Bevacizumab in neurofibromatosis type 2 (NF2) related vestibular schwannomas: a nationally coordinated approach to delivery and prospective evaluation

Abstract

Background

NF2 patients develop multiple nervous system tumors including bilateral vestibular schwannomas (VS). The tumors and their surgical treatment are associated with deafness, neurological disability, and mortality.

Medical treatment with bevacizumab has been reported to reduce VS growth and to improve hearing. In addition to evaluating these effects, this study also aimed to determine other important consequences of treatment including patient-reported quality of life and the impact of treatment on surgical VS rates.

Methods

Patients treated with bevacizumab underwent serial prospective MRI, audiology, clinical, CTCAE-4.0 adverse events, and NFTI-QOL quality-of-life assessments. Tumor volumetrics were classified according to the REiNs criteria and annual VS surgical rates reviewed.

Results

Sixty-one patients (59% male), median age 25 years (range, 10–57), were reviewed. Median follow-up was 23 months (range, 3–53). Partial volumetric tumor response (all tumors) was seen in 39% and 51% had stabilization of previously growing tumors. Age and pretreatment growth rate were predictors of response. Hearing was maintained or improved in 86% of assessable patients. Mean NFTI-QOL scores improved from 12.0 to 10.7 (P < .05). Hypertension was observed in 30% and proteinuria in 16%. Twelve treatment breaks occurred due to adverse events. The rates of VS surgery decreased after the introduction of bevacizumab.

Conclusion

Treatment with bevacizumab in this large, UK-wide cohort decreased VS growth rates and improved hearing and quality of life. The potential risk of surgical iatrogenic damage was also reduced due to an associated reduction in VS surgical rates. Ongoing follow-up of this cohort will determine the long-term benefits and risks of bevacizumab treatment.

Neurofibromatosis type 2 (NF2) is an autosomal dominant tumor predisposition condition with a birth incidence of 1 in 25–33 000.¹ Bilateral vestibular schwannomas (VS) are emblematic of NF2 but many patients develop schwannomas of other cranial, spinal, and peripheral nerves alongside multiple meningiomas and spinal ependymomas.² Patients are often young at presentation. While the majority of tumors are not malignant, management options are limited and the need for multiple interventions contributes to significant disability and a reduced life expectancy.^2,3

The natural history of NF2-associated schwannomas is of tumor growth over time.^4–7 A single group has reported spontaneous decreases in tumor size in 6% of patients in a study of small, early-stage tumors.⁶

In 2009, 6 of 10 reported NF2 patients treated with bevacizumab for VS had hearing improvement or greater than 20% tumor volume reduction.⁸ Subsequently, over 50% of 31 patients (including the 10 original) demonstrated radiographic and hearing responses at 1 and 3 years.⁹ Other groups have reported similar positive results in smaller (n < 15 patients) independent patient series.^10–14 In contrast, 7 pediatric patients in a retrospective series did not meet the criteria for a radiographic response.¹⁵

In England approximately 95% of NF2 patients are managed by centrally commissioned, specialist multidisciplinary teams working from 4 National Health Service (NHS) lead centers. Since 2010, the service has funded bevacizumab treatment for rapidly growing schwannomas, providing an alternative to surgery and stereotactic radiosurgery, both of which have significant associated morbidities. Patients who met defined criteria for treatment were monitored prospectively with recording of defined outcomes. We report here our national cohort results, constituting the largest prospective series of patients reported to date.

Methods

Treatment Protocol

Eligibility for treatment was defined by schwannoma growth of ≥4 mm by linear diameter or ≥60% by volume over the preceding averaged 12-month period, or an imminent threat to neurological function such as brainstem compression. Tumors meeting one or more of these criteria were defined as the index tumor. Exclusion criteria are described in Appendix 1.

Intravenous bevacizumab was given as a 5 mg/kg infusion every 2 weeks or a 7.5 mg/kg infusion every 3 weeks for at least 6 months or until there was no evidence of further tumor shrinkage on successive three-month-interval scans. For those with stable disease on imaging, the dosing schedule was then reduced to a maintenance regimen of 2.5 to 5 mg/kg every 4 weeks. Hematological and renal indices were monitored prior to each dose.

