). Facial sensation was regained in 3 patients treated with SRS and 1 patient treated with HypoFSRT. Significant facial nerve toxicity (increase of C2 points on House–Brackmann scale) developed in one patient treated with SRS, for an actuarial facial nerve toxicity rate of 2.1 ± 2.1 % at 5 years, not significantly different from patients treated with HypoFSRT (0.0 %) or ConFSRT (0.0 %; p = 0.6). One patient treated with SRS and one patient treated with ConFSRT had significant improvement in facial nerve function (decrease of C2 points on House– Brackmann scale). Self-reported vestibular function (i.e. dizziness, ataxia, vertigo) and tinnitus tended to improve with follow-up, but no formal vestibular testing was prospectively performed. New vestibular dysfunction did not develop in any SRS patient in our series, and occurred in 1 patients treated with HypoFSRT and 1 patient treated with ConFSRT, resulting in a 5-year actuarial toxicity rate of 2.9 ± 2.8 % for HypoFSRT and 5.6 ± 5.4 % for ConFSRT (p = 0.3). The 5-year actuarial estimate of new-onset tinnitus was 2.2 ± 2.2 % after SRS, 2.9 ± 2.9 % after HypoFSRT, and 0.0 % after ConFSRT (p = 0.7).
Discussion In the absence of data from randomized controlled trials, physicians treating vestibular schwannoma must rely upon published institutional outcomes to guide clinical decisionmaking. In modern series, local control rates are consistently greater than 90 % after surgical resection, FSRT or SRS, with no significant difference between these treatment methods [4–7, 10–12]. Our data indicate that the vast majority of failure events occur within the first 5 years, and therefore when reviewing the reported literature, one must cautiously interpret data where the follow-up is considerably shorter than 5 years. Given the relatively comparable local control outcomes, preservation of function becomes an important consideration. In our experience, there was no significant difference in hearing preservation or cranial nerve toxicity after SRS versus FSRT. These data must be interpreted cautiously, however, as the two patient groups were not comparable at baseline (Table 1). More patients in the FSRT group had serviceable hearing at the time of treatment, in spite of the fact that the FSRT patients had a nonsignificant trend toward larger tumor size and were more likely to have NF2. This could imply that the tumors in SRS-treated patients, which were predominantly intracanalicular, might have had faster growth rates, resulting in more rapid compression/compromise of the auditory nerve and/or its vasculature, and this rapid pressure change could be the
191
differentiating factor, rather than treatment. Also, the SRS group has longer follow-up and therefore statistically is more likely to demonstrate hearing loss, especially late events. It must however be borne in mind that for both SRS and FSRT, hearing loss plateaus after 3 years, and therefore, the differential in follow-up time may be irrelevant for this endpoint, given that all patients had longer than 3 years of follow-up. Meijer et al. [7] conducted a prospective study of 129 patients with vestibular schwannomas who were allocated to SRS versus hypofractionated FSRT on the basis of dentition, with dentate patients (n = 80) receiving FSRT and edentulous patients (n = 49) receiving SRS. Toxicity rates were low for both groups, but trigeminal nerve preservation was slightly higher in the FSRT group (98 vs. 92 % at 5 years, p = 0.048). A review of 125 patients treated with SRS (n = 69) or conventional FSRT (n = 56) at Thomas Jefferson University also showed comparable outcomes, with the exception for the parameter of functional hearing preservation, which was better for their patients treated with FSRT [6]. Analysis of 202 vestibular schwannomas treated with SRS (n = 30) versus conventional FSRT (n = 172) at the University of Heidelberg found equivalent outcomes between the two groups, with the exception of patients treated with SRS doses [13 Gy, who had lower rates of hearing preservation [8]. Recently, Collen et al. [9] compared the outcomes achieved with SRS (n = 78) or FSRT (n = 41) at the University of Brussels, and found superior facial nerve preservation rates with FSRT (97 vs. 83 %, p = 0.047). Multivariate analysis found that RT schedule and baseline Koos group predicted for facial nerve toxicity. These data, as well as ours are summarized in Table 3. Vestibular dysfunction is another symptom that merits attention, considering a recent study demonstrating that vertigo negatively impacts quality of life more so than hearing loss in patients with unilateral vestibular schwannomas [10]. These issues are less commonly addressed in the literature, perhaps due to the fact that most published series are retrospective in nature, and formal assessment of vestibular function and quality of life are infrequently performed in routine clinical care. In our experience, patients’ subjective reports of vestibular dysfunction (i.e. dizziness, ataxia, vertigo) tend to remain stable (90 %) or improve (7 %) after SRS or FSRT, and rarely worsen significantly (3 %). Another issue to consider, perhaps more from a societal perspective than from an individual patient perspective, is the overall cost of treatment. Banerjee et al. [11] recently analyzed the initial and follow-up costs over at least 36 months for 53 patients treated with microsurgery or SRS. Although the initial cost of surgical resection was higher, with a mean cost of $23,788 versus $16,143 for
123
192
J Neurooncol (2014) 116:187–193
Table 3 Single-institution comparisons of outcomes treating vestibular schwannoma with SRS, FSRT, or surgery Series
Treatment (no. pts)
Serviceable hearing preservation
CN VII toxicity
CN V toxicity
SRS (63) Surgery (28)
68 %
