Counseling patients who are diagnosed with vestibular schwannomas, formerly known as acoustic neuromas, can be challenging. The health care provider has the responsibility to explain, in understandable language, to the patient or legal representative the proposed treatment options, risks and complications associated with each form of treatment, and alternatives to treatment, including no therapy. Patients should be encouraged to gather information before making a treatment decision. For the physicians managing these patients, information should be delivered in a balanced way to ensure patient understanding of their options leading to adequate informed consent.
Counseling patients who are diagnosed with vestibular schwannomas (VS), formerly known as acoustic neuromas, can be challenging. These benign neoplasms, which originate from either the superior or inferior vestibular nerves, have an average growth rate of 1 mm 3 per year. Larger tumors can cause brainstem compression with few noticeable symptoms, whereas smaller VS can cause vertigo, tinnitus, and hearing loss.
Patients are confronted by the difficult task of choosing a treatment method based on the advice of caregivers, currently available literature, and Internet-based information sources. Patients often visit the Internet either before or during their decision period, which can be helpful or even more confusing for them as they weigh their options. The health care provider has the responsibility to explain, in understandable language, to the patient or legal representative governing the patient’s care the proposed treatment options, risks and complications associated with each form of treatment, and alternatives to treatment, including no therapy. The medical record must contain evidence of the patient’s informed consent, with the exception of emergency situations in which a delay in intervention could compromise outcomes in a life or limb-threatening situation. Aside from cases with brainstem compression and hydrocephalus, patients should be encouraged to gather information before making a treatment decision. For the physicians managing these patients, information should be delivered in a balanced way to ensure patient understanding of their options leading to adequate informed consent.
Options for treatment include radiation therapy, surgical excision, and observation with serial MRI. Discerning treatment advantages from a particular modality is made difficult because of nonstandardized definitions of tumor control and hearing preservation and varied posttreatment intervals presented in the medical literature. Surgical techniques and radiotherapy dosage paradigms have evolved considerably over the past two decades. Currently, no randomized, prospective clinical trial has compared the three treatment options and there are no clearly accepted, evidence-based, best practices for managing VS.
The treatment of VS requires a multidisciplinary team not only to deliver the chosen therapy but also to assist in the decision-making process. At our center, a nurse familiar with VS treatment coordinates appointments with a neurotologist, a neurosurgeon, and a radiotherapist. Patients are encouraged to take the amount of time necessary to make a decision with full understanding of potential risks and benefits. The amount of time necessary to come to a decision depends on the needs of a particular patient.
This article evaluates the English language literature, dating back to 1994, to present long-term results for tumor control and complication rates for the treatment of VS, excluding cases of neurofibromatosis type II. The goal is to provide a guide to otolaryngologists who provide initial counseling for patients with newly diagnosed VS.
Stereotactic radiation therapy
Radiation therapy was initially used as an adjunct to surgery in patients with incompletely resected VS. In a study reported from University of California San Francisco, Wallner and colleagues demonstrated that conventional fractionated radiation therapy to more than 45 Gy significantly reduced regrowth from 46% to 6% in incompletely resected VS. In this series, 31 patients treated between 1945 and 1983 were followed from 2.6 to 40.7 years. The authors concluded that postoperative radiation therapy reduced the “local recurrence rate” of VSs that were incompletely excised or only biopsied and demonstrated the effectiveness of radiation therapy in the treatment of acoustic neuromas.
Stereotactic radiosurgery, first developed in 1951 by Leksell , is a method of delivering a highly conformal single dose of ionizing radiation with submillimeter accuracy to an intracranial target. The goal of a single high-dose delivery was to cause tumor necrosis and control of growth as an alternative to surgery for patients who were suboptimal candidates for excision. The first stereotactic radiosurgery treatment for VS was performed in 1969 with the gamma knife (GK), also developed by Leksell . The GK is a highly specialized radiation delivery system that uses 201 radioactive cobalt 60 sources to deliver high-dose radiation accurately to tumor masses. A stereotactic frame is fixed to the skull and attached to the treatment table to provide rigid immobilization of the patient’s head and ensure accurate localization of the radiation dose.
Thousands of patients have been treated with the GK. Initially, doses were high, and although tumor control was excellent, toxicity was significant . Several refinements, including dose reduction, improved target definition, and treatment accuracy, have provided excellent tumor control while minimizing toxicities of treatment . Lower radiation doses were shown to be effective in controlling tumor growth, whereas tumor shrinkage could take several years to document radiographically.
