long article/nf2

[Guest guest]

This was written in 1998, but may be of interet to some.

The New England Journal of Medicine -- November 12, 1998 -- Vol.

339, No. 20

Treatment of Acoustic Neuromas

Acoustic neuromas (vestibular schwannomas) are benign

tumors of the eighth cranial nerve. They are unilateral and

typically become symptomatic after the age of 30 years.

Bilateral tumors usually are present in type 2

neurofibromatosis, are transmitted in an autosomal dominant

fashion, and have a deletion in the long arm of chromosome 22,

probably representing inactivation of a tumor-suppressor gene.

Acoustic neuromas usually present with tinnitus, a unilateral

reduction in hearing, and gait imbalance. If these symptoms are

disregarded and the tumor becomes large, it can lead to facial

numbness, weakness or twitching, or even signs of brain-stem

compression, such as numbness or weakness of the trunk or

extremities, difficulty swallowing, or hoarseness. Rarely is the

tumor large enough to cause hydrocephalus. Patients often

report hearing loss or " stuffiness " in one ear, which may be

slowly or, in some cases, rapidly progressive; these symptoms

can be mistaken for signs of occlusion of the eustachian tube or

otitis media and be treated with antibiotics or decongestants.

Persistence or worsening of symptoms usually prompts an

audiogram. Asymmetric reduction in hearing or speech

discrimination is then evaluated further with magnetic

resonance imaging (MRI), which should clearly demonstrate

the presence or absence of the acoustic neuroma or other lesion

adjacent to the eighth cranial nerve. The tumor results when

Schwann cells proliferate within the vestibular portion of the

nerve, usually beginning in the internal auditory canal and then

extending into the cerebellopontine angle. Occasionally,

schwannomas can arise on other cranial nerves, particularly the

fifth or seventh. These schwannomas usually grow extremely

slowly, over decades.

The first surgical removal of an acoustic neuroma was

performed in 1894. Surgical technique has improved

dramatically in this century, with the advent of the operating

microscope, neuroanesthesia, and intraoperative monitoring.

With optimal results, the tumor is removed completely and

safely and neurologic function is preserved at the preoperative

level, particularly with respect to hearing and facial movement

and sensation. In recent series (reviewed by Sekhar et al. (1)),

complete tumor removal was reported in 97 to 99 percent of

patients, mortality was below 1 percent, and facial movement

was normal or nearly normal in 94 to 97 percent of patients

with small tumors and 28 to 57 percent of those with large

tumors, as judged one year after surgery (some patients had

temporary weakness for weeks or months after the operation).

Patients with good hearing and small tumors retain their hearing

in 45 to 82 percent of cases. The frequency and severity of the

postoperative neurologic deficits increase with increasing

tumor size. (1)

Because of the morbidity and neurologic injury that can occur

with surgery, other treatments have been developed, among

them external-beam irradiation and stereotactic radiosurgery. In

1971, using multiple radiation sources and exposure ports,

Leksell (2) sought to maximize the irradiation of the tumor and

minimize the irradiation of the surrounding tissue. He called his

instrument a gamma knife, and the stereotactic technique is now

called radiosurgery. Great strides in instrument design,

imaging, and computational power now allow the delivery of

large doses of radiation to simple or complex targets with

relatively low exposure of surrounding structures.

In this issue of the Journal, Kondziolka and colleagues describe

their experience with radiosurgical treatment of acoustic

neuromas. (3) They report a low frequency of facial weakness

or numbness after treatment. The rate (47 percent) at which they

were able to maintain hearing levels that were useful before

treatment is similar to that reported in some surgical series, and

the 61 percent rate of preservation of some degree of hearing is

superior to the results in almost all surgical series. Clearly,

shorter hospitalization and less morbidity after radiosurgery

than after surgical removal make the procedure less

burdensome to patients. However, since radiosurgery does not

eliminate the tumor, it is imperative to know how effectively

irradiation prevents future growth of the tumor. Kondziolka et

al. address this issue by presenting data from patients who

were followed for at least five years after radiosurgery. In only

4 of their 162 patients was subsequent tumor growth sufficient

to require surgical removal; this recurrence rate is equal to or

better than that achieved in many surgical series.

This important study leaves several questions unanswered. The

treatment changed during the five years described in the report:

more than two thirds of the patients received doses of radiation

to the tumor edge that were higher than the 14 Gy that the

authors now use in the treatment of acoustic neuromas. In their

studies in animals, tumor regression measured three months

after irradiation was substantially less when the dose was 10

Gy than when it was 20 or 40 Gy, and there was no reduction in

the vascularity of the tumor at the lowest dose. (4) They state in

an earlier clinical study that control of tumor growth was

maintained with use of their current treatment dose of 14 Gy,

(5) although that study separated patients according to the

method of treatment planning (computed tomographic imaging

or MRI) rather than according to the radiation dose, and the

follow-up was shorter than two years for the group receiving

the lower dose and MRI-planned treatment.

In the current report, only 97 of 162 patients underwent

scanning more than five years after treatment, 20 patients were

lost to follow-up, and 32 patients either refused or otherwise

failed to undergo late MRI. These 32 patients were doing well

clinically, but a good clinical result can be expected even if

there is tumor recurrence. In the current report, we cannot

determine the duration of the actual radiologic follow-up

periods for the 46 patients treated with the 14-Gy dose. Recent

studies of acoustic tumors treated without surgery found no

growth in 26 to 83 percent of patients over one or two years of

follow-up. (6) Thus, some of the tumors in the series of

Kondziolka et al. probably would not have grown, even

without radiosurgery. Finally, we do not know the radiation

dose for the four patients who required surgery to manage

tumor growth after radiation.

Therefore, from the current study, we really do not know the

success of radiosurgery in controlling these tumors beyond a

few years. As the authors have stated, " Because slow-growing

tumors such as vestibular schwannomas often take years to

progress, any decrease in tumor control from using lower doses

could take many years to detect (possibly 5 to 10), while

decreased cranial neuropathy rates can be detected with 2

years of follow-up. " (5)

When tumors recur after radiosurgery, their surgical removal

without destruction of the facial or other cranial nerves is

substantially more difficult than it is when radiosurgery has not

been performed. (7,8,9) Radiosurgery conceivably may even

induce malignant transformation of the tumor or cause new

tumors, although this complication is apparently rare.

Irradiation of Schwann-cell tumors in type 1 neurofibromatosis

can lead to malignant transformation. (10) Typically, secondary

oncogenesis occurs well after the longest duration of follow-up

in the series reported by Kondziolka et al. External-beam

irradiation to the head can induce tumor growth at the base of

the skull (11) and triton tumors (malignant tumors with

rhabdoid features). (12) For instance, a case report documents

the presence and fatal expansion of a triton tumor within an

acoustic neuroma five years after radiosurgery. (9)

When radiotherapy is considered for a benign, surgically

curable tumor in a young patient, this risk of inducing a

secondary tumor must be seriously weighed. It will be decades

before the incidence of this complication is known.

External-beam irradiation can also cause intracranial arterial

occlusion, (13) although there are no reports to date of such

accelerated atherosclerosis after radiosurgery. The anterior

inferior cerebellar artery, which is the primary source of blood

supply to the lateral pons and upper medulla, lies right next to

the surface of acoustic neuromas.

Gamma-knife and linear-accelerator radiosurgery have

provided important new approaches to treatment for some

intracranial lesions and ultimately may prove to be valuable in

the treatment of acoustic neuromas. Although the report by

Kondziolka et al. lays the groundwork for determining the rate

of tumor control, the data presented cannot yet define just what

this rate is. Surgical removal can be incomplete and does not

guarantee protection against recurrence, but in the majority of

patients, surgical resection precludes the need for any further

treatment. One MRI scan is obtained three years

postoperatively to confirm curative resection. After stereotactic

irradiation, the tumor remains in situ and must be monitored

and imaged periodically for an as yet undetermined number of

years (and perhaps decades). Nonetheless, because morbidity

is lower and cranial-nerve function is as good after

radiosurgery as after surgical removal, it is imperative that

studies such as that by Kondziolka et al. continue.

Lawrence H. Pitts, M.D.

K. Jackler, M.D.

University of California, San Francisco

San Francisco, CA 94115

Table of Contents | Previous Article | Next Article

Copyright © 1998 by the Massachusetts Medical Society. All rights

reserved.

Marie Drew