Long-term Growth Rate of Vestibular Schwannoma in Neurofibromatosis 2: A Volumetric Consideration

Transcription

1 The Laryngoscope VC 2016 The American Laryngological, Rhinological and Otological Society, Inc. Long-term Growth Rate of Vestibular Schwannoma in Neurofibromatosis 2: A Volumetric Consideration Aurore Picry, MD; Nicolas-Xavier Bonne, MD; Juliette Ding, MD; Rabih Aboukais, MD; Jean-Paul Lejeune, MD, PhD; Marc Baroncini, MD; Frederique Dubrulle, MD, PhD; Christophe Vincent, MD, PhD Objectives/Hypothesis: To determine the long-term growth rate of vestibular schwannoma (VS) in neurofibromatosis type 2 (NF2) patients based on volumetric measurements. Study Design: Retrospective review. Methods: We retrospectively reviewed magnetic resonance imaging (MRI) scans acquired from patients with NF2 from 1999 to Patients with an annual radiologic follow-up over at least 3 consecutive years were included. Volumetric VS growth was prospectively measured using a three-dimensional imaging workstation and through manual contouring of the lesion. Time to tumor progression was assessed according to the Response Evaluation in Neurofibromatosis and Schwannomatosis Tumor Measurement Group. Results: Eighteen patients presenting with a total of 26 VSs were included. The mean age at diagnosis was 26.1 years (range, 7 to 49 years). One hundred five MRI scans were analyzed during a median radiological follow-up of 5.6 years (range, 3 to 12 years). The annual volume and diameter growth rates were respectively mm 3 /yr (range, 210 to 1,250 mm 3 / yr) and 0.9 mm/yr (range, 20.5 to 4.5 mm/yr). Time to tumor progression was 3 years (median survival). There was a weak correlation between volumetric and linear measurements (P <.0001, linear regression, n 5 26, r ). Among the 26 VSs, 76.9% (20/26) showed progression (VS growth more than 20%), 19.2% were stable, and 3.9% (1/26) exhibited shrinkage (spontaneous regression of more than 20% of the initial volume). Conclusions: This study helps to assess the long-term growth profile of VS in a population of NF2 patients with untreated VS. These data could help to better acknowledge VS natural growth history. Key Words: Neurofibromatosis type 2, vestibular schwannoma, growth rate, magnetic resonance imaging, volumetric. Level of Evidence: 4 Laryngoscope, 126: , 2016 INTRODUCTION Neurofibromatosis type 2 (NF2) is a rare autosomal dominant disorder characterized by bilateral vestibular schwannoma (VS), various brain and spinal tumors, peripheral nerve tumors, cutaneous tumors, and ocular abnormalities. 1 This condition is caused by mutations in a tumor suppressor gene located on chromosome 22 coding for a protein referred to as schwannomin or merlin. 2,3 Birth incidence is approximatively 1 in 33,000 and prevalence is 1 in 56, Hearing loss is the main complaint, with a risk of complete deafness. To date, there are four major treatment options available for NF2 VS: From the Department of Otology and Neurotology (A.P., N.-X.B., C.V.), Roger Salengro Hospital, Lille University Hospital, Lille, France; Department of Anatomy (A.P.), Henri Warembourg Faculty of Medicine, University of Lille, Lille, France; Department of Neurosurgery (R.A., J.-P.L., M.B.), Roger Salengro Hospital, and the Department of Radiology (F.D.), Claude Huriez Hospital, Lille University Hospital, Lille, France Editor s Note: This Manuscript was accepted for publication February 22, The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Aurore Picry, MD, Department of Otology and Neurotology, Roger Salengro Hospital, CHRU Lille, France. aurore.picry@chru-lille.fr DOI: /lary microsurgery, stereotactic radiation therapy, conservative management, and more recently, emerging use of targeted therapies. Accurate assessment of tumor growth and VS size is critical in the management of patients. In 1993, Sarazin et al. proposed a magnetic resonance imaging (MRI) semiautomated method for volume calculation to evaluate VS growth in nonoperated cases. 5 More recently, a few authors proposed a volume calculation based on linear measurements (anteroposterior, medial-lateral, super-inferior planes) on MRI slides according to Response Evaluation Criteria in Solid Tumors (RECIST criteria). 6 8 Today, it is commonly accepted that VS growth may be underestimated by RECIST criteria due to VS lesions being inherently irregular and oblong, and because of nonspherically symmetrical growth. Based this principle, Dombi et al. 9 recommended MRI with volumetric analysis to sensitively and reproducibly evaluate changes in tumor size during clinical trials. True volumetric measurement is an alternative to other proposed calculations, such as the two-box component used by Peyre et al. 10 or the ellipsoid model described by Dirks et al. 11 A 20% volume change is proposed to indicate a decrease or an increase in tumor size such as the time to tumor progression (TTP)

