Neurofibromatosis Type Two
by Sally Petrella

In a large Italian-American family first studied in 1992 (Narod 487), seven young adult cousins were seen at a clinic for hearing impairment and brown skin marks. The brown skin marks or "cafe au lait" spots are characteristic of an inherited dominant disorder called neurofibromatosis (NF) that predisposes individuals to develop swellings in the nerves called neurofibromas (Clayman 722-723). This family has a history of deafness and brain tumors: six of the seven cousins have or had a parent affected by one or both and they all shared a grandfather who was deaf and died from a brain tumor.

There are two types of neurofibromatosis: one affecting peripheral nerves and one affecting the central nervous system. Neurofibromatosis type one (NF1), also known as vonRecklinghausen or Peripheral NF is the most common neurofibromatosis (1/3000) and is characterized by skin marks, optic nerve gliomas and bony abnormalities (Rouleau 575). The less common (1/37,000) neurofibromatosis type two (NF2) or Central Bilateral Acoustic NF is associated with central nervous system tumors especially affecting both of the auditory nerves, as well as brain and spinal cord tumors and posterior capsular lens opacities (Bourn 69). Bilateral vestibular schwannomas (BVS) or unilateral vestibular schwannomas in a patient or in a patient's first degree relative are the hallmark of NF2 and are sufficient for diagnosis (Narod 487).

Small tumors in the fatty schwann cells covering the auditory nerves were found in the cousins of this family. This information combined with the cafe au lait skin marks and the family history of hearing impairment and brain tumors led doctors to suspect this family had familial NF2.

In 1987, linkage studies confirmed the distinction between NF1 and NF2 when NF1 was mapped to chromosome 17 and NF2 to chromosome 22. The NF2 gene was isolated in 1993 by several groups of researchers through the use of positional cloning and genetic linkage and deletion mapping analyses of affected individuals. A CA repeat polymorphism associated with the mutant allele was often used as a marker for the gene (Narod 486-487).

The NF2 gene is a tumor suppressor gene and encodes a 595-amino acid protein called merlin or schwannomin. Merlin is an acronym for moesin-ezrin-radixin-like protein, a family of proteins that play a role in cell surface dynamics and structure by linking the cytoskeleton to components of the plasma membrane (Bianchi 10854). In NF2, there is a nonsense or splice site mutation or frameshift deletions and insertions leading to truncated forms of merlin.

In 1995, Ruttledge et al. searched for a connection between the type of mutation in the NF2 gene and the severity of the disease. NF2 had previously been subdivided into two forms: the severe Wishart type and the more mild Gardner type. The Wishart type has an early onset, usually before age 25, a rapid course and multiple other tumors in addition to BVS. Wishart individuals undergo repeated surgeries and rarely survive past 50. The Gardner type is mild, has a later onset, progresses slowly and rarely has tumors other than BVS (Ruttledge 331). Ruttledge et al. discovered a correlation between protein truncating mutations (80% nonsense or frameshift causing deletion/insertion) and the severe Wishart type whereas the mild Gardner type is usually found in cases of single codon alterations, splice-site mutations, missense and 3-bp insertions (Ruttledge 341).

Our case study family was one of 73 families studied by Ruttledge et al. in 1995 (Ruttledge 331). The 1995 study pinpointed a 200 base pair insertion in the intron between exons 15 and 16 in eight affected family members and four at-risk individuals (Ruttledge 339). The family had been diagnosed with the mild Gardner type of NF2 because of the late onset of BVS and the lack of non-BVS tumors. Locating the causal mutation in a noncoding intron region fit with the hypothesis that less severe mutations result in less severe manisfestations. It is possible to have both types within a family yet all family members studied fit the Gardner type diagnosis. Since this mutation is in a splice site, partial splicing could cause the increased use of the normal gene, resulting in a more normal type of individual (Ruttledge 340) and helping to explain incomplete penetrance in some individuals who obviously inherited the mutation.

NF2 is incompletely penetrant and shows variation in its expression. Four of the seven affected cousins in this family came from one family with a father who had the disease. Since each sibling had a 50% chance of inheriting the defective gene, the probability of all four of them inheriting the gene is 1/16 or 0.0625. The fact that the mutant NF2 gene can be incompletely penetrant further reduces the likelihood that all four would both inherit the gene and exhibit the mutation. Because this is such an unlikely occurrence and the family had no unaffected offspring, perhaps the mother was also affected. The other possibility is that both of the father's genes were mutated. Either of these is possible since 50% of NF2 cases are due to new mutations (Narod 494).