1.6 %
NR
0 %, p \ 0.001
46 %, p \ 0.001
Surgery vs. SRS Myrseth et al. [18] Pollock et al. [19]
SRS (46)
63 %
2%
Surgery (36)
5 %, p \ 0.001
17 %, p = 0.04
NR
Re´gis et al. [20]
SRS (100)
40 %
0%
4%
Karpinos et al. [5]
Surgery (110) SRS (75)
5 %, p = 0.000001 44 %
47 %, p = 0.00005 35.3 %
29 %, p = 0.0009 22 %
Surgery (25)
40 %
6.1 %, p = 0.008
12.2 %, p = 0.009
SRS (48)
60.0 %
2.1 %
10.5 %
ConFSRT (19)
44.4 %
0%
5.3 %
HypoFSRT (37)
63.2 %
0%
25.0 %
SRS (78)
59 %
16 %
6%
ConFSRT (10) or HypoFSRT (32)
82 %
3%
3%
SRS (30)
NR
SRS vs. FSRT Anderson et al. current series
Collen et al. [9] Combs et al. [8]
ConFSRT (172) Meijer et al. [7] Andrews et al. [6]
17 %
7%
2%
3%
SRS (49)
75 %
7%
8%
HypoFSRT (80)
61 %
3%
2 %, p = 0.048
SRS (69)
33 %
2%
5%
ConFSRT (56)
81 %, p = 0.0228
2%
7%
p-values included for statistically significant differences SRS stereotactic radiosurgery, ConFSRT conventionally fractionated stereotactic radiation therapy, HypoFSRT hypofractionated FSRT, NR not reported (serviceable hearing preservation rates not compatible with table format)
SRS, follow-up costs decreased over time after surgery, and increased after SRS. The authors concluded that SRS is likely to be less expensive than surgery if long-term local control rates remain high at the lower doses currently prescribed, and the future frequency of follow-up MR imaging is decreased. Such a comparison is currently lacking between SRS and FSRT. What conclusions can be made, then, regarding the ideal treatment for vestibular schwannoma? First and foremost, considering the myriad of hypothesis-generating data that have emerged in recent years, it is clear that patients are best served by a multidisciplinary approach to individualize treatment recommendations for each patient on the basis of current knowledge combined with tumor characteristics, co-morbidities and patient-specific concerns. Our patient cohort was evaluated in multidisciplinary fashion, and includes only patients treated with a nonsurgical approach. Maximum tumor dimension greater than 3 cm is commonly considered to be a contraindication for SRS; indeed, our patient series included only four tumors greater than 3 cm in maximum dimension, of which only two were treated with SRS in the 1990s. The ideal marginal dose for SRS is most likely 12–13 Gy, since local control appears to
123
be compromised with lower doses [12, 13], and cranial nerve toxicity increases with higher doses [8, 14–16]. For FSRT, the best dose and fractionation regimen remains unknown, as good outcomes are being reported with both conventional and hypofractionated approaches. The small number of patients treated with hypofractionated versus conventionally fractionated FSRT in our series prohibits detailed comparison of the outcomes between these two patient groups. A recent analysis of 89 patients treated with 1.8 Gy fractions did find better hearing preservation and equal local control in patients treated to 46.8 versus 50.4 Gy [17]. We have tended to select patients with larger tumors and potential brainstem compression for FSRT, and smaller, predominantly intracanalicular tumors for SRS. In conclusion, vestibular schwannoma can be treated with SRS, conventionally fractionated FSRT and hypofractionated FSRT, resulting in high local control and low toxicity rates. For larger tumors, FSRT offers equivalent 5-year PFR compared with single fraction SRS. In order to truly ascertain a significant difference between these techniques, a formal randomized trial would be required, but given the relatively small differences in outcome, such a trial might require a very large number of patients. A
J Neurooncol (2014) 116:187–193
193
cost-effectiveness analysis of the various radiotherapy approaches would also be useful. Conflict of interest of interest.
The authors declare that they have no conflict
10.
11.
References 1. Plotkin SR, Stemmer-Rachamimov AO, Barker FG, Halpin C, Padera TP, Tyrrell A, Sorensen AG, Jain RK, di Tomaso E (2009) Hearing improvement after bevacizumab in patients with neurofibromatosis type 2. N Engl J Med 361:358–367 2. Tome´ WA, Meeks SL, Buatti JM, Bova FJ, Friedman WA, Li Z (2000) A high-precision system for conformal intracranial radiotherapy. Int J Radiat Oncol Biol Phys 47:1137–1143 3. Tome´ WA, Meeks SL, McNutt TR, Buatti JM, Bova FJ, Friedman WA, Mehta M (2001) Optically guided intensity modulated radiotherapy. Radiother Oncol 61:33–44 4. Pollock BE (2006) Management of vestibular schwannomas that enlarge after stereotactic radiosurgery: treatment recommendations based on a 15 year experience. Neurosurg 58:241–248 5. Karpinos M, Teh BS, Zeck O, Carpenter LS, Phan C, Mai WY, Lu HH, Chiu JK, Butler EB, Gormley WB, Woo SY (2002) Treatment of acoustic neuroma: stereotactic radiosurgery vs. microsurgery. Int J Radiat Oncol Biol Phys 54:1410–1421 6. Andrews DW, Suarez O, Goldman HW, Downes MB, Bednarz G, Corn BW, Werner-Wasik M, Rosenstock J, Curran WJ (2001) Stereotactic radiosurgery and fractionated stereotactic radiotherapy for the treatment of acoustic schannomas: comparative observations of 125 patients treated at one institution. Int J Radiat Oncol Biol Phys 50:1265–1278 7. Meijer OW, Vandertop WP, Baayen JC, Slotman BJ (2003) Single-fraction vs. fractionated linac-based stereotactic radiosurgery for vestibular schwannoma: a single-institution study. Int J Radiat Oncol Biol Phys 56:1390–1396 8. Combs SE, Welzel T, Schulz-Ertner D, Huber PE, Debus J (2010) Differences in clinical results after linac-based single-dose radiosurgery versus fractionated stereotactic radiotherapy for patients with vestibular schwannoma. Int J Radiat Oncol Biol Phys 76:193–200 9. Collen C, Ampe B, Gevaert T, Moens M, Linthout N, De Ridder M, Verellen D, D’Haens J, Storme G (2011) Single fraction
12.
