Neurofibromatosis, Type Ii

A number sign (#) is used with this entry because neurofibromatosis type II (NF2) is caused by heterozygous mutation in the gene encoding neurofibromin-2 (NF2; 607379), which is also called merlin, on chromosome 22q12.

Description

The central or type II form of neurofibromatosis (NF2) is an autosomal dominant multiple neoplasia syndrome characterized by tumors of the eighth cranial nerve (usually bilateral), meningiomas of the brain, and schwannomas of the dorsal roots of the spinal cord. The incidence of neurofibromatosis type II is 1 in 25,000 live births (Asthagiri et al., 2009). NF2 has few of the hallmarks of the peripheral or type I form of neurofibromatosis (NF1; 162200), also known as von Recklinghausen disease.

Asthagiri et al. (2009) provided a detailed review of neurofibromatosis type II.

Clinical Features

Gardner and Frazier (1933) reported a family of 5 generations in which 38 members were deaf because of bilateral acoustic neuromas; of these, 15 later became blind. The average age at onset of deafness was 20 years. The average age at death of affected persons in the second generation was 72, in the third generation 63, in the fourth 42, and in the fifth 28. Follow-up of this family (Gardner and Turner, 1940; Young et al., 1970) revealed no evidence of the systemic manifestations of neurofibromatosis I (NF1; 162200), also known as von Recklinghausen disease. Other families with no evidence of the latter disease were reported by Worster-Drought et al. (1937), Feiling and Ward (1920), and Moyes (1968). Worster-Drought et al. (1937) pointed out that Wishart (1822) was the first to report a case of bilateral acoustic neuroma. Wishart's patient, Michael Blair, was 21 years old when he consulted Mr. Wishart, president of the Royal College of Surgeons of Edinburgh, because of bilateral deafness. He had a peculiarly shaped head from infancy, and blindness in the right eye was discovered at about 4 months after birth. He became completely blind and deaf toward the end of his life. Autopsy revealed tumors of the dura mater and brain and also a 'tumour of the size of a small nut, and very hard, being attached to each of them (auditory nerves), just where they enter the meatus auditorius internus.'

Nager (1969) showed that in about 4% of cases acoustic neuroma is bilateral. In addition to their autosomal dominant inheritance and association with neurofibromatosis, bilateral tumors differ from unilateral ones in that they can reach a remarkably large size with extensive involvement of the temporal bone and the nerves therein. Fabricant et al. (1979) reported that more than 30 kindreds with 'central neurofibromatosis' had been described. Most patients with the central form (NF2) have no cafe-au-lait spots or peripheral neurofibromata, and no patients in one large series had 6 or more cafe-au-lait spots (Eldridge, 1981).

Kanter et al. (1980), who reviewed 9 personally studied kindreds and 15 reported ones, with a total of 130 cases, showed an increase only in antigenic activity of nerve growth factor (NGF; 162030) in central neurofibromatosis and only in functional activity in peripheral neurofibromatosis.

In a series reported by Mrazek et al. (1988), 1 of 41 acoustic neurinoma cases was bilateral. This was in a 10-year-old girl with von Recklinghausen neurofibromatosis, whose first tumor had been diagnosed at age 6.

Mayfrank et al. (1990) studied 10 patients with NF2 and found that all were sporadic cases, each presumably the result of a new mutational event. From a survey of these patients and those in the literature, they concluded that sporadic cases are characterized by a high incidence of multiple meningiomas and spinal tumors in addition to bilateral acoustic neurinomas.

Pulst et al. (1991) described a family with spinal neurofibromatosis without cafe-au-lait spots or other manifestations of either NF1 or NF2 such as cutaneous tumors, Lisch nodules, or acoustic tumors. Mutation at the NF1 locus was excluded with odds greater than 100,000:1. Markers with the NF2 locus were uninformative in this family.