2 Fig. 1. Magnetic resonance imaging scans on the three-dimensional imaging workstation. (A) Manual contouring on the axial plane and (B) on the coronal plane. (C) The largest diameter was measured in the axial and coronal plane. Postcontrast T1-weighted spin echo sequence. [Color figure can be viewed in the online issue, which is available at The present study aimed to calculate the long-term growth rate of VSs in NF2 patients based on volumetric measurements. Ultimately, these results would provide a finer characterization of VS natural growth history in our population and could improve the evaluation criteria for future therapeutic trials. MATERIALS AND METHODS Population We included patients harboring NF2 according to Manchester criteria and/or genetic analysis. Criteria of eligibility for volumetric analysis were patients with untreated VSs and who underwent annual MRI follow-up during at least 3 consecutive years (we considered a minimum of four measurements per tumor) in the period 1999 to Our MRI protocol was made in the department of radiology. MRI examinations performed within another center were excluded. To record all VS measures to assess each patient on every inclusion criteria, we retrospectively reviewed all radiological records. Exclusion criteria were VSs with volumetric measures 25 mm 3 or tumors identified on less than two slides, as their precise volumetric assessment is not reliable. All treated VSs were excluded: concomitant chemotherapy (bevacizumab [Avastin]), cochlear implant, or auditory brainstem implant recipients, internal auditory canal decompression and surgical management of VS, contralateral or ipsilateral recurrences, or another tumor during follow-up. We stratified the population in two groups, younger and older than 18 years. MRI Protocol MRI examinations were obtained using a standard head coil on a 1.5T MR Philips Ingenia (Philips, Amsterdam, the Netherlands). The protocol consisted in T1-weighted spin echo sequences (slice of 2 mm), before and after gadolinium injection in axial and coronal plane, and a fat-saturation threedimensional (3D) gradient-echo sequence was added for intracanicular tumor (slice of 1 mm). The postcontrast T1-weighted spin echo sequences were analyzed to perform linear and volumetric measurements. Each sequence was performed in axial and coronal imaging planes covering the internal auditory canal. Linear measurements were performed on axial and coronal planes with the largest VS diameter (Fig. 1C) by an experienced senior radiologist who could not see the patient. Coronal and axial measures were averaged for linear assessment. For volumetric analyses, MRI scans were loaded to a 3D imaging workstation (IntelliSpace Portal; Philips). The same radiologist manually contoured the lesion on each picture identifying the VS. This way, a lesional area was determined for each slice (Fig. 1A,B). The surfaces of each section were added to create a 3D volume image. The semiautomated software calculated the VS volume by multiplying the outlined areas by the slice thickness. As linear assessment, coronal and axial measures were averaged for volume. No automated procedure has been used for volumetric assessment. Average 2359

3 TABLE I. Clinical and Radiological Characteristics of 18 Patients Harboring 26 Vestibular Schwannoma. Patient Sex, Female/Male Age at Diagnosis, yr Family History, Y/N Mutation, C/M Diameter Baseline, mm, R/L Volume Baseline, mm 3, R/L Growth Rate, mm 3 /yr, R/L 1 Female 7 N Frame shift (C) 7/10 50/ / Male 13 N N/A X/9 X/80 X/88 3 Female 14 N Splice site (C) 18/X 400/X 140/X 4 Female 14 Y Frame shift (C) 12/X 300/X 1,250/X 5 Male 17 N Nonsense (C) 13/14 280/ /405 6 Male 17 Y Splice site (C) 9.5/X 240/X 28.33/X 7 Male 19 Y Splice site (C) 16/X 220/X 297.5/X 8 Female 20 N Frame shift (C) 22/9 1,030/ / Female 24 Y N/A (SNP) 8/12 99/ / Male 27 N N/A X/7 X/155 X/ Female 27 N Frame shift (C) 19/20 600/ / Male 28 N Frame shift (M) 23/26 1,600/4, / Male 30 Y Splice site (C) 13/23 420/3, / Female 34 Y N/A X/14.2 X/460 X/ Male 40 Y Nonsense (M) 13.5/X 390/X 172/X 16 Female 41 Y Splice site (C) X/15 X/147 X/ Male 49 N N/A 11/X 275/X /X 18 Male 49 Y Splice site (C) 28/20 2,800/ /25 C 5 constitutional; L 