A break in treatment was defined as at least one missed treatment dose (for example, if given infusions every 3 weeks, a patient would have a gap between doses of 6 or more weeks).

Treatment Outcomes

The following outcomes were recorded at baseline and at 3- to 6-month intervals thereafter, except toxicity, which was continuously monitored.

Volumetric VS Measurements

Measurements were made using Brainlab iPlan (Feldkirchen, Germany), a semiautomated software package, by a single operator under the supervision of a neuroradiologist. The coefficient of variation for this technique was 3.78% (unpublished data). Volumetric assessments were performed on the same MRI sequence for each patient using either a gadolinium-enhanced, T1-weighted sequence with a slice thickness of ≤3 mm (for n = 49 patients) or a ≤1 mm, high-resolution, heavily T2-weighted sequence (for n = 3 patients). Where there was a contralateral (nonindex) VS, it was also measured. To estimate pretreatment growth rate, at least one additional measurement was made in the 18-month period prior to treatment.

Radiological responses were defined according to the REiNS criteria¹⁶: partial response was a 20% or greater reduction in tumor volume, tumor progression as 20% or greater growth in volume, and stable disease as growth or reduction in tumor size of less than 20% compared with baseline.

Images were excluded from volumetric assessment if: (i) artifact from an auditory brainstem or cochlear implant prevented accurate volumetric assessment; (ii) imaging was incompatible with consistent serial volumetric measurement; or (iii) tumor volume was less than 1 cm³.

Audiometry

Maximum speech discrimination scores (SDS) were recorded using Arthur Boothroyd word list phoneme testing at multiple levels to determine the maximum SDS at each time point. SDSs were analyzed for a statistically significant change relative to baseline using binomial confidence intervals for Arthur Boothroyd words corresponding with the number of words on which each score was based.¹⁷

Quality of Life

NFTI-QOL (a disease specific quality-of-life measure)^18,19 assessment was used to assess patient-reported quality of life. The mean total NFTI-QOL scores in patients treated with bevacizumab were compared with independent scores from 117 previously reported NF2 patients attending the 4 specialist centers on 2 or more visits.¹⁹ This historical control population had a median age of 39 years (range, 10–87) and was 49% male and 51% female.

Clinical Status

Neurological symptoms and examination were recorded at each clinical review.

Toxicity Monitoring

Toxicity was recorded prior to each dose (2–4 weekly) of bevacizumab and graded according to the Common Toxicity Criteria for Adverse Events (CTCAE) version 4.03.²⁰

Genetic Severity

Genetic severity was rated as severe, moderate, or mild according to a previous published classification.¹⁹ In light of national cohort data, patients with exon 1 truncating mutations were classified as having mild rather than severe disease.²¹

1. Severe: full-germline truncating mutations in exons 2–13.

2. Moderate: deletions not involving the promoter region or exon 1, splice site mutations (exons 1–8), and mosaicism of truncating mutations in exons 1–13 in blood.

3. Mild: Truncating mutations in exon 1, missense mutations or in-frame deletions, large deletions involving the promoter region/exon 1, splice-site mutations (exons 9–15), mosaicism (excluding moderate criteria), and no mutation identified on blood analysis.

Statistical Analysis

All statistical analyses were carried out in SPSS version-20. Missing values were interpolated between adjacent observations but not extrapolated. This only applied to 10% values and there were no significant differences between analyses with or without these values. The Student t test was used to compare absolute and relative changes in tumor volume and changes in the total NFTI-QOL score, at 12, 18, and 24 months compared with baseline demographics. ANOVAs were used to examine changes in tumor volume and NFTI-QOL over time. Bivariate (Pearson's) correlations were used to examine associations between mean changes in tumor volume, changes in NFTI-QOL mean total score, and other variables. In addition to bivariate correlations, multiple linear regression was used to explore factors predictive of changes in tumor volume.

Change in Clinical Practice

The annual rates of surgical treatment of VS in NF2 patients in England were compared for the 5-year periods prior (2004–2009) and subsequent (2010–2014) to the introduction of treatment with bevacizumab in 2010.