The linear accelerator (LINAC) also can be adapted to perform stereotactic radiosurgery . Multiple beam positions or arcs are used to create a conformal dose distribution around the target. Although the tumor margin doses are similar to the GK, LINAC radiosurgery typically uses fewer isocenters and the dose to the tumor is more homogeneous. Similar to the GK, a non-relocatable invasive head frame is required for patient immobilization during treatment.
Fractionated stereotactic radiation therapy is the most recently developed technique for delivering high-dose and localized treatment. Unlike with GK or LINAC, a noninvasive relocatable head frame or thermoplastic mask is used for fractionated stereotactic radiation therapy. This head frame increases patient comfort but may result in less dose conformality when compared with other radiosurgery techniques. Subscribers of this technique believe that fractionation takes advantage of radiobiologic principles to reduce late toxicity while maintaining tumor control . The fractionated treatment regimens range from doses given over several days to standard fractionation given over 4 to 5 weeks, similar to the scheme originally used by Wallner.
Data regarding the outcomes GK, LINAC, and FSRT are reviewed in the next section. Because few centers in the world offer proton beam radiation therapy to treat VS, studies that used proton-based treatments were excluded. All pertinent papers published from 1994 to the present that assess local control rates with a median follow-up of at least 2 years were reported. Outcomes included tumor control, hearing preservation, facial neuropathy, and trigeminal neuralgia. In most patients, tumors were sporadic and had a maximum diameter of ≤3 cm.
The goal of radiation therapy is to arrest tumor growth. Local control rate in radiotherapy studies can be defined as the percentage of tumors that do not increase in size on follow-up imaging. Many researchers define the local control rate as the percentage of tumors that do not require salvage therapy. This determination could overestimate the control rate because some tumors may have progressed but are not symptomatic enough to require further treatment. All three techniques seem to achieve excellent local control with a range of 87% to 100% ( Tables 1–3 ) .
| Author (year) | Number of patients | Local control rate | Hearing preservation rate | Facial neuropathy rate | Trigeminal neuropathy rate | Other complications |
|---|---|---|---|---|---|---|
| Flickinger, 2004 | 313 | 98.6% @ 6 y | 78%@ 6 y | 0% @ 6 y | 4.4% @ 6 y | |
| Andrews, 2001 | 63 | 98% | 33% | 2% | 5% | hc 2.9% vertigo 1.4% ataxia 1.4% |
| Massager, 2006 | 82 | 98% | 65% | |||
| Ottaviani, 2002 | 30 | 87% | 73% | 3% temp | 16% mild symptoms | |
| Wackym, 2004 | 29 | 94% | 0% | 3% temp | ||
| Prasad, 2000 | 153 | 92% | 75% for tumors <1 cc, 57% for tumors >1 cc | 2.3% | 1.7% perm, 2.5% temp | |
| Chung, 2005 | 195 | 95% @ 7 y | 60% | 1.4% temp | 1.1% | hc 3.5% |
| Wowra, 2005 | 111 | 95% @ 6 y | 2.7% | 11.5% | ||
| Lundsford, 2005 | 829 | 97% @10 y | 78.6% | 0% | 4.4% | ventriculoperitoneal shunt 0.8%, tinnitus 0.2%, peritumoral cyst 3.6%, no radiation-induced malignancy |
| Petit, 2001 | 45 | 96% | 88% | 4% temp | 0% | |
| Hasegawa, 2005 | 317 | 92% @10 y | 68% | 1% temp | 2% | hc 4.1%, malignant transformation 0.3% |
| Pollock, 2006 | 46 | 100% | 63% | 4% | 2.1% | hc 4% |
| Mysreth, 2005 | 103 | 89.2% | 5.2% | hc 3.9% | ||
| Paek, 2005 | 25 | 100% | 46% | 0% | 5% | |
| Litvack, 2003 | 134 | 97.7% | 61.7% | 2.2% temp | 5.8% temp | hc 3.0% |