5 left; M 5 mosaic; N 5 no; N/A 5 not applicable; R 5 right; SNP 5 single nucleotid polymorphism; X - no VS Y 5 yes. growth rates were studied by age group (younger and older than 18 years). Genetic Analysis All patients underwent genetic analysis by sequencing of the 17 NF2 coding exons in circulating blood leukocytes and/or tumoral tissue. Correlation between VS growth rate profile and genotype was performed. Statistical Analysis Patient characteristics (age at diagnosis, sex, family history of NF2, genetic analysis) and VS measurements were downloaded within Microsoft Excel software (Microsoft Corp., Redmond, WA). Measurements from MRI scans were used for assessment of VS growth rate in the axial and coronal plane. Baseline VS size, VS growth, total percentage change between first and last scans, and growth rate per year were determined for volume and diameter measurements. VS growth was related to constitutional and mosaic mutations if possible. Correlation analyses for volumetric and linear growths were conducted with the Pearson test (using r 2 as the correlation coefficient, Graphpad Prism; GraphPad Software, Inc., La Jolla, CA). Nonparametric comparison of average growth rates by age group (younger or older than 20 years) was performed by Mann- Whitney test. TTP (median duration of follow-up corresponded to volume increase 20%) was calculated using the Kaplan- Meier technique. RESULTS Patients Eighteen NF2 patients were included with 26 VSs fulfilling study criteria. The characteristics of the patients and the VS growth are shown in Table I. There were 10 males and eight females. Average age at diagnosis was 26.1 years (range, 7 to 49 years). Nine subjects had a family history of NF2. There were eight subjects with bilateral schwannomas. One hundred five MRI scans from these patients, performed from 1999 to 2013, were available for analysis. Patients had a minimum of four and a maximum of 10 MRI scans during the study period. The mean duration of the radiologic follow-up was 5.6 years (range, 3 to 12 years). VS Growth The initial mean volume was mm 3, and the mean volume after 3 years of follow-up was 1, mm 3. The annual volume growth rate was mm 3 /yr (range, 210 to 1,250 mm 3 /yr) during the study period (Fig. 2A). For NF2 patients diagnosed before the age of 18 years, the average growth rate was mm 3 / year [210; 1,250] (n 5 13 tumors in nine individuals). In the group diagnosed after the age of 18 years, the average growth rate was mm 3 /year (28.33; ) (n 5 13 tumors in nine individuals). Meanwhile, there was no statistical difference between these two groups (Mann-Witney test, P >.05).The initial mean diameter was 15.1 mm (range, 8 to 28 mm). The annual diameter growth rate was 0.9 mm/yr (range, 20.5 to 4.5 mm/yr). Among the 26 VSs, 76.9% (20/26) showed progression (VS growth more than 20%), 19.2% were stable, and 3.9% (1/26) exhibited shrinkage (a spontaneous regression in at least 20% of the initial volume). There was a weak correlation coefficient between volumetric and linear measures (P <.0001, linear regression, n 5 26, r

4 Fig. 2. Vestibular schwannoma (VS) growth. (A) Volumetric growth rate (mm 3 /yr) for each of the 26 VSs. (B) Correlation analysis for volumetric and linear measurements by Pearson test (linear regression, n 5 26, r ). (C) Time to tumor progression is 3 years (median survival). 0.58) (Fig. 2B). Half of the tumors progressed to more than 20% in years (Fig. 2C). Genetic Analysis Mutation analysis was carried out for all patients. NF2 mutation was identified in 13 of the 18 patients (Table II). Five patients presented a frame-shift mutation, six splice-site, five were undetermined, two presented nonsense mutations, and there was no patient with a large deletion. The growth rate of VS in constitutional nonsense mutations was mm 3 /yr compared to 172 mm 3 /yr for mosaic nonsense mutations. The growth rate of VS in constitutional frame-shift mutations was mm 3 /yr compared to mm 3 /yr for mosaic frame-shift mutation. Volumetric comparison between constitutional and mosaic mutations could not be evaluated for frame-shift, splice-site, large-deletion, or undetermined mutations. DISCUSSION In the current study, we proposed to calculate the long-term VS growth rate in NF2 patients based on volumetric measurements. VS size was assessed from clinical MRI films on 18 subjects (26 VSs), and the mean duration of follow-up was 5.6 years (range, 3 to 12 years), with a minimum of 3 years. It is a large series of NF2 patients that aimed to assess natural growth history over an extended follow-up period based on volumetric analysis. The results of the present study show that VS progressed over mm 3 /yr on average, lower than in Harris et al. s study (700 mm 3 /yr). 