13.
14.
15.
16.
17.
18.
19.
20.
versus fractionated linac-based stereotactic radiotherapy for vestibular schwannoma: a single-institution experience. Int J Radiat Oncol Biol Phys 81:e503–e509 Myrseth E, Moller P, Wentzel-Larsen T, Goplen F, Lund-Johansen M (2006) Vestibular schwannoma: surgery or gamma knife radiosurgery? A prospective, nonrandomized study. Neurosurg 59:67–76 Banerjee R, Moriarty JP, Foote RL, Pollock BE (2008) Comparison of the surgical and follow-up costs associated with microsurgical resection and stereotactic radiosurgery for vestibular schwannoma. J Neurosurg 108:1220–1224 Foote KD, Friedman WA, Buatti JM, Meeks SL, Bova FJ, Kubilis PS (2001) Analysis of risk factors associated with radiosurgery for vestibular schwannoma. J Neurosurg 95:440–449 Chung WY, Liu KD, Shiau CY, Wu HM, Wang LW, Guo WY, Ho DM, Pan DH (2005) Gamma knife surgery for vestibular schwannoma: 10-year experience of 195 cases. J Neurosurg 102(Suppl):87–96 Flickinger JC, Kondziolka D, Niranjan A, Lunsford LD (2001) Results of acoustic neuroma radiosurgery: an analysis of 5 years’ experience using current methods. J Neurosurg 94:1–6 Flickinger JC, Kondziolka D, Pollock BE, Lunsford LD (1996) Evolution in technique for vestibular schwannoma radiosurgery and effect on outcome. Int J Radiat Oncol Biol Phys 36:275–280 Chihara Y, Ito K, Sugasawa K, Shin M (2007) Neurological complications after acoustic neurinoma radiosurgery: revised risk factors based on long-term follow-up. Acta Otolaryngol Suppl 559:65–70 Andrews DW, Werner-Wasik M, Den RB, Paek SH, DownesPhillips B, Willcox TO, Bednarz G, Maltenfort M, Evans JJ, Curran WJ (2009) Toward dose optimization for fractionated stereotactic radiotherapy for acoustic neuromas: comparison of two dose cohorts. Int J Radiat Oncol Biol Phys 74:419–426 Myrseth E, Moller P, Pedersen PH, Lund-Johansen M (2009) Vestibular schwannoma: surgery or gamma knife radiosurgery? A prospective, nonrandomized study. Neurosurg 64:654–661 Pollock BE, Driscoll CL, Foote RL, Link MJ, Gorman DA, Bauch CD, Mandrekar JN, Krecke KN, Johnson CH (2006) Patient outcomes after vestibular schwannoma management: a prospective comparison of microsurgical resection and stereotactic radiosurgery. Neurosurg 59:77–85 Regis J, Pellet W, Delsanti C, Dufour H, Roche PH, Thomassin JM, Zanaret M, Peragut JC (2002) Functional outcome after gamma knife surgery or microsurgery for vestibular schwannomas. J Neurosurg 97:1091–1100
123
CLINICAL STUDY
Single institution experience treating 104 vestibular schwannomas with fractionated stereotactic radiation therapy or stereotactic radiosurgery Bethany M. Anderson • Deepak Khuntia • Søren M. Bentzen • Heather M. Geye Lori L. Hayes • John S. Kuo • Mustafa K. Baskaya • Behnam Badie • Amar Basavatia • G. Mark Pyle • Wolfgang A. Tome´ • Minesh P. Mehta
•
Received: 21 January 2013 / Accepted: 14 October 2013 / Published online: 20 October 2013 Ó Springer Science+Business Media New York 2013
Abstract The pupose of this study is to assess the longterm outcome and toxicity of fractionated stereotactic radiation therapy (FSRT) and stereotactic radiosurgery (SRS) for 100 vestibular schwannomas treated at a single institution. From 1993 to 2007, 104 patients underwent were treated with radiation therapy for vestibular schwannoma. Forty-eight patients received SRS, with a median prescription dose of 12.5 Gy for SRS (range 9.7–16 Gy). For FSRT, two different fraction schedules were employed: a conventional schedule (ConFSRT) of 1.8 Gy per fraction (Gy/F) for 25 or 28 fractions to a total dose of 45 or 50.4 Gy (n = 19); and a once weekly hypofractionated course (HypoFSRT) consisting of 4 Gy/F for 5 fractions to
a total dose of 20 Gy (n = 37). Patients treated with FSRT had better baseline hearing, facial, and trigeminal nerve function, and were more likely to have a diagnosis of NF2. The 5-year progression free rate (PFR) was 97.0 after SRS, 90.5 % after HypoFSRT, and 100.0 % after ConFSRT (p = NS). Univariate analysis demonstrated that NF2 and larger tumor size (greater than the median) correlated with poorer local control, but prior surgical resection did not. Serviceable hearing was preserved in 60.0 % of SRS patients, 63.2 % of HypoFSRT patients, and 44.4 % of ConFSRT patients (p = 0.6). Similarly, there were no significant differences in 5-year rates of trigeminal toxicity facial nerve toxicity, vestibular dysfunction, or tinnitus.
