Neurofibromatosis 2

[Bilateral Acoustic Neurofibromatosis, Central Neurofibromatosis, NF2, Neurofibromatosis Type II]

Clinical Diagnosis

Modifications to NIH consensus diagnostic criteria for NF2 have been suggested [Evans, Huson, Donnai, Neary, Blair, Newton et al 1992; Gutmann et al 1997] to enable earlier diagnosis of founders. The Evans et al clinical diagnostic criteria for NF2 have been found to substantially improve sensitivity without affecting specificity [Baser et al 2002]. According to the modified criteria, NF2 is diagnosed in individuals with one of the following:

• Bilateral vestibular schwannomas

• A first-degree relative with NF2 AND

  • Unilateral vestibular schwannoma

    OR

  • Any two of: meningioma, schwannoma, glioma, neurofibroma, posterior subcapsular lenticular opacities*

• Unilateral vestibular schwannoma AND any two of: meningioma, schwannoma, glioma, neurofibroma, posterior subcapsular lenticular opacities*

• Multiple meningiomas AND

  • Unilateral vestibular schwannoma

    OR

  • Any two of: schwannoma, glioma, neurofibroma, cataract

*"Any two of" = two individual tumors or cataract

Testing

Cytogenetic testing.  A variety of chromosome abnormalities can be associated with NF2.

• Large submicroscopic deletions of 10-600 kb that may encompass the NF2 gene are present in 10-16% of families [Zucman-Rossi et al 1998, Kluwe et al 2005]. These are not associated with mental retardation even if quite large. Most large deletions and, less commonly, duplications of single or greater number of exons can be detected by MLPA [Kluwe et al 2005].

• Larger deletions cause mental retardation and can cause congenital abnormalities [Barbi et al 2002].

• Ring chromosome 22 can also cause multiple meningiomas and vestibular schwannomas fulfilling NF2 diagnostic criteria [Tsilchorozidou et al 2004]. The NF2 locus itself is usually present within the ring but the ring itself is frequently lost as a result of instability.

• Chromosome translocations that split the NF2 gene in two have also been described as causing NF2 [Tsilchorozidou et al 2004].

Note: Gross chromosomal changes detectable on normal cytogenetic analysis are fairly uncommon.

Molecular Genetic Testing

GeneReviews designates a molecular genetic test as clinically available only if the test is listed in the GeneTests Laboratory Directory by either a US CLIA-licensed laboratory or a non-US clinical laboratory. GeneTests does not verify laboratory-submitted information or warrant any aspect of a laboratory's licensure or performance. Clinicians must communicate directly with the laboratories to verify information.—ED.

Gene.   NF2 is the only gene known to be associated with neurofibromatosis 2.

Molecular genetic testing: Clinical uses

• Confirmatory diagnostic testing

• Predictive testing (The main role of molecular genetic testing is in early detection of at-risk individuals — primarily children of classically affected individuals — for management reasons.)

• Prenatal diagnosis

• Preimplantation genetic diagnosis

Molecular genetic testing: Clinical methods

Sequence analysis.  The mutation detection rate in leukocyte DNA depends on which generation is tested in a family.

In 62/86 (71%) of families with NF2, sequence analysis identified a mutation in a member of the second generation (unpublished data). Sixteen of 86 (19%) had deletions or duplications detectable on MLPA; in eight (9%), no mutation was identified.

In simplex cases (i.e., a single occurrence in a family), the mutation detection rate is around 60%. About 25-30% of mutations are not detected as a result of mosaicism [Kluwe et al 2003, Moyhuddin et al 2003].

Mutation scanning detects disease-causing mutations in up to two-thirds of cases [Parry et al 1996]. When mutation scanning is combined with tests to identify deletions or duplications of single exons, the mutation detection rate approaches 72% in simplex cases and exceeds 90% for familial cases [Wallace et al 2004, Kluwe et al 2005]. Other studies have reported lower mutation detection rates, which may reflect the inclusion of some more mildly affected individuals with somatic mosaicism (see Interpretation of test results).

Deletion/duplication testing. A more systematic approach to deletion detection using tests that can detect whole exon deletions and duplications suggests that at least 10-15% of NF2 constitutional aberrations are deletions [Zucman-Rossi et al 1998, Wallace et al 2004, Kluwe et al 2005].

Linkage analysis.  Linkage analysis can be considered in families in which no disease-causing mutation is identified and at least two family members of different generations are affected. Linkage studies are based upon an accurate clinical diagnosis of NF2 in the affected family members and accurate understanding of genetic relationships in the family. Linkage analysis is dependent on the availability and willingness of family members to be tested. The markers used for linkage analysis of NF2 are highly informative and very tightly linked to the NF2 locus; thus, they can be used in more than 95% of families with NF2 with greater than 99% accuracy. Linkage testing is not available to families with a single affected individual, a situation that often occurs when an individual has a de novo gene mutation and no affected offspring.

Table 1 summarizes molecular genetic testing for this disorder.

Table 1. Molecular Genetic Testing Used in Neurofibromatosis 2

Test Methods	Mutations Detected	Mutation Detection Rate	Test Availability
Sequence analysis	NF2 sequence variation	75%	Clinical
Mutation scanning		~65%
Deletion/duplication testing	NF2 deletion/duplication	10-15%
Linkage analysis	NA	NA

Interpretation of test results.  For issues to consider in interpretation of sequence analysis results, click here.

Research has shown that as many as 25-30% of individuals with NF2 caused by a de novo mutation are mosaic for the mutation [Kluwe et al 2003, Moyhuddin et al 2003]. Recognition of individuals who have mosaicism for an NF2 disease-causing mutation can be problematic as these individuals:

• May not have bilateral vestibular schwannomas;

• May have normal molecular genetic testing of the NF2 gene in unaffected tissue, such as lymphocytes; thus, molecular genetic testing of tumor tissue may be necessary to establish the presence of somatic mosaicism [Mohyuddin et al 2002].

Testing Strategy for a Proband

One of two sample types is used:

• Leukocyte DNA from an individual who has an affected parent

• Tumor DNA from an individual representing a simplex case

Molecular genetic testing is performed in the following order:

• 1

  Testing for large deletions using a technique such as MLPA

• 2

  Sequence analysis of exons 1-15 (Mutations have never been described in exons 16-17.)

• 3

  Linkage analysis in the rare instances in which a mutation cannot be identified in families with at least two affected family members of different generations

When tumor DNA is tested both mutations must be identified.

• This may mean testing for loss of one allele by assessing for loss of heterozygosity (LOH).

• Once both mutations are identified in the tumor, leukocyte DNA can be tested to determine which one of the mutations is constitutional and which is present in the tumor only.

Testing Strategy for Relatives at Risk

• At-risk relatives whose genetic status is unknown can be tested for presence of the constitutional or mosaic mutation identified in an affected relative such as the proband.

• Children of an individual representing a simplex case in whom molecular genetic testing of a tumor has revealed loss of heterozygosity (LOH) can be reassured if testing of their leukocyte DNA shows that they have inherited the allele that was lost in the parental tumor, as this allele is unlikely to have a disease-causing mutation.

Genetically Related (Allelic) Disorders

No other phenotypes are known to be associated with mutations in the NF2 gene.

Natural History

The average age of onset of findings in individuals with NF2 is 18 to 24 years. The age of onset ranges from two to 70 years [Evans, Huson, Donnai, Neary, Blair, Newton et al 1992; Parry et al 1994]. Almost all affected persons develop bilateral vestibular schwannomas by the age of 30 years. In addition to vestibular schwannoma, individuals with NF2 develop schwannomas of other cranial and peripheral nerves, meningiomas, and, rarely, ependymomas and astrocytomas.

Variable expressivity of NF2 among individuals results in varying size, location, and number of tumors. Although these tumors are not malignant, their anatomical location and multiplicity lead to great morbidity and early mortality. The average age of death is 36 years. Actuarial survival from the time of establishing the correct diagnosis is 15 years. Survival is improving with earlier diagnosis and better treatment in specialty centers [Baser et al 2002].

Because NF2 is considered an adult-onset disease, it may be under-recognized in children, in whom skin tumors and ocular findings may be the first manifestations [Evans et al 1999, Ruggieri et al 2005]. The presenting symptoms of 120 individuals with NF2 studied by Evans, Huson, Donnai, Neary, Blair, Newton et al (1992) in Great Britain are listed in Table 2. (This study did not include skin tumors or cataracts as a first symptom.)