| Hirato, 1996 | 29 | 93% | 59% | 10.3% temp | 3% temp | hc 6.8% |
| Muacevic, 2004 | 219 | 97% | 49% | 0.5% | 5% temp | |
| Iwai, 2003 | 51 | 96% | 56% | 0% | 4% | hc 8% |
| Kwon, 1998 | 88 | 95% | 67% | 4% temp, 4% perm | 3.4% temp | hc 3.5% |
| Author (year) | Number of patients | Local control rate | Hearing preservation rate | Facial neuropathy rate | Trigeminal neuropathy rate | Other complications |
|---|---|---|---|---|---|---|
| Friedman, 2006 | 390 | 90% @ 5 y, only 1% required surgery | 4.4% (0.7% w/dose ≤12.5 Gy) | 3.7% (0.7% w/dose dose ≤12.5 Gy) | ||
| Chung, 2004 | 45 | 100% | all patients deaf before tx | 4.4% perm | 7.5% perm | hc 4.4% |
| Combs, 2006 | 26 | 91% @ 10 y | 55% | 5% perm, 15% temp | 8% | 1 patient with radionecrosis of cerebellum |
| Meijer, 2003 | 49 | 100% | 75% | 7% | 8% | hc 2% |
| Speigelman, 2001 | 44 | 98% | 71% | 2.4% temp, 8% perm | 18% | |
| Foote, 2001 | 149 | 95.4% | 11.8% temp | 9.5% temp | ||
| Lee, 2001 | 42 | 100% | 82% | 2.5% perm | 2.5% perm | hc 7.5% |
| Author (year) | Number of patients | Dose | Local control rate | Hearing preservation rate | Facial neuropathy rate | Trigeminal neuropathy rate | Other complications |
|---|---|---|---|---|---|---|---|
| Andrews, 2001 | 46 | 50 Gy/25 fx | 97% | 81% | 2% | 7% | hc 3.6%, vertigo 1.7%, ataxia 3.6% |
| Combs, 2005 | 106 | 57.6 Gy/32 fx | 93% | 98% | 2.3% | 3.4% | |
| Chung, 2004 | 27 | 45 Gy/25 fx | 100% | 57% | 4% temp | 7% temp | hc 4% |
| Shirato, 1999 | 33 | 36–44 Gy/20–22 fx | 98% | 53% | 5% temp | 12% temp | |
| Williams/Shokek, 2004 | 375 | 25 Gy/5 fx (<3 cm diam), 30 Gy/10 fx (>3 cm diam), 40 Gy/20 fx | 97% | 59% | 1.5% temp 0.5% perm | 1.2% temp | |
| Meijer, 2003 | 80 | 20–25 Gy/5 fx | 94% @ 5 y | 61% | 3% | 2% | |
| Fuss, 2000 | 42 | 57.6 Gy/32 fx | 97.5% @ 5 y | 85% | 0% | 4.8% | |
| Selch, 2004 | 48 | 54 Gy/30 fx | 100% | 91.4% | 2.8% | 3.8% | Tinnitus 5%, hc 0%, ataxia 2% |
| Sawamura, 2003 | 101 | 40–50 Gy/20–25 fx | 97% | 71 | 4% temp | 14% | hc 11% disequilibrium 16.8% |
| Poen, 1999 | 31 | 21 Gy/3 fx/24 h | 97% | 77% | 3% | 16% | |
| Chan, 2005 | 70 | 54 Gy/30 fx | 92% | 84% | 1% | 4% | |
| Chang, 2005 | 61 | 21 Gy/3 fx/24 h | 98% | 74% | 0% perm | 0% | |
| Lederman, 1997 | 38 | 20 Gy/4 fx or 20 Gy/5 fx | 100% | 93.5% (does not distinguish useful from no useful hearing) | 2.6% temp | 0% | |
| Ishihara, 2004 | 38 | 15–20.5 Gy/1–3 fx | 94% | 93% | 2.6% temp | 2.6% |
Up to 50% of radiated tumors developed central necrosis that results in transient increase in tumor volume (23% of cases) . This phenomenon was observed up to 4 years after treatment and took from 6 months to 5 years to disappear. If tumor progression was defined as tumor growth, then some patients may have undergone surgery for salvage treatment unnecessarily .
Lundsford and colleagues and Hasegawa and colleagues reported 10-year local control rates of more than 90% in separate series of patients treated with GK. The median follow-up period in Hasegawa’s study was 7.8 years. Partial or complete radiographic response to treatment occurred in 62% of radiated tumors, and tumors <15 cm 3 had a better progression-free survival than tumors >15 cm 3 (96% versus 57%, P < .001). Tumors not compressing the brainstem or obstructing the fourth ventricle had a better progression-free survival (97% versus 74%, P < .008). Tumor progression occurred within 3 years from the time of treatment in most cases. Forster and colleagues documented local control rates for tumors >3 cm, 2 to 3 cm, and <2 cm at 33%, 86%, and 89%, respectively. Similar findings were reported by Kondziolka and colleagues .