B. M. Anderson (&) S. M. Bentzen H. M. Geye L. L. Hayes A. Basavatia W. A. Tome´ Department of Human Oncology, University of Wisconsin, 600 Highland Avenue, Madison, WI 53792, USA e-mail: [email protected]
M. K. Baskaya e-mail: [email protected]
S. M. Bentzen e-mail: [email protected] L. L. Hayes e-mail: [email protected] A. Basavatia e-mail: [email protected] W. A. Tome´ e-mail: [email protected] D. Khuntia Western Radiation Oncology, 2500 Grant Road, Mountain View, CA 94040, USA e-mail: [email protected]
B. Badie Department of Neurosurgery, City of Hope Cancer Hospital, 1500 East Duarte Road, Duarte, CA 91010, USA e-mail: [email protected] G. M. Pyle Division of Otolaryngology, Department of Surgery, University of Wisconsin, 600 Highland Avenue, Madison, WI 53792, USA e-mail: [email protected] W. A. Tome´ Department of Medical Physics, University of Wisconsin, 600 Highland Avenue, Madison, WI 53792, USA M. P. Mehta Department of Radiation Oncology, University of Maryland, 22 South Greene Street, Baltimore, MD 21201, USA e-mail: [email protected]
J. S. Kuo M. K. Baskaya Department of Neurological Surgery, University of Wisconsin, 600 Highland Avenue, Madison, WI 53792, USA e-mail: [email protected]
123
188
Conclusions:Equivalent 5-year PFR and toxicity rates are shown for patients with vestibular schwanoma selected for SRS, HypoFSRT, and ConFSRT after multidisciplinary evaluation. Factors correlating with tumor progression included NF2 and larger tumor size. Keywords Vestibular schwannoma Acoustic neuroma Stereotactic radiosurgery Fractionated stereotactic radiation therapy Radiotherapy
Introduction Vestibular schwannoma is a benign neoplasm derived from Schwann cells insulating the vestibular portion of the eighth cranial nerve. This tumor occurs with an incidence of 0.6–0.8 per 100,000 person-years. Over 90 % are unilateral; bilateral vestibular schwannoma is a pathognomonic feature of neurofibromatosis type II (NF2). Disease manifestations include hearing loss, tinnitus, vertigo, and cranial nerve deficits, especially nerves VIII, VII, and V. For patients with unilateral disease, vertigo may be the most troublesome symptom, whereas bilateral deafness is the most common disability for patients with neurofibromatosis. Common management options include observation, surgery, or radiation therapy. Radiation techniques employed to treat these patients include stereotactic radiosurgery (SRS) using various treatment systems (i.e. GammaKnife, CyberKnife, linear-accelerator based), or fractionated stereotactic radiotherapy (FSRT) delivered as a series of several small (‘‘conventional’’) or large (‘‘hypofractionated’’) precisely targeted treatments. Recent data suggests that anti-angiogenic agents such as bevacizumab can induce tumor regression and also, in some patients, restore hearing; larger trials with this agent are therefore planned for NF-2 associated vestibular schwannoma [1]. As randomized controlled trials on this patient population are lacking, the choice of treatment modality is based on physician familiarity, bias and choice, patient characteristics, such as age, symptoms, location and size of the lesion(s), and access to different technologies. In the absence of randomized data, comparisons between SRS and FSRT remain anecdotal. To add to the published data for this intercomparison, we present the long-term tumor control and late cranial nerve toxicity rates achieved in a cohort of 103 consecutively treated patients with 104 vestibular schwannomas who underwent SRS or FSRT at a single institution by a single multidisciplinary team. All patients included in this retrospective review, which spans a 14 year period of time, were treated a minimum of 5 years prior to this report to allow the opportunity for long-term follow-up data.
123
J Neurooncol (2014) 116:187–193
Materials and methods This series includes a total of 104 consecutively treated tumors occurring in 103 patients, from 1993 to 2007. The majority of lesions arose sporadically (n = 94), whereas the remainder occurred in patients with NF-2 (n = 10). 48 tumors were treated with SRS and 56 were treated with FSRT, using either a hypofractionated regimen (HypoFSRT) of 20 Gy in 5 once-weekly fractions (n = 37) or a conventionally-fractionated regimen (ConFSRT) of 45–50.4 Gy at 1.8 Gy/F (n = 19). Our patient cohort was evaluated in multidisciplinary fashion by an otolaryngologist and/or neurosurgeon with expertise in the management of vestibular schwannoma, as well as by radiation oncology. This analysis includes only patients treated with radiation therapy, either as primary management (n = 80) or due to disease progression after prior subtotal or gross total resection (n = 24). Patients were typically referred for radiotherapy on the basis of factors making them less well-suited for microsurgical resection, such as age and medical comorbidities, as well as by patient preference. SRS was most strongly considered for patients with sporadic intracanalicular vestibular schwannomas. HypoFSRT was the preferred regimen for patients with larger sporadic lesions located in close proximity to the brainstem. ConFSRT was recommended for patients with the largest tumors (up to 4 cm in maximum dimension) and those with NF-2, based upon the potential to reduce late neurologic toxicity through radiobiological dose sparing. Data was collected by means of retrospective chart review after obtaining institutional IRB approval. Pretreatment characteristics are summarized in Table 1. Chi squared analysis was used to identify significant differences in the pretreatment characteristics of patients treated with SRS versus FSRT. Patients treated with FSRT had better baseline hearing, facial, and trigeminal nerve function, and were more likely to have a diagnosis of NF2. Hearing was defined as serviceable or non-serviceable on the basis of the Gardner–Robertson classification, or subjectively by ability to use the telephone on the affected side if formal audiologic testing was unavailable. Facial nerve function was scored using the House–Brackmann scale. Trigeminal nerve function was assessed in all three divisions (V1–V3) via cranial nerve examination (facial sensation and corneal reflex), with patients scored as ‘‘normal’’ at baseline if no sensory deficits were detected. Vestibular function was scored as ‘‘normal’’ if no deficits were detected on cranial nerve examination and the patient denied symptoms of imbalance or vertigo. Tinnitus was assessed as present or absent, by patient report. SRS was delivered with a modified linear accelerator, with the patient immobilized in a modified Brown–