Evans et al. (1992, 1992) studied 150 patients. The mean age at onset was 21.57 years (n = 110) and no patient presented after 55 years of age. Patients presented with symptoms attributable to vestibular schwannomas (acoustic neuroma), cranial meningiomas, and spinal tumors. In 100 patients studied personally by the authors, 44 presented with deafness, which was unilateral in 35. Deafness was accompanied by tinnitus in 10. Muscle weakness or wasting was the first symptom in 12%. In 3 of the 100 patients, there was a distal symmetrical sensorimotor neuropathy, confirmed by nerve conduction studies and electromyography. Although similar features may result from the multiple spinal and intracranial tumors that occur in this condition, a generalized and isolated neuropathy appears to be a relatively common feature of NF2. Cafe-au-lait spots occurred in 43 of the 100 patients but only 1 had as many as 6 spots. Cataract was detected in 34 of 90 patients. Cataracts were probably congenital in 4 patients in this study. Three types of skin tumors were recognized. The first and least common was similar to the intradermal papillary skin neurofibroma with violaceous coloring occurring in NF1. The second type comprised subcutaneous well-circumscribed, often spherical, tumors that appeared to be located on peripheral nerves; the thickened nerve could often be palpated at either end of the tumor, the skin being mobile and separate from the tumor. The third and most frequent type, first described by Martuza and Eldridge (1988), was represented by discrete well-circumscribed, slightly raised, roughened areas of skin often pigmented and accompanied by excess hair. Skin tumors of some kind were found in 68% of patients, type 1 being present in 20%, type 2 in 33%, and type 3 in 47%. They could find no evidence that either pregnancy or contraceptive pills has adverse effects on vestibular schwannomas or other manifestations. Evans et al. (1992) provided useful advice on the follow-up of persons identified as having NF2 and the management of persons at risk of developing NF2.

Evans et al. (1992) divided their 120 cases of NF2 into 2 types: the Wishart (1822) type, with early onset, rapid course, and multiple other tumors in addition to bilateral vestibular schwannomas, and the Gardner type (1930, 1933, 1940), with late onset, more benign course, and usually only bilateral vestibular schwannomas. This classification had been suggested by Eldridge et al. (1991). Evans et al. (1992) found no evidence for the existence of a third type of generalized meningiomatosis that might be designated the Lee-Abbott type (Lee and Abbott, 1969). The age at onset of deafness and the age at diagnosis were almost identical in the 2 sexes. Birth incidence of NF2 was estimated to be 1 in 33,000-40,562. Evans et al. (1992) considered 49% of the 150 cases to represent new mutations. The mutation rate was estimated to be 6.5 x 10(-6). A maternal effect on severity was noted in that age of onset was 18.17 years in 36 maternally inherited cases and 24.5 years in 20 paternally inherited cases (p = 0.027). A preponderance of maternally inherited cases was also significant (p = 0.03). (A maternal effect on severity had been noted also for NF1.) Baser et al. (2001) studied 140 patients and found that maternal inheritance was not an independent correlate of NF2 disease severity.

Parry et al. (1994) assessed possible heterogeneity in NF2 by evaluating 63 affected members of 32 families. In addition to skin and neurologic examinations, workup included audiometry, complete ophthalmologic examination with slit-lamp biomicroscopy of the lens and fundus, and gadolinium-enhanced MRI of the brain and, in some, of the spine. Mean age-at-onset in 58 individuals was 20.3 years; initial symptoms were related to vestibular schwannomas (44.4%), other CNS tumors (22.2%), skin tumors (12.7%), and ocular manifestations including cataracts and retinal hamartomas (12.7%). Screening uncovered 5 affected but asymptomatic family members; vestibular schwannomas were demonstrated in 62 (98.4%). Other findings included cataracts (81.0%), skin tumors (67.7%), spinal tumors (67.4%), and meningiomas (49.2%). As a rule, clinical manifestations and clinical course were similar within families but differed among families. Parry et al. (1994) concluded that 2 subtypes but not 3 can be defined.