Friedman and colleagues reviewed the outcomes of 390 patients treated with LINAC radiosurgery for VS. With a median follow-up of 40 months and a median dose of 12.5 Gy, the 5- and 10-year local control rates were 90% with only 1% of patients requiring surgery for treatment failure during the follow-up period. With a median follow-up period of 48.5 months, Combs and colleagues also reported a local control rate of 91% at 10 years.
The longest follow-up periods for the FSRT studies are from Combs and colleagues , Sawamura and colleagues , and Chan and colleagues . The median follow-up periods in these studies averaged 48 months, and the 5-year local control rates in all three studies are more than 90% (see Table 3 ).
Unlike with use of GK, tumor volume or size has not been shown to be of prognostic value in predicting response to treatment with LINAC radiosurgery or fractionated stereotactic radiotherapy for acoustic neuromas. Different fractionation regimens were used depending on tumor size .
Hearing preservation has not been documented according to a consistent standard ( Table 4 ), and no randomized studies regarding tumor control and hearing preservation have been reported to date. The use of pure tone audiometry and discrimination testing before and at a standardized interval after treatment would be optimal. In most of the radiotherapy papers reviewed, the statistic that is most often reported is the percentage of patients who maintained or gained useful hearing (Gardner-Robertson classes 1 and 2). Intervals for testing hearing after treatment are not clear, and in some studies residual hearing was defined by whether patients could talk on the telephone on the side of the affected ear. Hearing loss is rare, tends to occur over 6 to 24 months, and can continue to decline for years after treatment. Prasad reported a useful hearing preservation rate of 58% with a median follow-up of 4.2 years. In most patients, hearing decline occurred after 2 years and continued up to 8 years after treatment, with a hearing preservation rate of 75% for tumors <1 cm 3 compared with 57% for tumors >1 cm 3 . Massager demonstrated better useful hearing preservation rates in tumors with intracanalicular volumes <100 mm 3 compared with larger tumors (82.6% versus 44.8%, P = .045).
| AAO-HNS classification | ||
|---|---|---|
| Class | Pure tone average (0.5, 1, 2, 3 kHz measured in dB HL) | Speech discrimination score (%) |
| A | 0–30 | 70–100 |
| B | 31–50 | 50–100 |
| C | >50 | 50–100 |
| D | Any | <50 |
| Gardner-Robertson Classification | ||
| Class | Pure Tone/Speech Reception Threshold (dB HL) | Speech Discrimination Score (%) |
| 1 | 0–30 | 70–100 |
| 2 | 31–50 | 50–69 |
| 3 | 51–90 | 5–49 |
| 4 | >90 | 1–4 |
| Word Recognition Scores | ||
| Class | Word Recognition Score (%) | |
| I | 70–100 | |
| II | 50–69 | |
| III | 1–50 | |
| IV | 0% | |
Improved hearing outcome has been demonstrated at doses ≤13 Gy. In an early paper from the University of Pittsburg, Kondziolka and colleagues reported a “useful” hearing preservation rate of 46% at an average marginal dose of 16 Gy. Later studies documented hearing preservation in up to 68% to 78% of cases treated in the range of 12 to 13 Gy . Because there were no differences in the local control rates with the lower doses, it is common practice to prescribe 12 to 13 Gy to the tumor margin when using stereotactic radiosurgery to control VS growth and maximize hearing preservation.
Others attempt to minimize toxicity and improve hearing preservation rates by fractionating the dose. This practice is based on a radiobiologic principle that there is a direct relationship between late normal tissue damage and dose per treatment delivered to the tissue. Andrews and colleagues performed a prospective, nonrandomized study to compare the outcomes of treatment for acoustic neuromas with GK radiation to FSRT. The GK dose was 12 Gy and the FSRT dose was 50 Gy in 25 fractions given daily. They found no difference in local control rates, but the hearing preservation rate was 33% for GK treatment compared with 81% for fractionated stereotactic radiotherapy. These results should be viewed with caution because they had relatively short mean follow-up of less than 3 years. Although most of the FSRT studies reported excellent hearing preservation rates, they have shorter follow-up than other radiosurgery series. The different fractionation regimens make it difficult to make any conclusions regarding dose and hearing preservation.