J Neurooncol (2014) 116:187–193
Roberts–Wells stereotactic frame. Each patient underwent a treatment-planning CT scan after frame placement, which was fused with a gadolinium-enhanced high resolution MRI scan of the skull base obtained no more than 30 days prior to the procedure. The target volume was delineated and prescribed a median peripheral dose of 12.5 Gy (range 9.7–16 Gy). FSRT was delivered on a linear accelerator with the patient immobilized in an aquaplast head mask. To ensure treatment delivery accuracy, pre- and intra-treatment real time patient localization was carried out using optical guidance. A patient-specific maxillary bite block was manufactured and attached to a fiducial array with passive optical markers prior to CT simulation. Prior to the planning CT and prior to each treatment fraction the fiducial array bite block complex was inserted and patient setup was continuously monitored with optical guidance. If the patient moved out of the predetermined error tolerance band of 0.3 mm and/or 0.3 degrees of rotation, treatment was interrupted and continued only after alignment correction. The interested reader is referred to references [2] and [3] for a more detailed description of the technique. A treatment-planning CT scan was performed and fused with a recent gadolinium-enhanced MRI scan. The target volume was delineated and prescribed either 20 Gy in 5 once-weekly fractions (37 tumors) or 45–50.4 Gy in 25–28 daily fractions (19 tumors). The fractionation regimen was individualized based on tumor size, patient preference, and the judgment of the prescribing radiation oncologist. Following treatment, patients were evaluated clinically for toxicity after 1 month, then with MRI scans at least annually through the first 5 years after treatment, at which point the interval of follow-up was typically extended to every other year. During each follow-up visit, a neurological examination was conducted with particular attention to cranial nerve function and balance. Formal audiologic testing was performed as clinically indicated. The Kaplan–Meier method was used to estimate the 5and 10-year progression free rate (PFR) and 5-year hearing preservation rates. Because some tumors enlarged slightly on MRI but then stabilized or decreased in size, disease progression was defined as the need for further therapy, or tumor growth in excess of 5 mm in any axis, or 25 % compared to any baseline dimension. Loss of functional hearing was defined using the most objective available data for each patient, preferably as increase in Gardner–Robertson score to [2 on post-treatment audiograms, or alternatively as loss of the ability to use the telephone on the affected side. Facial function was scored on the House– Brackmann scale. Other acute and late toxicities were scored using the Common Terminology Criteria for Adverse Events (Version 3.0).
189 Table 1 Pre-treatment characteristics of patients treated with SRS, HypoFSRT, and ConFSRT for vestibular schwannoma SRS (n = 48)
FSRT (n = 56)
p value
1.35 cm3
0.150
Median tumor volume 0.66 cm3
HypoFSRT = 0.89 cm3 ConFSRT = 2.94 cm3 Median tumor maximum dimension 1.5 cm
1.7 cm
0.575
HypoFSRT = 1.5 cm ConFSRT = 1.8 cm Serviceable hearing 25.0 %
53.6 %
0.003
HypoFSRT = 56.8 % ConFSRT = 47.4 % House–Brackmann Grade 1 72.9 %
89.3 %
0.031
HypoFSRT = 97.3 % ConFSRT = 73.7 % Normal trigeminal nerve function 62.5 %
82.1 %
0.024
HypoFSRT = 91.1 % ConFSRT = 73.7 % Normal vestibular function 37.5 %
35.7 % HypoFSRT = 32.4 %
0.704
ConFSRT = 36.8 % Without tinnitus 62.5 %
32.1 %
0.002
HypoFSRT = 35.1 % ConFSRT = 26.3 % Prior surgical resection 22.9 %
21.4 %
0.971
HypoFSRT = 13.5 % ConFSRT = 36.8 % NF2 2.1 %
16.1 %
0.016
HypoFSRT = 5.4 % ConFSRT = 36.8 % Tumor size was estimated both by target volume (cm3) from radiation treatment planning software and as maximum dimension (cm) from cross-sectional imaging. Chi squared analysis was used to compare the SRS and overall FSRT patient populations
Results Local control The median follow-up time was 52.1 months for the entire patient cohort; 83.6 months after SRS, 43.1 months after HypoFSRT, and 53.6 months after ConFSRT. A total of 7 patients experienced disease progression at a median of
123
190
J Neurooncol (2014) 116:187–193
53.6 months (range 7.4–84.4 months), as shown in Fig. 1. The actuarial 5-year PFR was 97.0 ± 3.0 % for patients treated with SRS, 90.5 ± 5.2 % after HypoFSRT, and 100 % after ConFSRT, with no statistically significant difference between the treatment groups (p = 0.09). The 10-year PFR for the SRS group was 85.2 ± 6.9 %. The majority of the progression-events occurred within 5 years, but a small number were delayed beyond 5 years. Univariate analysis demonstrated that NF2 correlated with poorer local control (p = 0.053), as did tumor size larger than the median (p = 0.026), but prior surgical resection did not. Hearing At the time of treatment, a total of 42 patients had serviceable hearing; 12 patients in the SRS group, 21 patients in the HypoFSRT group, and 9 in the ConFSRT group. Serviceable hearing was defined as Gardner–Robertson score B2, or ability to use the telephone in the affected ear if a baseline audiogram was unavailable. At 5 years, serviceable hearing was preserved in 60.0 ± 15.5 % of SRS patients, 63.2 ± 11.4 % of HypoFSRT patients, and 44.4 ± 16.6 % of ConFSRT patients (p = 0.6), as depicted in Fig. 2. Three patients treated with HypoFSRT reported regaining serviceable hearing after radiation therapy. The majority of changes in hearing status occurred within the first 3 years, irrespective of treatment modality.