Evans et al. (1999) studied the presentation of NF2 in childhood. A total of 334 cases of NF2 were identified from a comprehensive UK dataset, of which 61 (18%) had presented in childhood (0-15 years). Twenty-six of these children presented with symptoms of vestibular schwannoma, 19 with meningioma, 7 with a spinal tumor, and 5 with a cutaneous tumor. In addition, Evans et al. (1999) identified 22 children with a meningioma from the Manchester Children's Tumor Registry, a prospective database of children presenting with a tumor since 1954 within a defined population. At least 3 of these children subsequently developed classic NF2, and in none of them was there a family history suggestive of NF2. The authors concluded that NF2 should be considered in any child presenting with meningioma, vestibular schwannoma, or cutaneous symptoms such as neurofibroma or schwannoma, especially if they have fewer than 6 cafe-au-lait patches and therefore do not fulfill the diagnostic criteria for NF1.

Gijtenbeek et al. (2001) reported a patient with NF2, confirmed by genetic analysis, who presented with an axonal mononeuropathy multiplex with progression of axonal loss over several years. Sural nerve biopsy showed small scattered groups of Schwann cells transformed into irregular branching cells with abnormal cell-cell contacts. The authors hypothesized that defective Schwann cell function, due to inactivation of the NF2 gene product merlin, leads to changes in morphology, cell-cell contact, and growth, and finally to degeneration of axons.

Egan et al. (2001) reported 4 cases of NF2 with a monocular elevator paresis. Two of the patients had third nerve tumors demonstrable on MRI, which had not been present on earlier films. The other 2 patients may have had tumors too small for radiographic detection. The authors suggested that the isolated paresis may result from compression of particular fascicles of the third nerve that subserve the superior rectus and inferior oblique muscles as they exit the midbrain, and noted that ocular mobility defects should be closely monitored in patients with NF2.

To evaluate clinical and molecular predictors of the risk of mortality in persons with NF2, Baser et al. (2002) analyzed the mortality experience of 368 patients from 261 families in the United Kingdom NF2 registry. Age at diagnosis, intracranial meningiomas, and type of treatment center were informative predictors of the risk of mortality. The relative risk of mortality increased 1.13-fold per year decrease in age at diagnosis and was 2.51-fold greater in people with meningiomas compared with those without meningiomas. The relative risk of mortality in patients treated at specialty centers was 0.34, compared with those treated at nonspecialty centers. The relative risk of mortality in people with constitutional NF2 missense mutations was very low compared with those with other types of mutations (nonsense, frameshift, or splice site mutations, and large deletions), but the confidence interval could not be quantified because there was only 1 death among people with missense mutations.

Ocular Abnormalities

Pearson-Webb et al. (1986) pointed out that Lisch nodules, which are iris hamartomas that are frequently found in NF1, are not found in NF2. They found, however, an apparently high frequency of presenile posterior subcapsular and nuclear cataracts which sometimes required surgery and/or predated the symptoms of bilateral acoustic neurofibromatosis. Landau et al. (1990) described combined pigment epithelial and retinal hamartoma (CEPRH) in NF2.

Kaiser-Kupfer et al. (1989) found posterior capsular lens opacities in 20 NF2 patients in 11 families. Parry et al. (1991) extended these observations. In 26 persons who were first-degree relatives of an affected individual, they found posterior capsular cataracts in 21. Of 14 at-risk individuals, i.e., persons with mild changes of NF but not NF1, persons under age 40 with unilateral acoustic neuroma, a child with meningioma and/or schwannoma, and a person with multiple meningioma, they found posterior capsular lens opacities in 13. These patients probably represented new mutations. The presence of posterior capsular opacities in a relative of persons with NF2 was suggestive of NF2. Furthermore, NF2 should be considered in young persons without NF1 but with mild skin findings of NF or CNS tumors with posterior capsular opacities. Bouzas et al. (1993) found posterior subcapsular/capsular cataracts in 36 (80%) of 45 affected individuals in 29 families. In addition, the association of peripheral cortical lens opacities with NF2 was found to be statistically significant: such cataracts were found in 17 of the patients (37.8%) but in none of the unaffected family members (p less than 0.0001). In 3 patients, peripheral cortical opacities were present despite the absence of posterior subcapsular/capsular cataracts. Bouzas et al. (1993), reporting further on the NIH experience, reviewed visual impairment in 54 NF2 patients, 51 of whom had bilateral vestibular schwannomas. Causes of decreased vision were cataracts, damage in the optic pathways, macular hamartomas, and corneal opacities. Although lens opacities are an important marker for NF2, they usually do not interfere with vision; some progress, requiring cataract extraction. In 6 patients, decreased visual acuity was due to corneal opacifications secondary to either seventh or fifth cranial nerve damage, or both. Damage to the seventh cranial nerve caused lagophthalmos and decreased lacrimal secretion; damage to the fifth cranial nerve caused corneal hypesthesia. The nerves were damaged by the growth of vestibular tumors in 1 patient, but in most patients they were damaged during neurosurgical procedures.