The study was approved by the Westminster Ethics Committee as a multiple site study and informed consent was obtained.

Results

Cohort Characteristics

At the time of analysis, 744 patients were managed by the specialist NF2 service in England. Of these, 61 commenced bevacizumab between September 2010 and December 2013 (Table 1). Data on all were prospectively collected and are reported here. Median age at treatment was 25 years (range, 10–57). The majority of index tumors were VS (n = 59 tumors in 52 patients). Nine patients with index tumors other than VS are not reported here. All patients were given intravenous bevacizumab at a dose of 5 mg/kg every 2 weeks (n = 48) or 7.5 mg/kg every 3 weeks (n = 14) according to patient preference. No patients had been previously exposed to chemotherapy. No tumors were irradiated in the 6 months prior to treatment. Three patients had previously received SRS to 5 vestibular schwannomas, 11 to 219 months (median 201 months) prior to treatment.

Table 1.

Cohort characteristics at baseline

Demographics
Patients (n)	61 (36 men)
Age (years)	25 (range, 10–57)
Index tumor volume (cc)	7.26 (range, 1.04–34.31)
Geneticsa
Truncating mutation (n)	33 (54%)
Deletion (n)	13 (22%)
Other mutation (n)	11 (18%)
Not identified (n)	3 (5%)
Genetic severitya
Severe (n)	31 (52%)
Moderate (n)	18 (31%)
Mild (n)	11 (18%)
Index tumor
Unilateral VS	45 (74%)
Bilateral VS	7 (11%)
Lower CN schwannoma	2 (3%)
Extracranial schwannoma	7 (11%)
Hearing status at start of treatment
Unilateral hearing (n)	22 (36%)
Bilateral deafness (n)	20 (33%)
Bilateral hearing (n)	19 (31%)

Abbreviations: VS, Vestibular schwannoma; CN, Cranial nerve.

^aOne patient declined genetic testing.

Treatment Outcomes

Imaging

Twelve index VS tumors from 10 patients were excluded from volumetric analysis due to their size being less than 1 cm³ (n = 5 patients) and scan artifact (n = 5 patients). Following these exclusions, 47 index VS tumors and 14 nonindex VS tumors from 42 and 14 patients, respectively, were suitable for radiographic assessment. Median radiological follow-up was 23 months (range, 3–56 months).

Pretreatment Growth Rate

The median annual growth rate prior to treatment in 38 eligible index and 9 nonindex VS tumors was 57% (range, 12%–179%) and 21% (range, 2%–114%), respectively. Younger patients' tumors grew more quickly before treatment than older patients' tumors, but the difference did not reach statistical significance (mean ± standard deviation = 55% ± 34% aged ≥18; 107% ± 45% aged <18; P = .29).

The Effect of Treatment With Bevacizumab on VS Tumor Volumes

In the index VS tumors, the best observed responses were 15 (32%) partial responses, 24 (51%) cases of stable disease, and 8 (17%) cases of progressive disease (Fig. 1A). There were no complete responses. Partial responses were maintained for a median of 15 months (range, 3–33 months), equivalent to the duration of follow-up on drug (Fig. 1B). Radiological progression occurred at a median of 12.5 months (range 3–28) and while on treatment in 11 of 13 patients. Dose escalation to 10 mg/kg was attempted in 2 patients with progressive disease but was unsuccessful.

Fig. 1.

Volumetric tumor responses. (A) Waterfall plot of the maximum change in vestibular schwannoma (VS) tumor volume during treatment with bevacizumab. Index and nonindex tumors are shown with black and grey bars, respectively. (B) The percentage change in tumor volume relative to baseline (y-axis) for individual patients over time in months (x-axis) from start of treatment, including any periods off treatment. Index VS are shown with solid lines and nonindex VS with dotted lines.

Despite failing to meet standard criteria for treatment, responses were also seen in nonindex tumors; there were 9 (64%) partial responses, 4 (29%) cases of stable disease, and 1 (7%) case of progressive disease.