Fig. 2 Hearing preservation after SRS, or HypoFSRT, and ConFSRT amongst patients who had serviceable hearing at the time of treatment for vestibular schwannoma
Table 2 Trigeminal nerve, facial nerve, vestibular function and tinnitus after SRS, HypoFSRT or ConFSRT for vestibular schwannoma Improved no. pts (crude %)
Worsened no. pts (crude %)
SRS
3 (6.3)
4 (8.3)
HypoFSRT ConFSRT
1 (2.7) 0 (0.0)
0 (0.0) 1 (5.3)
Facial sensation (V)
Trigeminal neuralgia (V)
Trigeminal nerve, facial nerve, vestibular function and tinnitus Overall, SRS and FSRT had little impact on other cranial nerve function, as shown in Table 2. Grade 2 trigeminal
SRS
0 (0.0)
2 (4.2)
HypoFSRT
0 (0.0)
1 (2.7)
ConFSRT
0 (0.0)
0 (0.0)
SRS
1 (2.1)
1 (2.1)
HypoFSRT
0 (0.0)
0 (0.0)
ConFSRT
1 (5.3)
0 (0.0)
Facial weakness (VII)
Vestibular function (VIII) SRS
3 (6.3)
0 (0.0)
HypoFSRT
3 (8.1)
1 (2.7)
ConFSRT
1 (2.7)
1 (2.7)
SRS HypoFSRT
4 (8.3) 4 (10.8)
1 (2.1) 1 (2.7)
ConFSRT
3 (15.8)
0 (0.0)
Tinnitus (VIII)
Fig. 1 Local control (progression free rate) after SRS, HypoFSRT, or ConFSRT for vestibular schwannoma
123
nerve symptoms developed in 7 patients (6.7 %) after treatment, including lack of sensation (3 patients SRS, 1 patient ConFSRT), neuralgia (1 patient SRS, 1 patient HypoFSRT), or both (1 patient SRS). The actuarial 5-year rate of new CN V toxicity was 10.5 ± 5.1 % for SRS, 25.0 % ± 21.7 % for HypoFSRT and 5.3 ± 5.1 % for
J Neurooncol (2014) 116:187–193
ConFSRT (p = 0.8). Facial sensation was regained in 3 patients treated with SRS and 1 patient treated with HypoFSRT. Significant facial nerve toxicity (increase of C2 points on House–Brackmann scale) developed in one patient treated with SRS, for an actuarial facial nerve toxicity rate of 2.1 ± 2.1 % at 5 years, not significantly different from patients treated with HypoFSRT (0.0 %) or ConFSRT (0.0 %; p = 0.6). One patient treated with SRS and one patient treated with ConFSRT had significant improvement in facial nerve function (decrease of C2 points on House– Brackmann scale). Self-reported vestibular function (i.e. dizziness, ataxia, vertigo) and tinnitus tended to improve with follow-up, but no formal vestibular testing was prospectively performed. New vestibular dysfunction did not develop in any SRS patient in our series, and occurred in 1 patients treated with HypoFSRT and 1 patient treated with ConFSRT, resulting in a 5-year actuarial toxicity rate of 2.9 ± 2.8 % for HypoFSRT and 5.6 ± 5.4 % for ConFSRT (p = 0.3). The 5-year actuarial estimate of new-onset tinnitus was 2.2 ± 2.2 % after SRS, 2.9 ± 2.9 % after HypoFSRT, and 0.0 % after ConFSRT (p = 0.7).
Discussion In the absence of data from randomized controlled trials, physicians treating vestibular schwannoma must rely upon published institutional outcomes to guide clinical decisionmaking. In modern series, local control rates are consistently greater than 90 % after surgical resection, FSRT or SRS, with no significant difference between these treatment methods [4–7, 10–12]. Our data indicate that the vast majority of failure events occur within the first 5 years, and therefore when reviewing the reported literature, one must cautiously interpret data where the follow-up is considerably shorter than 5 years. Given the relatively comparable local control outcomes, preservation of function becomes an important consideration. In our experience, there was no significant difference in hearing preservation or cranial nerve toxicity after SRS versus FSRT. These data must be interpreted cautiously, however, as the two patient groups were not comparable at baseline (Table 1). More patients in the FSRT group had serviceable hearing at the time of treatment, in spite of the fact that the FSRT patients had a nonsignificant trend toward larger tumor size and were more likely to have NF2. This could imply that the tumors in SRS-treated patients, which were predominantly intracanalicular, might have had faster growth rates, resulting in more rapid compression/compromise of the auditory nerve and/or its vasculature, and this rapid pressure change could be the
191
differentiating factor, rather than treatment. Also, the SRS group has longer follow-up and therefore statistically is more likely to demonstrate hearing loss, especially late events. It must however be borne in mind that for both SRS and FSRT, hearing loss plateaus after 3 years, and therefore, the differential in follow-up time may be irrelevant for this endpoint, given that all patients had longer than 3 years of follow-up. Meijer et al. [7] conducted a prospective study of 129 patients with vestibular schwannomas who were allocated to SRS versus hypofractionated FSRT on the basis of dentition, with dentate patients (n = 80) receiving FSRT and edentulous patients (n = 49) receiving SRS. Toxicity rates were low for both groups, but trigeminal nerve preservation was slightly higher in the FSRT group (98 vs. 92 % at 5 years, p = 0.048). A review of 125 patients treated with SRS (n = 69) or conventional FSRT (n = 56) at Thomas Jefferson University also showed comparable outcomes, with the exception for the parameter of functional hearing preservation, which was better for their patients treated with FSRT [6]. Analysis of 202 vestibular schwannomas treated with SRS (n = 30) versus conventional FSRT (n = 172) at the University of Heidelberg found equivalent outcomes between the two groups, with the exception of patients treated with SRS doses [13 Gy, who had lower rates of hearing preservation [8]. Recently, Collen et