Ragge et al. (1995) concluded that the most common ocular abnormalities in NF2 are posterior subcapsular or capsular, cortical, or mixed lens opacities, found in 33 of 49 patients (67%), and retinal hamartomas found in 11 of 49 patients (22%). The types of cataract that were most suggestive of NF2 were plaque-like posterior subcapsular or capsular cataract and cortical cataract with onset under the age of 30 years.

Baser et al. (2003) confirmed the high prevalence of cataracts in young NF2 patients. They suggested that the frequent occurrence of cataracts before the tumor manifestations of NF2 indicated the usefulness of this non-eighth nerve feature in the diagnosis of NF2 in children and adolescents.

McLaughlin et al. (2007) identified 3 types of NF2-associated ocular manifestations: juvenile posterior subcapsular cataract, epiretinal membrane, and intrascleral schwannoma. Their histopathologic analysis revealed that dysplastic lens cells accumulated just anterior to the posterior lens capsule in juvenile posterior subcapsular cataract, and that dysplastic Muller cells might be a major component of NF2-associated epiretinal membrane. McLaughlin et al. (2007) concluded that their findings suggested that a subset of glial cells with epithelial features (Schwann cells, ependymal cells, and Muller cells) might be particularly sensitive to loss of the NF2 gene.

Diagnosis

In a review of NF2, Martuza and Eldridge (1988) defined criteria for the diagnosis of both NF1 and NF2. An NIH Consensus Development Conference (1988) concluded that the criteria for NF2 are met if a person is found to have '(1) bilateral eighth nerve masses seen with appropriate imaging techniques (e.g., CT or MRI); or (2) a first-degree relative with NF2 and either unilateral eighth nerve mass, or two of the following: neurofibroma, meningioma, glioma, schwannoma, or juvenile posterior subcapsular lenticular opacity.' Pastores et al. (1991) demonstrated that small (less than 8 mm) acoustic neuromas can be detected in asymptomatic individuals by the use of gadolinium-enhanced MRI. They demonstrated such neuromas in 2 asymptomatic children, aged 7 and 11 years, one of whom had normal audiometric and brainstem-evoked response testing.

Using polymorphic DNA markers in a study of 13 NF2 kindreds, Ruttledge et al. (1993) concluded that it is possible to determine, with a high degree of certainty, the carrier status of about 85% of persons at risk. Risk prediction was possible in every case in which DNA was available from both parents. In 76% of informative individuals, it was possible to assign a decreased risk of being carriers. Thus, the use of probes for construction of chromosome 22 haplotypes for risk assessment should result in a greatly reduced number of individuals who will require periodic screening.

Gutmann et al. (1997) provided guidelines for the diagnostic evaluation and multidisciplinary management of both NF1 and NF2. The criteria for definite NF2 were bilateral vestibular schwannomas; or family history of NF2 in 1 or more first-degree relative(s) plus (a) unilateral vestibular schwannomas at age less than 30 years, or (b) any two of the following: meningioma, glioma, schwannoma, or juvenile posterior subcapsular lenticular opacities/juvenile cortical cataract. The criteria for presumptive or probable NF2 was unilateral vestibular schwannomas at age less than 30 years, plus at least one of the following: meningioma, glioma, schwannoma, or juvenile posterior subcapsular lenticular opacities/juvenile cortical cataract; or multiple meningiomas (two or more) plus (a) unilateral vestibular schwannomas at age less than 30 years, or (b) one of the following: glioma, schwannoma, or juvenile posterior subcapsular lenticular opacities/juvenile cortical cataract.