12 However, even presenting with a slow growth rate, most of them showed progression (76.9% [20/26], tumoral growth more than 20% of the initial volume), 19.2% (5/26) were stable, and only 3.9% (1/ 26) exhibited shrinkage (spontaneous regression of more than 20% of the initial volume). This is probably related to the average age in the population (26.1 years; range, 7 to 49 years). Some authors found that VS growth rate tends to decrease with increasing patient age, reflecting more active NF2 disease in young patients. 6,7,9,13 In contrast, Slattery et al. found that VS growth rate did not change with increasing age, but rather thought of a possible selection bias, as faster growing VSs were more likely to be excised with increasing age. Moreover, excised VSs grew more rapidly prior to excision than nonexcised VSs, and there was a small proportion of young patients. 14 They suggested that the decrease in VS growth rate with increasing age in previous studies could be due to the small sample size. Meanwhile, the relationship was later confirmed by combining datasets obtained from Baser et al. and Mautner et al. studies (reaching n 5 55 patients), consistent with our results. 7,13 The natural growth rate of VS is highly variable from one study to another. A trend could be that VS growth rate decreases with increase in age. In our study, it seemed that this trend was verified with higher annual growth rate among younger patients (20 years). 15 This might be partly a result of the NF2 mutations spectrum (e.g., nonsense, frame-shift, large deletion, or splice-site mutations). The growth profile of VS is also linked to the type of inheritance of NF2 (i.e., germline or mosaic). In our study, all patients underwent genetic screening, and VS growth was assessed accordingly. We found that VS progressed for nonsense, frame-shift, splice-site, and undetermined mutations, but growth rate varied between mm 3 /yr (frameshift, mosaic) and mm 3 /yr (nonsense, TABLE II. Genetic Analysis. Mutation Nonsense (n 5 3) Frame Shift (n 5 9) Splice Site (n 5 8) Large Deletion (n 5 0) Undetermined (n 5 6) Constitutional (n 5 2) (n 5 7) (n 5 8) N/A (n 5 6) Mosaic 172 (n 5 1) (n 5 2) N/A N/A Vestibular schwannoma growth (mm 3 /yr) and genetic analysis by sequencing of the 17 NF2 coding exons on circulating blood leukocytes and /or tumoral tissue N/A 5 not applicable. 2361

5 constitutional). It seems that germline mutations, particularly nonsense mutations, have a more severe evolution with higher VS growth rate. However, the existence of different growth rates within affected family members with the same mutation emphasizes the fact that other major factors could be also involved (genetic and/or environmental). The strength of this work is based on the reliability of volumetric analysis as recommended by the Response Evaluation in Neurofibromatosis and Schwannomatosis (REiNS) Tumor Measurement Group. 9 Herein, volumetric measurements were obtained using postcontrast T1- weighted spin echo sequences (slice of 2 mm) and a fatsaturation 3D gradient-echo sequence was added for intracanalicular tumor (slice of 1 mm). 16 As a result, the relatively thin slice of this type of sequence enabled us to measure the smaller tumors. To our knowledge, this is the first study to report double volumetric assessment in axial and coronal plane using the average of these two measurements. A recent study by Harris et al. proposed a long-term follow-up of VS in NF2 patients. The authors suggested that VS growth was higher (700 mm 3 /yr) than in our study, but MRI protocol was different, with thicker cut (3 mm) and shorter radiologic follow-up (mean years, ranging from 1 to 5.75 years). 12 We were able to highlight the weak correlation between volumetric and linear measurement in VS growth. Due to the irregular shape of VSs, linear measurements can be highly variable, and long time periods are required to detect measurable changes, because VSs are slow-growing tumors. That being said, there is inevitably a recruitment bias in our population, given that tumors with the largest follow-up are those with the slowest growth and therefore do not require surgical management. 13 Finally, our volumetric analysis included the TTP as recommended for current response evaluation of benign tumors in clinical trials. Widemann et al. first set a phase II clinical trial using volumetric MRI analysis with TTP demonstrating that volumetric assessment was suitable to detect tumor progression (volume increase 20%) much earlier than linear measurements. 