al. [9] compared the outcomes achieved with SRS (n = 78) or FSRT (n = 41) at the University of Brussels, and found superior facial nerve preservation rates with FSRT (97 vs. 83 %, p = 0.047). Multivariate analysis found that RT schedule and baseline Koos group predicted for facial nerve toxicity. These data, as well as ours are summarized in Table 3. Vestibular dysfunction is another symptom that merits attention, considering a recent study demonstrating that vertigo negatively impacts quality of life more so than hearing loss in patients with unilateral vestibular schwannomas [10]. These issues are less commonly addressed in the literature, perhaps due to the fact that most published series are retrospective in nature, and formal assessment of vestibular function and quality of life are infrequently performed in routine clinical care. In our experience, patients’ subjective reports of vestibular dysfunction (i.e. dizziness, ataxia, vertigo) tend to remain stable (90 %) or improve (7 %) after SRS or FSRT, and rarely worsen significantly (3 %). Another issue to consider, perhaps more from a societal perspective than from an individual patient perspective, is the overall cost of treatment. Banerjee et al. [11] recently analyzed the initial and follow-up costs over at least 36 months for 53 patients treated with microsurgery or SRS. Although the initial cost of surgical resection was higher, with a mean cost of $23,788 versus $16,143 for
123
192
J Neurooncol (2014) 116:187–193
Table 3 Single-institution comparisons of outcomes treating vestibular schwannoma with SRS, FSRT, or surgery Series
Treatment (no. pts)
Serviceable hearing preservation
CN VII toxicity
CN V toxicity
SRS (63) Surgery (28)
68 %
1.6 %
NR
0 %, p \ 0.001
46 %, p \ 0.001
Surgery vs. SRS Myrseth et al. [18] Pollock et al. [19]
SRS (46)
63 %
2%
Surgery (36)
5 %, p \ 0.001
17 %, p = 0.04
NR
Re´gis et al. [20]
SRS (100)
40 %
0%
4%
Karpinos et al. [5]
Surgery (110) SRS (75)
5 %, p = 0.000001 44 %
47 %, p = 0.00005 35.3 %
29 %, p = 0.0009 22 %
Surgery (25)
40 %
6.1 %, p = 0.008
12.2 %, p = 0.009
SRS (48)
60.0 %
2.1 %
10.5 %
ConFSRT (19)
44.4 %
0%
5.3 %
HypoFSRT (37)
63.2 %
0%
25.0 %
SRS (78)
59 %
16 %
6%
ConFSRT (10) or HypoFSRT (32)
82 %
3%
3%
SRS (30)
NR
SRS vs. FSRT Anderson et al. current series
Collen et al. [9] Combs et al. [8]
ConFSRT (172) Meijer et al. [7] Andrews et al. [6]
17 %
7%
2%
3%
SRS (49)
75 %
7%
8%
HypoFSRT (80)
61 %
3%
2 %, p = 0.048
SRS (69)
33 %
2%
5%
ConFSRT (56)
81 %, p = 0.0228
2%
7%
p-values included for statistically significant differences SRS stereotactic radiosurgery, ConFSRT conventionally fractionated stereotactic radiation therapy, HypoFSRT hypofractionated FSRT, NR not reported (serviceable hearing preservation rates not compatible with table format)
SRS, follow-up costs decreased over time after surgery, and increased after SRS. The authors concluded that SRS is likely to be less expensive than surgery if long-term local control rates remain high at the lower doses currently prescribed, and the future frequency of follow-up MR imaging is decreased. Such a comparison is currently lacking between SRS and FSRT. What conclusions can be made, then, regarding the ideal treatment for vestibular schwannoma? First and foremost, considering the myriad of hypothesis-generating data that have emerged in recent years, it is clear that patients are best served by a multidisciplinary approach to individualize treatment recommendations for each patient on the basis of current knowledge combined with tumor characteristics, co-morbidities and patient-specific concerns. Our patient cohort was evaluated in multidisciplinary fashion, and includes only patients treated with a nonsurgical approach. Maximum tumor dimension greater than 3 cm is commonly considered to be a contraindication for SRS; indeed, our patient series included only four tumors greater than 3 cm in maximum dimension, of which only two were treated with SRS in the 1990s. The ideal marginal dose for SRS is most likely 12–13 Gy, since local control appears to
123
be compromised with lower doses [12, 13], and cranial nerve toxicity increases with higher doses [8, 14–16]. For FSRT, the best dose and fractionation regimen remains unknown, as good outcomes are being reported with both conventional and hypofractionated approaches. The small number of patients treated with hypofractionated versus conventionally fractionated FSRT in our series prohibits detailed comparison of the outcomes between these two patient groups. A recent analysis of 89 patients treated with 1.8 Gy fractions did find better hearing preservation and equal local control in patients treated to 46.8 versus 50.4 Gy [17]. We have tended to select patients with larger tumors and potential brainstem compression for FSRT, and smaller, predominantly intracanalicular tumors for SRS. In conclusion, vestibular schwannoma can be treated with SRS, conventionally fractionated FSRT and hypofractionated FSRT, resulting in high local control and low toxicity rates. For larger tumors, FSRT offers equivalent 5-year PFR compared with single fraction SRS. In order to truly ascertain a significant difference between these techniques, a formal randomized trial would be required, but given the relatively small differences in outcome, such a trial might require a very large number of patients. A
J Neurooncol (2014) 116:187–193
193
cost-effectiveness analysis of the various radiotherapy approaches would also be useful. Conflict of interest of interest.