Kluwe et al. (2000) studied 40 skin tumors (36 schwannomas and 4 neurofibromas) from 20 NF2 patients, 15 of whom had NF2 mutations previously identified in blood leukocytes. The detection rate of constitutional mutations was higher in patients with skin tumors (65%) than in patients without skin tumors (40%). Alterations in both NF2 alleles were found in 17 (43%) of the tumors. They concluded that loss of a functional NF2 gene product is a critical event in the generation of skin schwannomas and that mutation detection in skin tumors may be a useful diagnostic tool in patients with skin tumors where the clinical diagnosis of NF2 is ambiguous, or in unclear cases in which NF1 must be excluded.

Baser et al. (2002) evaluated 4 previous sets of clinical diagnostic criteria for NF2 developed by groups of experts: the NIH Consensus Development Conference (1988), the Consensus Development Panel (1994) of the NIH, the Manchester Group criteria reported by Evans et al. (1992), and the National Neurofibromatosis Foundation (NNFF) criteria reported by Gutmann et al. (1997). Baser et al. (2002) concluded that none of the existing sets of criteria was adequate at initial assessment for diagnosing people who present without bilateral vestibular schwannomas, particularly people with a negative family history of NF2.

Baser et al. (2011) empirically developed and tested an improved set of diagnostic criteria that used understanding of the natural history and genetic characteristics of NF2 to increase sensitivity while maintaining very high specificity. They used data from the UK Neurofibromatosis 2 Registry and Kaplan-Meier curves to estimate frequencies of clinical features at various ages among patients with or without unequivocal NF2. On the basis of this analysis, Baser et al. (2011) developed the Baser criteria, a diagnostic system that incorporates genetic testing and gives more weight to the most characteristic features and to those that occur before 30 years of age. In an independent validation subset of patients with unequivocal NF2, the Baser criteria increased diagnostic sensitivity to 79% (9-15% greater than previous sets of criteria) while maintaining 100% specificity at the age of onset of the first characteristic sign of NF2.

Mosaicism in NF2

Evans et al. (2007) showed that the chances of a de novo patient with NF2 being mosaic for the underlying mutation in the NF2 gene increased with age at presentation with vestibular schwannoma and was particularly high in patients with unilateral presentation of vestibular schwannoma, but who still had at least 2 further NF2-related tumors in order to fulfill the Manchester criteria.

Evans and Wallace (2009) analyzed the mosaic risk in de novo patients with NF2 by age at the time of vestibular schwannoma diagnosis. They analyzed this risk in 4 age cohorts to derive figures for mosaicism and offspring risk both before and after lymphocyte DNA testing with sequencing and multiple ligation-dependent probe amplification. The study was based on actual genetic testing of lymphocyte DNA in 402 de novo patients and subsequent tumor testing in 51 patients with negative blood analysis. The risk of NF2 to an offspring of a patient presenting with bilateral vestibular schwannoma at less than 20 years of age was 29.3%, whereas the offspring risk for a patient presenting with asymmetric disease after 40 years of age was only 5.5%, as there is a 99% chance that they are mosaic.

Clinical Management

Stereotactic radiosurgery is the principal alternative to microsurgical resection for acoustic neuromas. The goals of radiosurgery are the long-term prevention of tumor growth, maintenance of neurologic function, and prevention of new neurologic deficits. Kondziolka et al. (1998) evaluated 162 consecutive patients who underwent radiosurgery for acoustic neuromas between 1987 and 1992, surveying the results between 5 and 10 years after the procedure. Resection had been performed previously in 42 patients; in 13 patients, the tumor represented a recurrence of disease after a previous total resection. The rate of tumor control (with no resection required) was 98%. Radiosurgery was believed to have been successful by all 30 patients who had undergone surgery previously and by 81 (95%) of the 85 who had not. Pitts and Jackler (1998) pointed out that when radiotherapy is considered for a benign, surgically curable tumor in a young patient, the risk of inducing a secondary tumor must be seriously weighed. The risk of intracranial arterial occlusion from external-beam irradiation must also be considered, although there had been no reports of 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.