All tumors that attained partial response did so within the first 6 months of treatment, most within the first 3 months (Fig 1B).

The portion of index tumors with partial response or stable disease at 12, 24, and 36 months was 35 of 39 (90%), 12 of 19 (63%), and 5 of 8 (63%). (Fig. 2: Progression-free survival using the Kaplan-Meier method.)

Fig. 2.

Progression-free survival of index tumors using the Kaplan-Meier method (n = 47 at time 0 and n = 6 at 36 months).

Predictors of Radiological Response

Age and pretreatment growth rate both predicted imaging response (Pearson coefficient −0.36 and 0.36 respectively, P < .05) but were intercorrelated; younger patients had a tendency to faster growing tumors and poorer responses (Table 2). Multiple linear regression modeling suggested age to be the stronger factor. When the population of patients 18 years of age and over (n = 39) was analyzed separately, no statistically significant predictors of response were identified. Neither gender, genetic severity, nor toxicity indices (hypertension and/or proteinuria ≥ grade 1) correlated with imaging response.

Table 2.

Pearson correlation of descriptives vs maximum percent change and percent change at 12 months (both relative to baseline)

	Maximum Change	Change at 12 Months
Age	−0.325*	−0.355*
Sex	−0.017	0.082
Pretreatment growth rate	0.333*	0.361*
Genetic severity	−0.035	−0.208
Hypertension	−0.259	−0.274
Proteinuria	0.103	−0.018

Two-tailed P value significant to .05 marked with *.

Differences Between Pediatric and Adult Patients

Six patients with 8 VS tumors eligible for radiological response assessment were under 18 years of age at the start of treatment. The median tumor volume prior to treatment was 5.0 cm³ (range, 1.6–8.9) for children under 18 and 8.1 cm³ (range, 1.0–32.3) for those 18 years of age or older (P = .2). Only 1 of the 15 partial responses in the cohort occurred in the younger age group. There was a tendency for patients under 18 years to have ongoing tumor growth despite treatment (Fig. 3).

Fig. 3.

Estimated marginal means of tumor volume over time (baseline to month 24) for patients by age. VS in pediatric patients (under 18 years) have a tendency to grow and those in adult patients (18 years and over) show decrease then stabilization. Bars are SEM.

The Effect of Breaks in Treatment on Tumor Volume

There were 36 breaks in treatment in 28 patients, of median duration 3.5 months (range, 1–10 months). Reasons for treatment breaks are detailed in Table 3. Five patients did not restart treatment and have been followed for 5 to 14 months. We divided treatment breaks into those lasting less than 3 months (‘short’, n = 15) and 3 months or longer (‘long’, n = 16). No changes in the imaging response categories resulted from short breaks in treatment. Long treatment breaks were associated with tumor growth (median +12%; range, −11% to +79%). Of note, 2 patients with partial responses prior to the treatment break did not regain their previous partial responses, despite re-treatment.

Table 3.

Reasons for breaks in treatment

Reason for Break	Number of Events
Surgical or dental procedure	14
Adverse events	12
Planned treatment holiday in a stable patient	4
Personal reasons	4
Accidental injury	2

Three patients in the cohort had previously been treated with stereotactic radiation. Two (treated with SRS 11 and >200 months prior to bevacizumab, respectively) had ongoing tumor stability or shrinkage despite long treatment breaks. The third patient, who had previously received stereotactic radiation (>200 months prior), did not have breaks in treatment and had stable disease throughout treatment.

Hearing Outcome

Of the 61 patients, 52 had sufficient data to assess hearing over time (median follow-up 13.5 months; range, 3–42 months). Nineteen (32%) were bilaterally deaf at the start of treatment and remained deaf. Nine patients had insufficient data for longitudinal assessment. We report here the hearing results from 48 hearing ears in the remaining 33 patients. (Fig. 4)

Fig. 4.

Flow diagram of hearing assessment and outcomes (*One of the patients who had initial improvement had a subsequent decline in hearing to baseline while on treatment).