The authors declare that they have no conflict
10.
11.
References 1. Plotkin SR, Stemmer-Rachamimov AO, Barker FG, Halpin C, Padera TP, Tyrrell A, Sorensen AG, Jain RK, di Tomaso E (2009) Hearing improvement after bevacizumab in patients with neurofibromatosis type 2. N Engl J Med 361:358–367 2. Tome´ WA, Meeks SL, Buatti JM, Bova FJ, Friedman WA, Li Z (2000) A high-precision system for conformal intracranial radiotherapy. Int J Radiat Oncol Biol Phys 47:1137–1143 3. Tome´ WA, Meeks SL, McNutt TR, Buatti JM, Bova FJ, Friedman WA, Mehta M (2001) Optically guided intensity modulated radiotherapy. Radiother Oncol 61:33–44 4. Pollock BE (2006) Management of vestibular schwannomas that enlarge after stereotactic radiosurgery: treatment recommendations based on a 15 year experience. Neurosurg 58:241–248 5. Karpinos M, Teh BS, Zeck O, Carpenter LS, Phan C, Mai WY, Lu HH, Chiu JK, Butler EB, Gormley WB, Woo SY (2002) Treatment of acoustic neuroma: stereotactic radiosurgery vs. microsurgery. Int J Radiat Oncol Biol Phys 54:1410–1421 6. Andrews DW, Suarez O, Goldman HW, Downes MB, Bednarz G, Corn BW, Werner-Wasik M, Rosenstock J, Curran WJ (2001) Stereotactic radiosurgery and fractionated stereotactic radiotherapy for the treatment of acoustic schannomas: comparative observations of 125 patients treated at one institution. Int J Radiat Oncol Biol Phys 50:1265–1278 7. Meijer OW, Vandertop WP, Baayen JC, Slotman BJ (2003) Single-fraction vs. fractionated linac-based stereotactic radiosurgery for vestibular schwannoma: a single-institution study. Int J Radiat Oncol Biol Phys 56:1390–1396 8. Combs SE, Welzel T, Schulz-Ertner D, Huber PE, Debus J (2010) Differences in clinical results after linac-based single-dose radiosurgery versus fractionated stereotactic radiotherapy for patients with vestibular schwannoma. Int J Radiat Oncol Biol Phys 76:193–200 9. Collen C, Ampe B, Gevaert T, Moens M, Linthout N, De Ridder M, Verellen D, D’Haens J, Storme G (2011) Single fraction
12.
13.
14.
15.
16.
17.
18.
19.
20.
versus fractionated linac-based stereotactic radiotherapy for vestibular schwannoma: a single-institution experience. Int J Radiat Oncol Biol Phys 81:e503–e509 Myrseth E, Moller P, Wentzel-Larsen T, Goplen F, Lund-Johansen M (2006) Vestibular schwannoma: surgery or gamma knife radiosurgery? A prospective, nonrandomized study. Neurosurg 59:67–76 Banerjee R, Moriarty JP, Foote RL, Pollock BE (2008) Comparison of the surgical and follow-up costs associated with microsurgical resection and stereotactic radiosurgery for vestibular schwannoma. J Neurosurg 108:1220–1224 Foote KD, Friedman WA, Buatti JM, Meeks SL, Bova FJ, Kubilis PS (2001) Analysis of risk factors associated with radiosurgery for vestibular schwannoma. J Neurosurg 95:440–449 Chung WY, Liu KD, Shiau CY, Wu HM, Wang LW, Guo WY, Ho DM, Pan DH (2005) Gamma knife surgery for vestibular schwannoma: 10-year experience of 195 cases. J Neurosurg 102(Suppl):87–96 Flickinger JC, Kondziolka D, Niranjan A, Lunsford LD (2001) Results of acoustic neuroma radiosurgery: an analysis of 5 years’ experience using current methods. J Neurosurg 94:1–6 Flickinger JC, Kondziolka D, Pollock BE, Lunsford LD (1996) Evolution in technique for vestibular schwannoma radiosurgery and effect on outcome. Int J Radiat Oncol Biol Phys 36:275–280 Chihara Y, Ito K, Sugasawa K, Shin M (2007) Neurological complications after acoustic neurinoma radiosurgery: revised risk factors based on long-term follow-up. Acta Otolaryngol Suppl 559:65–70 Andrews DW, Werner-Wasik M, Den RB, Paek SH, DownesPhillips B, Willcox TO, Bednarz G, Maltenfort M, Evans JJ, Curran WJ (2009) Toward dose optimization for fractionated stereotactic radiotherapy for acoustic neuromas: comparison of two dose cohorts. Int J Radiat Oncol Biol Phys 74:419–426 Myrseth E, Moller P, Pedersen PH, Lund-Johansen M (2009) Vestibular schwannoma: surgery or gamma knife radiosurgery? A prospective, nonrandomized study. Neurosurg 64:654–661 Pollock BE, Driscoll CL, Foote RL, Link MJ, Gorman DA, Bauch CD, Mandrekar JN, Krecke KN, Johnson CH (2006) Patient outcomes after vestibular schwannoma management: a prospective comparison of microsurgical resection and stereotactic radiosurgery. Neurosurg 59:77–85 Regis J, Pellet W, Delsanti C, Dufour H, Roche PH, Thomassin JM, Zanaret M, Peragut JC (2002) Functional outcome after gamma knife surgery or microsurgery for vestibular schwannomas. J Neurosurg 97:1091–1100
123