Mapping

Seizinger et al. (1986) found loss of genes on chromosome 22 in acoustic neuromas; i.e., whereas normal tissue was heterozygous, tumor tissue was hemizygous (or homozygous) for the polymorphic markers SIS (190040), IGLC (147220), and the anonymous DNA locus D22S1. They were prompted to undertake the study by analogy to retinoblastoma and Wilms tumor and by the facts that meningioma occurs in association with familial acoustic neuroma and that cytologic change in chromosome 22 is frequent in meningioma (see 607174). Seizinger et al. (1987) found specific loss of alleles from chromosome 22 in 2 acoustic neuromas, 2 neurofibromas, and 1 meningioma from patients with bilateral acoustic neurofibromatosis. In each case, a partial deletion occurred with a breakpoint distal to the D22S9 locus in band 22q11. Wertelecki et al. (1988) confirmed localization of the causative gene on chromosome 22 (22q11.21-q13.1) by demonstration of linkage in family studies to markers on chromosome 22. Wertelecki et al. (1988) also presented the clinical data on 15 affected male and 8 affected female members of the 1 large kindred they studied for linkage data.

Rouleau et al. (1990) identified markers on chromosome 22 bracketing the NF2 gene which are therefore useful for accurate presymptomatic and prenatal diagnosis, as well as for isolating the defective gene. Through linkage analysis on 12 families with NF2, Narod et al. (1992) confirmed the assignment of the NF2 gene to chromosome 22 and concluded that there is no evidence of genetic heterogeneity in NF2. They indicated that the presence of bilateral vestibular schwannomas, as they termed the acoustic neuromas, is sufficient for the diagnosis.

Using 8 polymorphic loci on chromosome 22 to study tumor and constitutional DNAs isolated from 39 unrelated patients with sporadic or NF2-associated acoustic neuromas, meningiomas, schwannomas, and ependymomas, Wolff et al. (1992) found 2 tumors with loss of heterozygosity (LOH) patterns consistent with the presence of chromosome 22 terminal deletions. By use of additional polymorphic markers, the terminal deletion breakpoint in one of the tumors, an acoustic neuroma from an NF2 patient, was mapped within the previously defined NF2 region. In addition, they identified a sporadic acoustic neuroma with an LOH pattern consistent with mitotic recombination or deletion and reduplication. The findings lent further support to the recessive tumor suppressor model for the NF2 gene. Arai et al. (1992) described a patient with bilateral acoustic neurinomas and other tumors in the central nervous system and a constitutional translocation t(4;22)(q12;q12.2). Thus, 22q12.2 is a refined localization for the NF2 gene. The same karyotype that was seen in cultured peripheral lymphocytes was found in a paraspinal neurinoma. The patient's father was also a carrier of the translocation but he had no clinical symptoms of NF2, nor did other relatives. As explanation for the failure of expression in the father, Arai et al. (1992) suggested various possibilities including nonpenetrance, mosaicism, or genetic imprinting. They quoted Kanter et al. (1980) as demonstrating earlier onset of symptoms when NF2 is transmitted by the mother. Bovie et al. (2003) also reported a case of neurofibromatosis 2 in a patient with a balanced X;22 translocation. The patient presented with a large abdominal schwannoma and intellectual disability. A clinical diagnosis of NF2 was made when bilateral vestibular schwannomas were found on MRI. With demonstration of a de novo balanced reciprocal translocation between chromosome X and 22, the disorder in this patient was initially assumed to have been caused by the loss of NF2 at the translocation breakpoint. This was found, however, not to be the case; the breakpoint was 6 Mb centromeric to the NF2 gene and no mutations or deletions were found in the germline NF2 gene of the patient. The X-inactivation pattern in lymphocytes was 100% skewed to inactivate the normal X chromosome as predicted for X;autosome translocations whereas in tumor tissue there was aberrant X inactivation of the opposite derivative X chromosome. The mechanism of the disease in this case was thought to be that a proportion of Schwann cells had 1 NF2 allele acting as a functional null by virtue of NF2 being translocated to the X chromosome and aberrant X inactivation of the X;autosome.