17 In the same study, Widemann et al. showed that TTP with volumetric analysis was more effective to detect tumor progression than the formerly described method based on World Health Organization (WHO) criteria. The median TTP was 14.3 months, compared to 52.2 months using WHO criteria; it could not be determined by RECIST criteria. 8,18 In our work, we found attp of 36 months (with MRI scans done 1 year apart). Having no point of comparison in the literature, this result could serve as a baseline for future clinical trials. CONCLUSION This study offers a volumetric characterization of the long-term natural growth history in NF2 patients. In our population, natural VS growth rate tends to progress over the long-term (median years; range, 3 to 12years), with an average growth rate of about mm 3 /yr (range, 210 to 1,250 mm 3 /yr), but it is highly variable depending on type of NF2 mutation (range, to mm 3 /yr). Our volumetric analysis included TTP (volume increase 20%) of 3 years, as recommended by the REiNS tumor measurement group for current response evaluation of benign tumors in clinical trials. These results could serve as the new baseline for further studies. Acknowledgments The help of Dr Frederique Dubrulle, Department of Radiology, Lille University Hospital, for technical support and study design is gratefully acknowledged. BIBLIOGRAPHY 1. Evans DG, Huson SM, Donnai D, et al. A genetic study of type 2 neurofibromatosis in the United Kingdom. II. Guidelines for genetic counselling. J Med Genet 1992;29: Rouleau GA, Merel P, Lutchman M, et al. Alteration in a new gene encoding a putative membrane-organizing protein causes neuro-fibromatosis type 2. Nature 1993;363: Trofatter JA, MacCollin MM, Rutter JL, et al. A novel moesin-, ezrin-, radixin-like gene is a candidate for the neurofibromatosis 2 tumor suppressor. Cell 1993;75: Evans DG, Howard E, Giblin C, et al. Birth incidence and prevalence of tumor-prone syndromes: estimates from a UK family genetic register service. Am J Med Genet A 2010;152A: Sarazin L, Jolivet O, Doyon D. Computerized evaluation of tumor volume with MRI. Applications to the surveillance of acoustic neurinoma. J Radiol 1993;74: Baser ME, Makariou EV, Parry DM. Predictors of vestibular schwannoma growth in patients with neurofibromatosis Type 2. J Neurosurg 2002;96: Mautner V-F, Baser ME, Thakkar SD, Feigen UM, Friedman JM, Kluwe L. Vestibular schwannoma growth in patients with neurofibromatosis Type 2: a longitudinal study. J Neurosurg 2002;96: Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer Oxf Engl 2009;45: Dombi E, Ardern-Holmes SL, Babovic-Vuksanovic D, et al. Recommendations for imaging tumor response in neurofibromatosis clinical trials. Neurology 2013;81:S33 S Peyre M, Goutagny S, Bah A, et al. Conservative management of bilateral vestibular schwannomas in neurofibromatosis type 2 patients: hearing and tumor growth results. Neurosurgery 2013;72: Dirks MS, Butman JA, Kim HJ, et al. Long-term natural history of neurofibromatosis type 2-associated intracranial tumors. J Neurosurg 2012; 117: Harris GJ, Plotkin SRMD, MacCollin M, et al. Three-dimensional volumetrics for tracking vestibular schwannoma growth in neurofibromatosis type II. Neurosurgery 2008;62: Baser ME, Mautner V-F, Parry DM, Evans DGR. Methodological issues in longitudinal studies: vestibular schwannoma growth rates in neurofibromatosis 2. J Med Genet 2005;42: Slattery WH, Fisher L, Iqbal Z, Oppenhiemer M. Vestibular schwannoma growth rates in neurofibromatosis type 2 natural history consortium subjects. Otol Neurotol 2004;25: Parry DM, Eldridge R, Kaiser-Kupfer MI, Bouzas EA, Pikus A, Patronas N. Neurofibromatosis 2 (NF2): clinical characteristics of 63 affected individuals and clinical evidence for heterogeneity. Am J Med Genet 1994; 52: Dubrulle F, Delomez J, Kiaei A, et al. Mass screening for retrocochlear disorders: low-field-strength (0.2-T) versus high-field-strength (1.5-T) MR imaging. AJNR Am J Neuroradiol 2002;23: Widemann BC, Dombi E, Gillespie A, et al. Phase 2 randomized, flexible crossover, double-blinded, placebo-controlled trial of the farnesyltransferase inhibitor tipifarnib in children and young adults with neurofibromatosis type 1 and progressive plexiform neurofibromas. Neuro Oncol 2014;16: Miller AB, Hoogstraten B, Staquet M, Winkler A. Reporting results of cancer treatment. Cancer 1981;47: