Low-dose bevacizumab induces radiographic regression of vestibular schwannomas in neurofibromatosis type 2: A case report and literature review.

The current case study aimed to explore the efficacy of a low-dose bevacizumab regimen in inhibiting tumor growth and minimizing adverse effects. A 55-year-old man with neurofibromatosis type 2 (NF2) suffered bilateral vestibular schwannomas (VS) measuring 5.25 and 2.54 cm3 on the left and right, respectively. His capacity for bilateral language recognition was impaired. However, the patient refused microsurgical tumor resection and gamma knife therapy. Low-dose bevacizumab regimen (3.3-2.2 mg/kg every 2-4 weeks) was administered by intravenous injection for ~1.5 years to inhibit tumor growth and avoid further deterioration of hearing. Compared with baseline measurements prior to treatment, the bilateral VS regressed to 3.59 cm3 (68%) and 2.08 cm3 (82%) on the left and right, respectively. No hearing improvement was detected; however, the patient subjectively experienced a significant hearing improvement as his ability to communicate with people and distinguish voices was restored. No adverse effects were observed. Bevacizumab provides an alternative treatment option for those who refuse surgical intervention. Given the adverse effects commonly induced by bevacizumab, the use of a low-dose regimen would appear to be promising with regard to tumor regression and hearing preservation for patients with VS in NF2. However, the minimum dose required to sustain a response to bevacizumab in NF2 patients remains unknown. Finding the minimum effective dose sufficient to sustain hearing and/or volumetric response for individual patients is required.

Introduction

Neurofibromatosis type 2 (NF2) is a rare autosomal dominant syndrome that is characterized by the presence of bilateral vestibular schwannomas (VS; acoustic neuroma). In certain patients, the disease may be complicated by meningioma, ependymoma, spinal schwannomas or vitreous opacities. The growth of VS may lead to brain stem compression, tinnitus, progressive hearing loss, deafness, ataxia and eventual mortality. The incidence of NF2 in the population is 1 in 40,000–33,000, and 50% of the cases represent sporadic mutation, while the other half inherit the disease from their parents (1). Microsurgery for tumor resection and stereotactic radiotherapy are the standard treatment strategies; however, these may sacrifice hearing to achieve tumor control.

A vast amount of data has indicated that angiogenesis is pivotal for tumor growth, and vascular endothelial growth factor (VEGF) is a vital factor affecting angiogenesis and vascular permeability. Therefore, molecular targeted treatment may be promising for NF2-related tumors. Bevacizumab is the first Food and Drug Administration-approved monoclonal antibody that is able to neutralize the activity of VEGF in order to inhibit angiogenesis, tumor metastasis and growth, specifically (2). Bevacizumab has been used to treat malignant tumors, including metastatic rectal carcinoma and advanced non-small cell lung cancer. Plotkin et al (3) previously achieved a favorable outcome by treating NF2 with bevacizumab [5 mg/kg/2 weeks; intravenous injection (IV)], including radiographic tumor regression and hearing improvement. To the best of our knowledge, the optimal regimen remains to be defined. A dosage of 5 mg/kg/2 weeks is widely used; however, this may induce adverse effects, including hypertension, thrombosis or delayed wound healing. Thus, establishing a well-tolerated therapy would be promising for patients, given the adverse effects of bevacizumab.

In the present case involving a patient with bilateral VS in NF2, a lowered dose of bevacizumab and protracted infusion interval was trialled so as to explore the efficacy of low-dose bevacizumab regimen in inhibiting tumor growth and minimizing the adverse effects.

Case report

A 55-year-old man who was suffering with left progressive hearing loss and facial weakness was misdiagnosed with nerve deafness in 2007 at Binzhou Medical University Hospital (Binzhou, China). The patient presented right hearing loss in April 2009. On February 25, 2013, contrast-enhanced magnetic resonance imaging (MRI; Signa HDx 3.0T and 1.5T MR; GE Healthcare Life Sciences) (Fig. 1) revealed bilateral VS measuring 5.25 and 2.54 cm3 on the left and right, respectively (Fig. 1a–b). No mass effects were detected on MRI of the spinal cord. Café au lait spots (3×2 cm) were noted in the left anterior tibial skin. The patient was conscious without nystagmus, and his left nasolabial groove was shallower than the right. These findings confirmed a diagnosis of NF2. The pure tone average (PTA) threshold of 0.5, 1 and 2 kHz tones presented by bone-conduction indicated an impaired capacity for bilateral language recognition (Fig. 2). However, the patient refused microsurgical tumor resection and gamma knife therapy due to fears of surgery complications. On February 28, 2013, bevacizumab (Roche Diagnostics, Indianapolis, IN, USA) was administered by IV to inhibit tumor growth and avoid further deterioration of hearing initially. The drug dose was calculated according to patient's body weight (91 kg). Initially, 300 mg (91 kgx3.3 mg/kg) was administered every 2 weeks for 3 months. Subsequently, the dose of bevacizumab was gradually lowered to 2.2 mg/kg every 4 weeks in order to avoid adverse effects. Given that the patient had history of hypertension, felodipine (5 mg/day) was administered orally to maintain the patient's blood pressure within a normal range.

Figure 1.

Bilateral vestibular schwannomas are shown in the area selected by the red line. The maximal perpendicular diameters (R and r) of the tumor were measured on axial magnetic resonance imaging scans, and the extent in the coronal direction (L) was measured on coronal scans. The total volumes (V) of the tumor were then approximated by using the formula for a spheroid: V = R × r × L × π/6. (a and c) T1WI (axial) + contrast + fat suppression; (b and d) T1WI (coronal) + contrast; (e, g, i and k) T1WI (axial) + contrast + flash; and (f, h, j and l) T1WI (coronal) + contrast + flash.

Figure 2.

Compared with baseline measurements prior to treatment, the vestibular schwannomas regressed by 69% and 80% in the left and right, respectively, following bevacizumab treatment for a year.

Following bevacizumab treatment (3.3 mg/kg, every 2 weeks) for a period of 3 months, the patient's hearing level remained stable without deterioration and the bilateral VS regressed by 56% and 76% on the left and right, respectively (Figs. 1c–d and 2). The central part of tumor was not well enhanced due to the rapid inhibition of angiogenesis by bevacizumab. On June 17, 2013, the patient was recommenced on bevacizumab (3.3 mg/kg, every 3 weeks), administered by IV, for the next 3 months. His hearing remained stable as before and the left VS further regressed by 13%, while the right VS remained stable (Figs. 1e–f and 2). Given the adverse effects induced by bevacizumab, including anemia, neutropenia and lymphocytopenia (4), on September 14, 2013, the dose was further lowered to 2.2 mg/kg and the infusion interval protracted to 4 weeks so as to reduce the risk of hypertension aggravated by bevacizumab whilst preventing tumor recurrence following drug discontinuation.

After a constant treatment (2.2 mg/kg) for 1 year, the patient's hearing was successfully preserved by low-dose bevacizumab, without further progression (Fig. 3). Although no hearing improvement was detected by pure tone audiometry (using the Madsen Midimate 622 audiometer; GN Otometrics, Taastrup, Denmark), the patient subjectively experienced a significant hearing improvement as his ability to communicate with people and distinguish voices was restored. Compared with baseline measurements prior to treatment, the bilateral VS regressed by 3.59 cm3 (68%) and 2.08 cm3 (82%) on the left and right, respectively (Fig. 1g–l; Table I). At the time of writing, the patient was continuing to receive bevacizumab treatment (2.2 mg/kg, every 4 weeks) by IV without any significant adverse effects observed, and with no signs of tumor progression or hearing deterioration.

Figure 3.

PTA on the right was marginally improved from 57.0 dB HL to 48.3 dB HL following bevacizumab treatment. PTA on the left was also slightly improved from 57.6 dB HL to 38.3 dB HL following treatment for a year (a PTA <25 dB HL is normal for the majority of people). PTA, pure tone average; dB HL, decibels hearing level. AC, air conduction; BC, bone conduction; NR, no response.

Table I.

Tumor size analysis.

	Baseline		3 months		6 months		9 months		12 months		16 months
Dimension	Left	Right	Left	Right	Left	Right	Left	Right	Left	Right	Left	Right
R, mm	27.5	18.3	20.0	11.0	17.7	11.1	17.8	11.6	17.2	10.9	18.2	9.8
r, mm	18.7	15.6	15.1	10.5	12.1	10.6	13.4	10.6	12.6	10.3	11.3	9.4
L, mm	19.5	17.0	14.6	10.1	14.0	11.3	13.2	9.5	13.9	9.3	15.4	9.5
V, cm3	5.25	2.54	2.31	0.61	1.57	0.70	1.65	0.61	1.58	0.55	1.66	0.46

R / r, maximal perpendicular diameters; L, length in coronal direction; V, volume.

Discussion

A complete literature review was conducted using computer search engines in PubMed (http://www.ncbi.nlm.nih.gov/pubmed) to identify all cases of VS in NF2 treated with bevacizumab. The search was performed using single or combined search terms, including ‘bevacizumab’, ‘bilateral vestibular schwannomas’, ‘neurofibromatosis 2’ and ‘adverse effects’. In total, 6 relevant reports comprising 39 cases of VS in NF2, published between 2010 and 2014, were included (3,5–9). The clinical characteristics are summarized in Table II, including drug dose, number of patients, treatment duration, mean age and adverse effects. Radiological response and hearing status are detailed in Table III. All patients received bevacizumab 5 mg/kg/2 weeks by IV. The mean age was 26±2.2 years (range, 12–73 years). The median duration of treatment was 12 months, and 85% of the patients (33/39) experienced tumor regression. The median tumor volume reduction was 27.5% (range, 3–91%). The solid component regressed less than the cystic component of the tumor (8). However, the findings also indicate that tumor regrowth may occur after drug discontinuation (9). The rate of hearing improvement was 45.2% (14/31), whilst the rate of hearing stability was 48.4% (15/31). Hearing loss was observed during treatment intermission and disappeared after treatment resumed (7). Plotkin et al (7) reported bevacizumab treatment for progressive VS in 31 patients with a 3-year follow-up. The rates of tumor stability or regression were 88% at 1 year, 67% at 2 years and 54% at 3 years. The rates of hearing stability or improvement were 90% at 1 year, 81% at 2 years, and 61% at 3 years. However, 168 adverse events were identified during 572 patient-months of treatment. The frequency of adverse events was high, demonstrating how adverse effects have become an obstacle for the clinical use of bevacizumab.

Table II.

Summary of patient characteristics and adverse effects in previous studies.

Author, year	Dose	Patients, n	Median duration (range), months	Mean age (range), years	Adverse effects	Ref.
Subbiah et al, 2012	5 mg/kg IV/2 wk	2[a]	9.5 (9–10)	28.5 (16–41)	No significant adverse effects	(5)
Eminowicz et al, 2012	5 mg/kg IV/2 wk	2	3.5 (3–4)	34 (31–37)	No significant adverse effects	(6)
Plotkin et al, 2012	5 mg/kg IV/2 wk	31	14 (6–41)	26 (12–73)	Hypertension, proteinuria, menorrhagia, epistaxis, pneumonia	(7)
Plotkin et al, 2009	5 mg/kg IV/2 wk	10	12 (3–19)	25 (16–53)	Hypertension, proteinuria, menorrhagia, epistaxis, pneumonia	(3)
Mautner et al, 2010	5 mg/kg IV/2 wk	2	4.5 (3–6)	32 (24–40)	No significant adverse effects	(8)
Mautner et al, 2010	5 mg/kg IV/2–4 wk	2	15 (12–18)	30 (22–38)	Hypertension	(9)

Only 2 of 6 patients were treated by bevacizumab alone. IV, intravenous injection; wk, weeks.

Table III.

Summary of radiological response and hearing status after the treatment.

			Hearing status, % (n)[a]
Author, year	Tumor reduction, median (range)	No. of patients experiencing tumor regression	Improved	Stable	Declined	Ref.
Subbiah et al, 2012	Stable	0 of 2	–	100 (2 of 2)	–	(5)
Eminowicz et al, 2012	30.5% (10–52%)	2 of 2	–	100 (2 of 2)	–	(6)
Plotkin et al, 2012[b]	26% (3–91%)	27 of 31	57 (13 of 23)	35 (8 of 23)	8 (2 of 23)	(7)
Plotkin et al, 2009	26% (5–44%)	9 of 10	57 (4 of 7)	29 (2 of 7)	14 (1 of 7)	(3)
Mautner et al, 2010	42% (41–43%)	2 of 2	–	100 (2 of 2)	–	(8)
Mautner et al, 2010	47.5% (43–52%)	2 of 2[c]	50 (1 of 2)[d]	50 (1 of 2)	–	(9)

Patients not eligible for a hearing response were excluded.

This is an extended follow-up study of Plotkin's previous study.

Tumor regrowth following drug discontinuation in 1 patient; tumor regression can be sustained under low-dose bevacizumab.

Hearing improved during treatment but worsened following drug discontinuation.

The NF2 tumor suppressor gene has been shown to reside on chromosome 22q12.2, with the highest mutation rate among human genetic diseases (10). The aberration of the NF2 gene may lead to neoplasia of ectodermal and mesodermal tissues during embryonic development phases by affecting a variety of signal transduction pathways associated with tumor genesis and progression, including the phosphoinositide 3-kinase/protein kinase B/mammalian target of rapamycin pathway and the Raf/Ras/mitogen-activated protein kinase kinase pathway (5,11). Loss of a functional NF2 gene product, Merlin (a tumor suppressor protein), may inhibit the expression of anti-angiogenesis semaphorin-3F by up regulating the Rac1 pathway in VS, resulting in a relative surplus of VEGF (12). Furthermore, tumor necrosis occurs when a tumor's blood supply cannot satisfy its growth, which may activate the compensatory overexpression of VEGF to improve the hypoxic microenvironment (12). Hence, VEGF blockade using bevacizumab may normalize the vascularity and decrease edema (Fig. 4) (12).

Figure 4.

Mechanism underlying effect of VEGF blockade with bevacizumab on vascularity and edema. VEGF, vascular endothelial growth factor; VEGFR, VEGF receptor; NF2, neurofibromin 2; SEMA3F, semaphorin-3F; P, phosphate; PI3K, phosphoinositide 3-kinase; AKT, protein kinase B; mTOR, mammalian target of rapamycin; MEK, mitogen-activated protein kinase kinase.

Microsurgery for tumor resection and stereotactic radiotherapy are usually recommended for VS in NF2 (13). However, patients must abandon hearing in order to achieve tumor control, and are also subject to possible postoperative complications, such as cerebrospinal fluid leak and facial weakness. Unfortunately, postoperative tumor recurrence may induce the need for reoperation in some patients, resulting in a decline in quality of life (14). In the current study, anti-angiogenesis therapy with bevacizumab objectively induced tumor regression and hearing preservation in a case of VS in NF2. A low-dose regimen was beneficial for avoiding significant adverse effects, providing an alternative treatment strategy instead of surgical intervention.

Compared with baseline measurements prior to treatment, the bilateral VS regressed by 3.59 cm3 (68%) and 2.08 cm3 (82%) on the left and right, respectively (Fig. 1; Table I). The absolute change in tumor volume was not as significant as the relative change; as the tumors in the current patient were not very large, a small absolute volume change may result in a large relative percentage reduction in tumor volume compared with the baseline. To avoid the potential bias in tumor volume analysis, relative percentage and absolute volume changes from baseline were used for comparison. The tumor volume remained stable in last 10 months without further regression. This suggests that tumor stability may be a compromise between tumor regression induced by low-dose regimen and tumor growth.

Although no objective hearing improvement was detected in the patient, subjective hearing improvement did occur. Furthermore, the patient's hearing declined during treatment intermission and recovered after treatment was resumed, while the tumor volume remained stable. Thus, we hypothesized that hearing stability was drug-dependent. It is possible that higher-dose regimens may be helpful for hearing improvement, whilst low-dose regimens are not. In addition, the biological mechanisms underlying the effects of bevacizumab for hearing stability may be different from those for tumor regression, which have been reported previously (15).

Whether bevacizumab may be used for long-term treatment remains controversial due to the risk of adverse effects, which include hypertension, proteinuria, thrombosis and hemorrhage (7). Furthermore, considering tumor recurrence following drug discontinuation (9), the optimal dose and duration of treatment is undetermined. As reported by Zuniga et al (16), disease recurrence and more aggressive progression were also detected in the treatment of malignant glioma following bevacizumab discontinuation. However, no evidence has indicated that bevacizumab discontinuation leads to accelerated disease progression. Therefore, the current patient is undergoing close monitoring to observe the efficacy of low-dose regimen in avoiding severe adverse effects and tumor recurrence following drug discontinuation.

Given the adverse effects induced by systemic IV administration of bevacizumab and the insufficient drug concentration at the tumor site due to the blood-tumor barrier (BTB), Riina et al (17) reported a novel approach involving super-selective intra-arterial cerebral infusion of bevacizumab following BTB disruption. Of 3 patients, 1 experienced tumor regression of 11% and 19% on the left and right, respectively. All 3 patients presented hearing improvement or stability. Appropriate cerebrovascular interventional therapy may be better in certain circumstances. For example, with regard to local chemotherapy, cerebrovascular interventional therapy results in less neurotrauma than craniotomy. Furthermore, cerebrovascular interventional therapy may benefit patients exhibiting tumors which extend into the internal auditory canal, as gross total resection is difficult, patients that cannot tolerate the adverse effects induced by systemic chemotherapy, and patients with a Karnofsky performance scale score of ≥60 (18).

As the pathogenesis of NF2 depends on multiple signaling pathways, the combination of bevacizumab with other molecular targeted drugs, such as erlotinib and lapatinib (inhibitors of epidermal growth factor receptor), may generate synergistic effects (5,19). However, the efficacy of this treatment must be confirmed by clinical trials containing a larger cohort of patients.

For the patients who exhibit no response to bevacizumab, a surgical intervention is recommended following a 4-6-week drug metabolism phase to minimize the risk of postoperative hemorrhage and delayed wound healing (6,8).

In summary, a low-dose regimen of bevacizumab (2.2 mg/kg/4 weeks, IV) would appear to be promising for patients with VS in NF2, given the adverse effects triggered by bevacizumab (4). However, the minimum dose required to sustain a response to bevacizumab in NF2 patients is still unknown, and may differ from patient to patient. Finding the minimum effective dose for individual patients sufficient to sustain hearing and/or volumetric response would aid in decreasing toxicity and long-term tolerability. With the scientific advancements, more comprehensive and safe regimens may be determined.