Molecular Genetics

Rouleau et al. (1993) provided incontrovertible evidence that the NF2 gene (607379) is the site of the mutations causing neurofibromatosis II by demonstrating germline and somatic SCH mutations in NF2 patients and in NF2-related tumors. For description of the mutations identified in the NF2 gene and for a discussion of somatic mosaicism, see 607379.

Wu et al. (1998) identified 15 patients from a series of 537 with unilateral vestibular schwannomas who also had 1 or more of the following: other tumors (10 of 15), features of NF2 (3 of 15), or a family history of neurogenic tumors (5 of 15). No germline NF2 mutations were detected, and in 7 of 9 cases where tumor material was available for analysis, a germline mutation in NF2 was excluded. Wu et al. (1998) concluded that most instances of unilateral vestibular schwannoma which do not fulfill criteria for NF2 represent chance occurrences.

Baser et al. (2002) reported a patient with NF2 who developed malignant mesothelioma after a long occupational exposure to asbestos. Genetic analysis of the tumor tissue showed loss not only of chromosome 22 but also of chromosomes 14 and 15, and gain of chromosome 7. Baser et al. (2002) suggested that an individual with a constitutional mutation of an NF2 allele, as in NF2, is more susceptible to mesothelioma. Although mesothelioma is not a common feature in NF2, the authors cited the observation of Knudson (1995) that somatic mutations of a tumor suppressor gene, such as NF2, RB1 (614041), or p53 (191170), can be common in a tumor type that is not characteristic of the hereditary disorder, perhaps due to the proliferative timing of the cells involved.

In a family with the mild or so-called Gardner type of neurofibromatosis type II, Watson et al. (1993) defined a submicroscopic deletion on chromosome 22q which involved the neurofilament heavy chain locus (NEFH; 162230) but did not extend as far as the Ewing sarcoma region (EWSR1; 133450) proximally or the leukemia inhibitory factor locus (LIF; 159540) distally. They estimated that the deletion was about 700 kb long.

Mohyuddin et al. (2002) identified 45 patients aged 30 years or less at the onset of symptoms of unilateral vestibular schwannoma. Molecular genetic analysis of the NF2 gene was performed in all 45 patients and on 28 tumor samples. No pathogenic NF2 mutations were identified in any of the blood samples. NF2 point mutations were identified in 21 of 28 (75%) tumor samples and LOH in 21 of 28 (75%) tumor samples. Overlap, i.e., both mutational hits, were identified in 18 of 28 (65%) tumor samples. They observed 1 multilobular tumor in which 1 (presumably first hit) mutation was confirmed which was common to different foci of the tumor, while the second mutational event differed between foci. The molecular findings in this patient were consistent with somatic mosaicism for NF2 and a clinical diagnosis was confirmed with the presence of 2 meningiomas on a follow-up MRI scan.

Tsilchorozidou et al. (2004) reported 5 NF2 patients with constitutional rearrangements of chromosome 22 and vestibular schwannomas, multiple intracranial meningiomas, and spinal tumors. The authors noted that an additional 10 NF2 patients with constitutional NF2 deletions had been discovered using NF2 FISH in their laboratory, and suggested that chromosome analysis with FISH might be a useful first screen prior to molecular testing in NF2 patients.

Genotype/Phenotype Correlations

Parry et al. (1996) identified mutations in the NF2 gene in 66% of 32 patients; 20 different mutations were found in 21 patients. They suggested that their results confirmed the association between nonsense and frameshift mutations and clinical manifestations compatible with severe disease. They stated that their data raised questions regarding the role of other factors, in addition to the intrinsic properties of individual mutations, that might influence the phenotype. Ruttledge et al. (1996) reported that when individuals harboring protein-truncating mutations are compared with patients having single codon alterations, a significant correlation (p less than 0.001) with clinical outcome is observed. They noted that 24 of 28 patients with mutations that cause premature termination of the NF2 protein presented with severe phenotypes. In contrast, all 16 cases from 3 families with mutations that affect only a single amino acid had mild NF2.