Of 11 patients with hearing impairment at the start of treatment, 5 (45%) had a statistically significant improvement in their SDS and 3 had nonsignificant improvements. Of 33 patients eligible for assessment of hearing loss, only 2 (6%) had a decline in hearing on treatment. Two had statistically significant hearing losses after initial improvement, both after 24 months on treatment. Two patients had nonsustained decreases in SDS at 6 months of treatment, but were stable relative to baseline at 12 months after commencement of treatment.

Within the 48 hearing ears, 29 hearing ears had 12-month pretreatment SDSs. Three (10%) of the 29 demonstrated significant hearing deterioration over the 12-month pretreatment period. Of 3 ears with significant deteriorations in hearing prior to treatment, 1 ear had a sustained improvement for 24 months on treatment. The remaining 2 ears had temporary improvements for up to 18 months, followed by deterioration. Of those 26 ears that had stable pretreatment hearing, 1 had a nonsustained, significant hearing deterioration at 6 months, otherwise none had significant deterioration at 6 months (25 ears), 12 months (21 ears), 18 months (17 ears), and 24 months (13 ears). Two ears recorded significant hearing improvement, which was maintained to 12 and 24 months respectively.

Quality of Life

The mean total NFTI-QOL score improved significantly from baseline at both 3 months (10.1 vs 11.5 at baseline, P < .01, n = 37) and 6 months (10.7 vs 12.0 at baseline, P < .05, n = 34), and then remained stable to 12 months. The mean total NFTI-QOL score for 117 NF2 historical controls did not change significantly at an average of 8 months between visits (9.3/9.4 at first/second visits). Pearson's correlations for change in volume compared with change in NFTI-QOL over 12 months (n = 30) and beyond were not significant.

Adverse Events

All 61 patients were eligible for adverse event evaluation, of whom 50 experienced at least 1 event. Treatment was generally well tolerated; grade 3 or grade 4 toxicities were experienced in 8 patients (Table 4). There were no deaths attributed to bevacizumab. The most commonly reported side effects were fatigue, menorrhagia, and hypertension. Treatment was interrupted for toxicity in 10 patients due to hypertension (n = 4), bleeding (n = 2), wound healing (n = 2), fatigue and anemia (n = 1), and infection (n = 1).

Table 4.

Adverse events graded by CTCAE 4.03

Adverse Event	Overall		Grade 3/4
	N	%	N	%
Proteinuria	10	16%	1	2%
Hypertension	18	30%	1	2%
Epistaxis	6	10%
Menorrhagia	7	28%
Irregular menstruation	2	8%
Fatigue	21	34%
Mucositis	11	18%
Nausea	12	20%
Psychosis	1	2%	1	2%
Anemia	1	2%	1	2%
Depression	2	3%	1	2%
Infections	11	18%	2	3%
Raised alanine aminotransferase	1	2%	1	2%
Delayed wound healing	5	8%
Urinary tract infection	4	7%
Vascular access	2	3%
Diarrhea	6	10%
Rash	6	10%
Constipation	3	5%
Palpitations	2	3%

The percentage figure (%) is calculated from the total 61-person cohort with the exception of menorrhagia and irregular menses where percentage is calculated for the 25 female patients.

Vestibular Schwannoma Resection Following Bevacizumab Treatment

Three patients required resection of a growing vestibular schwannoma after a period of treatment with bevacizumab. In all cases these were planned procedures and bevacizumab was withheld for at least 4 weeks prior to surgery and no change to usual perioperative care of NF2 patients was required. In 2 patients, an auditory brainstem implant was inserted at the same procedure. The histopathology of the resected VS tissue was schwannoma grade 1 in all cases.

Surgical Rates

On average, 22 VS resections were performed per year in England in the 5 years prior to introduction of bevacizumab, and an average of 18 per year in the 5 years subsequent to introduction of bevacizumab, with a further reduction to 13 per year (a 40% reduction) in the most recent 2 years as bevacizumab use has become more widespread.