Vestibular schwannoma.  Initial symptoms include tinnitus, hearing loss, and balance dysfunction. Disability is usually insidious in onset, although occasionally hearing loss may occur suddenly, presumably as a result of vascular compromise by the tumor. Affected individuals often report difficulty in using the telephone in one ear, or unsteadiness when walking at night or on uneven ground.

With time, vestibular tumors extend medially into the cerebellar pontine angle and, if left untreated, cause compression of the brain stem and hydrocephalus. Significant facial palsy is rare even in large tumors.

Schwannomas may also develop on other cranial and peripheral nerves, with sensory nerves more frequently affected than motor nerves [Evans et al 1992].

Spinal tumors.  At least two-thirds of individuals with NF2 develop spinal tumors, which are often the most devastating and difficult aspects of this disease to manage [Parry et al 1994, Mautner et al 1995]. The most common spinal tumors are schwannomas, which usually originate within the intravertebral canal on the dorsal root and extend both medially and laterally, taking the shape of a "dumbbell". Intramedullary tumors of the spinal cord such as astrocytoma and ependymoma occur in 5-33% of individuals with NF2. Most individuals with spinal cord involvement have multiple tumors. Although multiple tumors are often present on imaging studies, many individuals with NF2 never develop symptoms from them.

Meningioma.  Approximately half of individuals with NF2 develop meningiomas [Evans, Huson, Donnai, Neary, Blair, Newton et al 1992; Parry et al 1994]. Most are intracranial; however, spinal meningiomas occur. NF2 meningiomas tend to occur less frequently in the skull base and are usually of the fibroblastic variety [Evans et al 2000, Kros et al 2001]. Meningiomas in the orbit may compress the optic nerve and result in visual loss. Those at the skull base may cause cranial neuropathy, brain stem compression, and hydrocephalus. Meningioma may be the presenting feature of NF2 particularly in childhood [Evans et al 1999, Perry et al 2001].

Ocular involvement.  One-third of individuals with NF2 have decreased visual acuity in one or both eyes [Ragge et al 1995]. Posterior subcapsular lens opacity rarely progressing to a visually significant cataract is the most common ocular finding. Lens opacities may appear prior to the onset of symptoms from vestibular schwannoma and can be seen in children. Retinal hamartoma and epiretinal membrane are seen in up to one-third of individuals. Intracranial and intra-orbital tumors may result in decreased visual acuity and diplopia. Rarely, other ocular manifestations may occur; hyperplastic persistent primary vitreous (HPPV) has been reported in a father and son [Nguyen et al 2005].

Intracranial and intra-orbital tumors may result in decreased visual acuity and diplopia.

Mono/polyneuropathy.  An increasingly recognized feature of NF2 is a mononeuropathy occurring particularly in childhood [Evans et al 1999] and frequently presenting as a facial palsy that usually only partially recovers, a squint (third nerve palsy) or a foot or hand drop. The foot drop may mimic polio.

A progressive polyneuropathy of adulthood not directly related to tumor masses is also being increasingly recognized [Sperfeld et al 2002].

Further evidence for the mononeuropathy of childhood and the polyneuropathy of adulthood has come from sural nerve biopsies [Hagel et al 2002].

Other.  Renal vascular disease similar to that occurring in NF1 has been reported once [Cordiero et al 2006].

Somatic mosaiciam for disease-causing mutations in the NF2 gene.   Mosaicism has been suspected in individuals with unilateral vestibular schwannoma and multiple other ipsilateral tumors [MacCollin et al 1997, Moyhuddin et al 2003]. This has now been confirmed for most cases in which tumor DNA from multiple tumors is available for analysis [Moyhuddin et al 2003, Wallace et al 2004, Aghi et al 2006].

Histopathology.  The tumors of NF2 are derived from Schwann cells, meningeal cells, and glial cells. They are uniformly benign. About 40% of NF2 vestibular tumors have a lobular pattern that is uncommon in tumors from individuals who have no known family history of NF2. NF2-associated vestibular schwannomas tend to be more invasive than non-NF2 tumors and to have a higher degree of dividing cells. Meningiomas from individuals with NF2 also have a higher degree of dividing cells. NF2 meningiomas are usually of the fibroblastic variety. No histologic differences have been observed between glial tumors in individuals with NF2 and individuals who do not have NF2.

Genotype-Phenotype Correlations

Intrafamilial variability is much lower than interfamilial variability, suggesting a strong effect of the underlying genotype on the resulting phenotype.

Unlike neurofibromatosis type 1 (NF1), large deletions of the NF2 gene have been associated with a mild phenotype [Baser et al 2004].

Nonsense and frame-shifting mutations have been associated with severe disease regardless of their position within the gene [Ruttledge et al 1996, Baser et al 2004].

Splice site mutations have been associated with both mild and severe disease [Ruttledge et al 1996, Kluwe et al 1998, Baser et al 2005], and may be milder if occurring in the 3' half of the gene [Baser et al 2005].

Missense mutations are usually mild, often causing the mildest form of NF2 [Evans, Trueman et al 1998; Baser et al 2002].

In general, truncating mutations show a significantly more severe pattern of tumors and are associated with greater disease-related mortality [Baser et al 2002, Baser et al 2005]. Truncating mutations are also associated with increased prevalence of spinal tumors [Patronas et al 2001, Dow et al 2005].

Somatic mosaicism for typical truncating mutations that would normally cause severe NF2 may also mitigate the effect, resulting in a milder phenotype [Evans, Trueman et al 1998]. The type of NF2 constitutional mutation is an important determinant of the number of NF2-associated intracranial meningiomas, spinal tumors, and peripheral nerve tumors [Baser et al 2004].

Penetrance

Penetrance is 100%. All individuals who have a pathogenic mutation develop the disease in an average lifetime. Age at onset can vary with mutation type as described above.

Anticipation

Although some reports suggested anticipation in NF2, it is likely that these were instances of milder disease associated with mosaicism for an NF2 mutation in the first generation and, more severe disease associated with an NF2 germline mutation in the second and subsequent generations.

Nomenclature

The term "neurofibromatosis" is a misnomer as the primary tumor types in NF2 are schwannoma and meningioma. Vestibular schwannoma (previously termed acoustic neuroma) was initially considered part of von Recklinghausen neurofibromatosis type 1, leading to multiple instances in which individuals with NF2 were included in series of individuals with NF1.

Since 1987 the great majority of reports have correctly distinguished between NF1 and NF2, with NF2 described as "bilateral acoustic" or "central" neurofibromatosis.

Prevalence

The incidence of NF2 was initially reported as 1:33,000-40,000 individuals. Disease prevalence is somewhat lower at 1:200,000. However, a recent update suggests that the incidence may be as high as 1:25,000 and the prevalence higher than 1:100,000 [Evans, Moran et al 2005].

NF2 has no ethnic or racial predilections.

Evaluations at Initial Diagnosis to Establish the Extent of Disease

• Head MRI

• Hearing evaluation, including BAER

• Ophthalmologic evaluation

• Cutaneous examination

Treatment of Manifestations

Vestibular schwannoma.  Untreated tumors may be slow growing and not require active intervention in the short term [Masuda et al 2004, Slattery et al 2004]. Therapy remains primarily surgical. Small vestibular tumors (<1.5 mm) that are completely intercanalicular can often be completely resected with preservation of both hearing and facial nerve function. Larger tumors are probably best managed expectantly with debulking or decompression carried out only when brain stem compression, deterioration of hearing, or facial nerve dysfunction occurs [Evans, Baser et al 2005].

Stereotactic radiosurgery, most commonly with the gamma knife, has been offered as an alternative to surgery in select individuals with vestibular schwannoma. However, the outcomes from radiation treatment in individuals with NF2 are not as good as for individuals with sporadic unilateral vestibular schwannoma, with only approximately 60% long-term tumor control [Rowe et al 2003].

Malignant transformation is a possible, but probably not common, sequelum [Baser et al 2000]; however, it should be noted that tumor development following radiation may take 15 years [Evans et al 2006].

Management of individuals with vestibular tumors should include counseling for insidious problems with balance and underwater disorientation, which can result in drowning [Kanter et al 1980].

Other tumors.  Other intracranial, cranial nerve, or spinal nerve tumors are very slow growing and surgical intervention for a tumor producing little impairment may cause disability years before it would occur naturally.

Although ependymoma in individuals without NF2 is optimally treated with complete resection, and occasionally with radiotherapy and chemotherapy, it is unclear whether ependymoma in individuals with NF2 warrants aggressive management.

Radiation therapy of NF2-associated tumors should be carefully considered since radiation exposure may induce, accelerate, or transform tumors in an individual with an inactive tumor suppressor gene, especially in childhood [Baser et al 2000, Evans et al 2006].

Hearing.  Hearing preservation and augmentation are important in the management of individuals with NF2. All affected individuals and their families should be referred to an audiologist to receive training in optimization of hearing and speech production. Hearing aids may be helpful early in the course of the disease [Evans, Baser et al 2005]. Auditory rehabilitation with a cochlear or brain stem implant should be discussed with those who have lost hearing [Evans, Baser et al 2005].

Rarely, individuals who have had vascular insult to the cochlea, but otherwise are without nerve damage, may benefit from a cochlear implant. Lip-reading skills may be enhanced by instruction, and sign language may often be more effectively acquired before the individual loses hearing.

Ocular involvement.  Early recognition and management of visual impairment from other manifestations of NF2 is extremely important. Routine complete eye examinations should be part of the care of all individuals with NF2.

Prevention of Secondary Complications

Prevention of substantial handicap from the disease can be achieved by appropriate expert treatment of tumors.

• A cervical spinal scan should be performed before cranial surgery to prevent complications from manipulation under anesthesia [Evans, Baser et al 2005].

• Spinal tumors may make epidural analgesia difficult; therefore, lumbosacral imaging should be performed before regional analgesia is given [Spiegel et al 2005; Sakai et al 2005].

Surveillance

For at-risk individuals who have either tested positive for the known disease-causing mutation in their family or whose genetic status cannot be clarified by molecular genetic testing:

• MRI is usually begun at age ten to 12 years but can be delayed in families in which the onset is known to be later [Evans, Baser et al 2005]. MRI should be continued on an annual basis until at least the fourth decade of life. It is not clear if earlier studies are beneficial and it is not known at what age monitoring can be safely stopped. Although some individuals with NF2 do not have symptoms until they are in their fifties, it is likely that "silent" tumors would be detected on an MRI performed at a younger age.

• Hearing evaluation, including BAER testing, may be useful in detecting changes in auditory nerve function before changes can be visualized by MRI.

Agents/Circumstances to Avoid

Radiotherapy should be avoided in children with NF2 [Evans et al 2006].

Testing of Relatives at Risk

Consideration of molecular genetic testing of at-risk family members (see Genetic Counseling) during childhood is appropriate for surveillance.

• Early identification of those who have inherited the family-specific NF2 mutation allows for appropriate screening using MRI for neuroimaging and brain stem auditory evoked response (BAER) testing for audiologic evaluation, thus resulting in earlier detection of disease manifestations and improved final outcome [Evans, Baser et al 2005].

• Early identification of those who have not inherited the family-specific NF2 mutation eliminates the need for costly screening with MRI and BAER testing.

Therapies Under Investigation

The search for an effective medical treatment for NF2-related tumors continues. It is possible that promising agents such as PAK-1 blocking drugs [Hirokawa et al 2004], could be tried on the NF2 mouse model in the first instance.

Search ClinicalTrials.gov for access to information on clinical studies for a wide range of diseases and conditions.

Mode of Inheritance

NF2 is inherited in an autosomal dominant manner.

Risk to Family Members

Parents of a proband

• About 50% of individuals with NF2 have an affected parent and 50% have NF2 as the result of a de novo gene mutation [Parry et al 1996]. However, 25-30% of those representing simplex cases (i.e., individuals with no family history of NF2) are mosaic for an NF2 mutation or deletion [Kluwe et al 2003, Moyhuddin et al 2003].

• Recommendations for the evaluation of parents of a proband with an apparent de novo mutation include a clinical history and, if any suspicion of NF2 exists, an MRI scan. A parent can be excluded as having NF2 if his/her offspring is shown to be mosaic, but absence of a mutation detected in the child does not eliminate the possibility of mosaicism in the parent. Because the age of onset of symptoms is consistent within families, it is usually not necessary to offer surveillance to asymptomatic parents.

Sibs of a proband

• The risk to the sibs of the proband depends upon the genetic status of the parents.

• If a parent of the proband is affected, the risk to the sibs is 50%.

• If neither parent of an individual with NF2 is symptomatic, the risk to the sibs of the affected individual is extremely low because the age of onset of symptoms is relatively uniform within families.

• However, a single case of germline mosaicism in a clinically normal individual has been reported [Parry et al 1996].

Offspring of a proband

Each child of an individual with NF2 has up to a 50% chance of inheriting the mutation.

• If the proband has other affected family members, each child of the proband has a 50% chance of inheriting the mutation.

• If the proband is the only affected individual in the family, two possibilities exist:

  • The proband may have somatic mosaicism for the disease-causing mutation or deletion. Offspring of an individual who is mosaic may have less than a 50% risk of inheriting the disease-causing mutation or deletion.

  • The proband may have a de novo germline mutation (i.e., present in the egg or sperm at the time of conception). Each offspring of an individual with a de novo germline mutation has a 50% chance of inheriting the mutation.

• Persons with somatic mosaicism and bilateral vestibular tumors have less than a 50% chance of having an affected child [Evans, Wallace et al 1998]. If the mutation is detected in DNA from multiple tumors, but not in DNA from leukocytes, the risk to offspring is probably less than 5%.

Other family members of a proband.  The risk to other family members depends upon the genetic status of the proband's parents. If a parent is found to be affected, his or her family members may be at risk, depending on the family structure.

Related Genetic Counseling Issues

Considerations in families with an apparent de novo mutation.  When the parents of a proband with an autosomal dominant condition are unaffected, possible non-medical explanations include alternate paternity or undisclosed adoption.

Testing of at-risk asymptomatic children and adults.  Consideration of molecular genetic testing of at-risk family members during childhood is appropriate for surveillance (see Management). Molecular genetic testing used in early identification of at-risk family members may either be mutation analysis or linkage analysis. Mutation analysis can only be used for testing of at-risk relatives if a disease-causing mutation has been identified in an affected family member. Linkage analysis is the preferred method of testing in families with more than one affected family member.

Because early detection of at-risk individuals affects medical management, testing of at-risk asymptomatic individuals during childhood is beneficial. Parents often want to know the genetic status of their children prior to initiating screening in order to avoid unnecessary procedures for a child who has not inherited the altered gene. Special consideration should be given to education of the children and their parents prior to genetic testing. A plan should be established for the manner in which results are to be given to the parents and children.

DNA banking.  DNA banking is the storage of DNA (typically extracted from white blood cells) for possible future use. Because it is likely that testing methodology and our understanding of genes, mutations, and diseases will improve in the future, consideration should be given to banking DNA of affected individuals. DNA banking is particularly relevant in situations in which the sensitivity of currently available testing is less than 100% or when linkage analysis is utilized. See DNA Banking for a list of laboratories offering this service.

Prenatal Testing

Prenatal diagnosis of pregnancies at 50% risk for NF2 is possible by analysis of DNA extracted from fetal cells obtained by amniocentesis usually performed at about 15-18 weeks' gestation or chorionic villus sampling (CVS) at about 10-12 weeks' gestation. The disease-causing allele of an affected family member must be identified or linkage established in the family before prenatal testing can be performed.

Note: Gestational age is expressed as menstrual weeks calculated either from the first day of the last normal menstrual period or by ultrasound measurements.

Preimplantation genetic diagnosis (PGD) may be available for families in which the disease-causing mutation has been identified in an affected family member in a research or clinical laboratory. For laboratories offering PGD, see .

Literature Cited

Published Statements and Policies Regarding Genetic Testing

No specific guidelines regarding genetic testing for this disorder have been developed.

Suggested Readings

Author History

D Gareth Evans, MD, FRCP (2004-present)
Mia MacCollin, MD; Harvard Medical School (1998-2004)

Initial posting: 14 October 1998

Revision History

• 6 June 2006 (me) Comprehensive update posted to live Web site

• 6 April 2004 (me) Comprehensive update posted to live Web site

• 29 October 2001 (me) Comprehensive update posted to live Web site

• 14 October 1998 (pb) Review posted to live Web site

• 5 August 1998 (mm) Original submission