Evans et al. (1998) reported 42 cases of NF2 from 38 families with truncating mutations. The average age of onset of symptoms was 19 years and age at diagnosis 22.4 years. Fifty-one cases from 16 families (15 with splice site mutations, 18 with missense mutations, and 18 with large deletions) had an average age of onset of 27.8 years and age at diagnosis of 33.4 years. Subjects with truncating mutations were significantly more likely to develop symptoms before 20 years of age (p less than 0.001) and to develop at least 2 symptomatic CNS tumors in addition to vestibular schwannoma before 30 years (p less than 0.001). There were significantly fewer multigenerational families with truncating mutations.

Kehrer-Sawatzki et al. (1997) reported a patient with NF2 and a ring chromosome 22 (46,XX,r(22)/45,XX,-22). Severe manifestations included multiple meningiomas, spinal and peripheral neurinomas, and bilateral vestibular schwannomas. The patient was also severely mentally retarded, a feature not usually associated with NF2. The authors hypothesized that a mutation in the NF2 gene of the normal chromosome 22, in addition to the loss of the ring 22 in many cells during mitosis, could explain the presence of multiple tumors. Using a meningioma cell line lacking the ring chromosome, Kehrer-Sawatzki et al. (1997) searched for deletions, rearrangements, or other mutations of the NF2 gene on the normal chromosome 22; no such alterations were found. The authors concluded that the loss of the entire chromosome 22 and its multiple tumor suppressor genes may have led to the severe phenotype in this patient.

In 406 patients from the population-based United Kingdom NF2 registry, Baser et al. (2004) evaluated genotype/phenotype correlations for various types of non-VIII nerve tumors using regression models with the additional covariates of current age and type of treatment center (specialty or nonspecialty). The models also permitted consideration of intrafamilial correlation. The authors found statistically significant genotype/phenotype correlations for intracranial meningiomas, spinal tumors, and peripheral nerve tumors. People with constitutional NF2 missense mutations, splice site mutations, large deletions, or somatic mosaicism had significantly fewer tumors than did people with constitutional nonsense or frameshift NF2 mutations. In addition, there were significant intrafamilial correlations for intracranial meningiomas and spinal tumors, after adjustment for the type of constitutional NF2 mutation. Baser et al. (2004) concluded that the type of constitution NF2 mutation is an important determinant of the number of NF2-associated intracranial meningiomas, spinal tumors, and peripheral nerve tumors.

In 831 patients from 528 NF2 families, Baser et al. (2005) analyzed location of splice site mutations and severity of NF2, using age at onset of symptoms and number of intracranial meningiomas as indicators. They found that individuals with splice site mutations in exons 1 to 5 had more severe disease than those with splice site mutations in exons 11 to 15. Baser et al. (2005) confirmed the previously reported observation that missense mutations are usually associated with mild NF2.

History

Baser et al. (2004) noted that initial genotype/phenotype correlation studies of NF2 were limited by the generality of the definition of disease severity, which was often reported only as 'mild,' 'moderate,' or 'severe.' The mild and severe disease categories corresponded to the historical nomenclature of 'Gardner' (mild) and 'Wishart' (severe) subtypes, which were based on the clinical observation that the severity of NF2 tended to 'run true' within a family (Wishart, 1822; Gardner and Frazier, 1930). Another category, 'Lee-Abbott' (Lee and Abbott, 1969), which corresponds to very severe NF2, was not consistently adopted by subsequent studies.

Cytogenetics

Krone and Hogemann (1986) found monosomy 22 as a predominant numerical anomaly in cultured cells grown from peripheral neurofibromas in patients described simply as suffering 'from sporadic peripheral NF.' Duncan et al. (1987) observed a ring chromosome 22 in a man with an atypical form of neurofibromatosis. He lacked a family history of NF, cafe-au-lait spots, and axillary freckling. He had multiple neurofibromas and a plexiform neuroma. By in situ hybridization, Duncan et al. (1987) showed that both the normal chromosome 22 and the ring chromosome 22 carried this gene.