Discussion

We report here the largest series to date of NF2 patients treated with bevacizumab for growing vestibular schwannomas. The 39% radiological response rate in our series for index and nonindex tumors combined is lower than the partial response rate of 55% that the next largest series reported.⁹ Differences in the pretreatment growth characteristics of the population of tumors, in the eligibility for treatment, and in the frequency and completeness of follow-up may have contributed to the discrepancy in observed response rates between the 2 series. In addition the former series included a higher proportion of patients pretreated with radiotherapy (5 of 31).⁹ Only 3 of 61 of our patients had previously had radiation to treat a VS. It is possible that the effects of previous stereotactic radiosurgery may confound treatment outcome response in this patient subgroup.

Although the administration of bevacizumab was not undertaken as part of a clinical trial, the patients in our series were treated according to prospectively defined criteria and all treated patients were followed up closely, according to protocol, within the NHS. Since approximately 95% of affected patients are managed by the specialized NF2 service in the UK, our results are a nearly population-based estimate of response to bevacizumab in this setting. The pattern of response timing is consistent with descriptions of other NF2 patients with VS treated with bevacizumab.⁹ Patient selection was defined by protocol and as such, while there is a lack of nontreatment control group, this study has allowed the evaluation of treatment response in a real-world clinical environment.

Notwithstanding differences in response metrics between cohorts, our series and other published series all have similar proportions of patients who have achieved either imaging responses or stable disease on treatment. For many patients in the series, disease stabilization may be viewed as a clinically significant response to treatment since all tumors were growing rapidly prior to treatment. However, since salutatory growth has been described in some NF2-associated tumors, including VS,⁴ with periods of growth interspersed by periods of quiescence, some caution must be exercised when interpreting these results.

Age and pretreatment growth rate were intercorrelated predictors of response. While absolute predictors of response to bevacizumab remain elusive, our experience is that tumor behavior in the first 6 months of treatment is helpful in predicting future clinical response. All patients who obtained a response did so by 6 months, while patients who progressed within the first 6 months of treatment did not subsequently respond.

The radiographic response rates in our pediatric subgroup were lower than in our adult group. While this observation may be expected given that younger patients are more likely to have a more severe phenotype and higher tumor growth rates,^1,6,7 our findings are similar to those of a recent report from a small retrospective cohort of children treated with bevacizumab, in which there were no radiographic responses.¹⁵ This finding is important given the known toxicity profile of bevacizumab and the potential for long-lasting effects on renal function. Continued reporting of the schwannoma response rates and toxicity of bevacizumab in childhood NF2 will be essential to inform future treatment strategies for this patient group.

We are the first to show that treatment of VS with bevacizumab resulted in an improvement in patient-reported quality of life. The pattern we observed was of an improvement in overall score at 3 months that was then maintained over time while on treatment. However, individual patient data showed a more complex pattern: although hearing and imaging responses were also most striking within the first 3 months, improvements in patient-rated quality of life did not simply parallel radiological and hearing responses. However, the changes in NFTI-QOL suggest that it also reflects the benefits of regular contact with health care providers, patients' perspectives on their life role, and control of other symptoms such as pain. Nonetheless, our finding reinforces the vital importance of measuring patient-reported outcomes when evaluating novel treatments.

NF2 is associated with a high probability of progressive hearing loss, which is unpredictable and rarely recovers spontaneously. Profound or total deafness is unfortunately common although cochlear implantation and auditory brainstem implantation does offer some hearing rehabilitation in select patients. Maintaining hearing must therefore be the goal and early outcome data suggest that bevacizumab may be helpful. In NF2 patients, bilateral complete hearing loss unfortunately remains extremely common. While the impact on hearing in our population may not be indefinite, the maintenance or improvement of hearing at follow-up longer than 12 months (range, 12–42 months) in 86% of assessable patients is encouraging. Changes in hearing did not correlate well with imaging response in our study (data not reported), a trend that has also been reported in natural history studies.²² Our observation of an improvement in hearing seen in a small group of patients with recent hearing loss prior to commencement of treatment is consistent with the work of Blakely et al.¹³