Facial neuropathy is a potential complication with all three treatment options for VS. Kondziolka and colleagues performed a multivariate analysis of 162 patients treated with GK and found that tumor volume and dose of radiation to the tumor margin were associated with the risk of neuropathy ( P < .001). With an average marginal dose of 16 Gy, the overall rate of facial neuropathy was 15%. In subsequent reports, when 12 to 13 Gy were prescribed to the margin, the facial neuropathy rate dropped to 0%. Friedman and colleagues presented a similar finding when they lowered their LINAC radiosurgery dose to 12.5 Gy (4.4% versus 0.7%). As seen in Tables 1 and 2 , the risk of facial neuropathy with radiosurgery is minimal and often temporary with a dose of 12 to 13 Gy. Similar rates of facial neuropathy are reported with FSRT despite variable doses and fractionation regimens (see Table 3 ). Unfortunately, no consistent facial nerve function scale was used for reporting.
After performing a multivariate analysis, Kondziolka and colleagues documented that tumor volume and radiation dosage to the tumor margin were associated with the risk of trigeminal neuropathy ( P < .001). With average marginal doses of 16 Gy, the overall rate of trigeminal neuropathy was 16% compared with 4.4% at 12 to 13 Gy. There was no evidence of trigeminal nerve damage for intracanalicular tumors. Friedman and colleagues noted a similar finding when they lowered their LINAC radiosurgery dose to 12.5 Gy (3.7% versus 0.7%).
Meijer and colleagues reported the outcomes of treatment with LINAC compared with FSRT for their patients with VS. They found a statistically significant increased incidence of trigeminal neuropathy in patients treated with FSRT (8%) versus LINAC radiosurgery (2%) ( P = .048).
Other potential side effects from radiation therapy include vertigo, tinnitus, ataxia, headache, hydrocephalus, cyst formation, radiation-induced edema or necrosis, intratumoral bleeding, and malignant transformation (see Tables 1–3 ). The reporting of these toxicities has not been consistent. The rate of hydrocephalus ranges from 0 to 11% . Sawamura and colleagues reported the highest incidence of hydrocephalus, with 11% of 101 patients treated with FSRT 40 to 50 Gy in 20 to 25 daily fractions manifesting communicating hydrocephalus that requires ventriculoperitoneal shunting. The hydrocephalus resolved in all patients with shunt placement and was assumed to be caused by cerebrospinal fluid (CSF) malabsorption associated with VS.
Delayed malignant transformation or radiation-induced malignancy may occur with radiation therapy for acoustic neuromas. Malignant transformation occurred in one patient 51 months after radiotherapy treatment. The malignant transformation rate in this study was 0.3% among patients who were followed longer than 5 years after GK . Two other case reports of malignant transformation have been reported with GK . One case was reported of a patient who developed a glioblastoma multiforme adjacent to an acoustic neuroma that was treated with GK 7.5 years earlier . Another patient treated with FSRT developed a low-grade malignant nerve sheath tumor 216 months after initial treatment . Delayed malignant transformation or radiation-induced malignancies are rare. Long-term, yearly follow-up of these patients provides a more accurate assessment of incidence.
Recent papers compared the outcomes of microsurgery to stereotactic radiosurgery to attempt to determine which treatment is better for sporadic VS ≤3 cm. Myrseth and colleagues reported a retrospective study of 189 consecutive patients—86 treated by microsurgery and 103 by GK. The mean follow-up period was 5.9 years. In addition to local control and cranial nerve preservation, they also evaluated quality of life through standardized questionnaires. The overall local control rates for microsurgery and GK were 89.2% versus 94.2%, which was not statistically significant. Facial nerve function (House-Brackmann grade 1-2) was preserved in 79.8% of the microsurgery group and in 94.8% of the GK group ( P = .0026). The middle fossa approach was not used in these patient cohorts. Overall, the quality-of-life scores were significantly lower in the microsurgery group compared with the GK group. The authors concluded that these results favored GK as the treatment of choice for this group of patients.
Pollock and colleagues presented a prospective cohort study of 82 patients with unilateral, unoperated VS <3 cm in greatest dimension undergoing surgical resection ( n = 36) or GK ( n = 46). Other than age (patients undergoing microsurgery were younger), all other pretreatment characteristics were matched in the two treatment groups. The mean follow-up period was 42 months. There was no difference in the local control rate between microsurgery and GK (100% versus 96%, P = .50). They found that facial nerve preservation (96% versus 75%, P < .01) and serviceable hearing rates (63% versus 5%, P < .001) were better in the GK group than microsurgery group. With regard to quality of life, patients who underwent microsurgery had a decline in physical functioning and bodily pain scores, whereas patients who had GK had lower Dizziness Handicap Inventory scores. All of these differences in quality of life were statistically significant. Although the authors concluded that radiosurgery should be considered the best treatment for this group of patients, they commented on the necessity of longer follow-up.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