There are two individuals in this family in whom the mutant NF2 gene appears nonpenetrant. One of these is in the first filial generation (individual II4) and is unaffected yet he has a son with the disease. Narod et al. mentioned that this individual did have progressive deafness though it could have been occupational in origin and he died in his fifties. Although this individual never had a confirmed BVS, his phenotype remains ambiguous as he was never tested (Narod 487). The other seemingly nonpenetrant family member is in the same generation as the seven cousins and already has a son with the disorder. This nonpenetrance is probably due to the mildness of the Gardner type and the partial splicing effect of this family's particular mutation.

Neurofibromatosis is a genetic disease that currently has no cure, the only therapy involves treatment of symptoms. Treatment consists of removal of harmful tumors and is most effective if done early in the growth stage of the tumor. Since NF2 is incompletely penetrant and age of onset can be late, early screening is important in affected families. The CA repeat polymorphism is the most commonly used test for the mutant NF2 gene but is not completely reliable because it is ten map units from the NF2 gene and recombination occurs 10% of the time giving a false reading. This test is still fairly reliable for NF2 diagnosis in combination with pedigree analysis but it will not diagnose the NF1 disorder because NF1 does not share the same locus or chromosome.

Families with a known history of NF2 need to be carefully screened and would benefit from genetic counseling if they choose to have children. Individual III10 has been diagnosed with NF2 and married a man who also has NF2. Their child has a 25% chance of not inheriting the mutation. The fact that the mother comes from a family where there are no unaffected offspring may decrease the chances further of having an unaffected child. Her mother, though she does not exhibit the phenotype, could also have the mutation, making individual III10 homozygous and giving her child a 100% chance of inheriting the mutation. Testing her mother (individual II7) for the CA repeat polymorphism would help to clear this up.

Like all cancers, NF2 has no known cure. Genetic analysis techniques have improved dramatically over the past ten years and it is only recently that we have had the technology available to pinpoint the mutant gene and discover the types of mutations causing dysfunction in its encoded protein. Until we learn how to repair these mutations in the DNA or in the resulting protein, we are capable only of testing for the mutation, genetic counseling of affected individuals to try to avoid its inheritance, and treatment of symptoms. We have come far in our understanding of all the complex interactions involved in gene regulation and cancer, yet there is still so much to be learned.


Bianchi, Albert, Shin-Ichiro Mitsunaga, Jin Quan Cheng, et al. "High Frequency of Inactivating Mutations in the Neurofibromatosis Type 2 Gene (NF2) in Primary Malignant Mesotheliomas." Proceedings of the National Academy of Sciences, USA. Vol 92, 10854-10858, Nov. 1995.

Bourn, D., S. A. Carter, D.G.R. Evans et al. "A Mutation in the Neurofibromatosis Type 2 Tumor-suppressor Gene, Giving Rise to Widely Different Clinical Phenotypes in Two Unrelated Individuals." American Journal of Human Genetics. 55: 69-73, 1994.

Clayman, Charles B., M. D. The American Medical Association Encyclopedia of Medicine. New York: Random House, 1989.

Narod, Steven A., Dilys M. Parry, Jillian Parboosingh, et al. "Neurofibromatosis Type 2 Appears to be a Genetically Homogenous Disease." American Journal of Human Genetics. 51:486-496, 1992.

Rouleau, Guy A., Philippe Merel, Mohini Lutchman, et al. "Alteration in a New Gene Encoding a Putative Membrane-organizing Protein Causes Neurofibromatosis type 2." Nature. Vol. 363, 10 June 1993.

Ruttledge, Martin H., Anne A. Andermann, Catherine M. Phelan, et al. "Type of Mutation in the Neurofibromatosis Type 2 Gene (NF2) Frequently Determines Severity of Disease." American Journal of Human Genetics. 59: 331-342, 1996.

Work Consulted

Wolff, Roger K., Kelly A. Frazer, Robert K. Jackler, et. al. "Analysis of Chromosome 22 Deletions in Neurofibromatosis Type 2-Related Tumors." American Journal of Human Genetics. 51: 478-485, 1992.

Return Case Studies in Virtual Genetics 1996-1997