In our cohort bevacizumab was well tolerated. The frequency of adverse events was similar to other, smaller series in this setting.^9,11 Fatigue was commonly experienced in our population although rarely required dose modification. Proteinuria was less frequently observed in our series than in some other reported series, and may be due to differences in dosing regimes. All of the patients we report were given the equivalent of 5 mg/kg of bevacizumab every 2 weeks for 6 months, after which the majority had reductions in treatment intensity, in contrast to some other reported and unreported series that involve doses of up to 10 mg/kg every 2 weeks. Toxicities in our NF2 series are much less pronounced than previous reports of bevacizumab chemotherapy for cancer.²³ This finding is not surprising since the relatively young NF2 cohort who are likely to benefit most from this treatment have generally not yet acquired significant comorbidities, are generally not taking multiple other medications, and the potential for drug interactions is therefore less. However, given the relatively lower response rates observed in young patients, and the risks of increasing toxicity with long-term use,²⁴ it is imperative that this treatment is formally evaluated in young patients treated for this indication, particularly in children, and that they continue to be followed for toxicity into the long term. This evaluation should ideally take place as part of a rigorously conducted clinical trial.

Changes in Healthcare Practices

The number of patients proceeding to surgical VS resections has dropped by 20% to 40% since the introduction of bevacizumab treatment for managing NF2 in the UK. In this young but morbid population, avoidance of iatrogenic damage is a vitally important endpoint. However while bevacizumab treatment responses remain durable in the majority of patients, we and other groups have seen breakthrough growth in some patients treated for prolonged periods. Furthermore, adverse events were the second highest cause of interruptions and discontinuation of treatment in our cohort, and long-term toxicity may limit the prolonged use of the drug.²⁴ It is uncertain whether this will lead to a rise in surgical procedure rates in future. It is reassuring that in the very small number of patients who have proceeded to surgery after treatment with bevacizumab, there did not appear to be an increased risk of perioperative difficulties.

Conclusion

Since the introduction of bevacizumab for NF2 patients with growing schwannomas in the UK, over 60 patients have been treated. The majority have had either stabilization or reduction in VS size and in those with recent hearing loss, hearing has improved. Patient-reported quality of life has had multifactorial improvement as measured by the NFTI-QOL and surgical intervention has decreased.

The benefits bevacizumab may offer to this group of young people with a high burden of disability will not be indefinite, and must be weighed against potential risks of long-term toxicity. However, it has proved a useful addition to our treatment armamentarium. Going forward, investigation of those most likely to benefit from treatment and to understand the long-term risks of treatment is essential for future patient selection and management.

Funding

Dr Morris: NIHR RCF grant from the University of Oxford (AC12/092).

Unpublished work

Morris, KA, Parry A, Pretorius Pieter M, A comparison of linear and volumetric tumor measurement techniques in the monitoring of NF2 related tumor bevacizumab responses.

McCabe MG et al, Outcomes of extra-cranial schwannomas in NF2 patients treated with bevacizumab

This material has been presented in part elsewhere

Posters: Morris KA, Afridi S, Axon P, Evans DG, Halliday D, Parry A Clinical and Radiological Response of NF2 Associated Tumours to Bevacizumab in an English Cohort, AAN 2014, Philadelphia (abstracts published online) Neurology 2014;82 (Meeting Abstracts): P3.326.

Morris KA, Golding JF, Parry A, Halliday D, Durie-Gair J, Evans DG, Ferner RE, Use of the Neurofibromatosis 2 Impact on Quality of life questionnaire (NFTI-QOL) in monitoring NF2 patients treated with Bevacizumab.CTF NF Conference 2014, Washington DC, June 7–9 2014.

Morris KA, McCabe MG, Afridi S, Jena R, Blesing C, Halliday D, Evans DG, Axon P, Pretorius P, Parry A, Bevacizumab treatment interruption in NF2: the effect of treatment break duration on the likelihood of vestibular schwannoma regrowth CTF NF Conference 2014, Washington DC, June 7–9 2014.

Oral Presentation: Katrina Morris and D Gareth Evans on behalf of the UK Nationally commissioned NF2 services Roundtable discussion: Update on bevacizumab in NF2: Comparing results across the Atlantic – the UK experience. 4th NF2 State of the Art Conference, Boston, June 4-5 2014.