Clinical and Genetic Overview of Neurofibromatosis Type 2 (NF2)

Clinical features

The first description of bilateral VS with multiple intracranial tumors, subsequently known as NF2, was reported by Wishart in 1822 [19]. A patient showed up with bilateral hearing loss that gradually worsened from several weeks ago, however, eventually expired 1 year later at the age of 21 with pain attacks and swallowing disability. Two centuries later with further research, the mean age at diagnosis of NF2 patients is reported as 24–28 years, mostly presenting with bilateral/unilateral hearing loss, tinnitus, and imbalance [14,15]. Bilateral VS and other CNS tumors are frequently found via computed tomography (CT) scans or magnetic resonance imaging (MRI). However, the mean age at onset of initial symptoms associated with NF2 is reported as 17–20 years, which is 7–8 years earlier than the age at diagnosis [14,15]. Initial symptoms include painful or growing skin tumors, vision loss, muscle weakness, headaches, and seizures, mostly caused by other CNS tumors rather than VS [15]. These manifestations should be carefully evaluated in patients who might be at risk for NF2 but without VS yet in CT scans or MRI.

Pediatric patients with NF2

As stated above, NF2 is commonly presented among young adults. However, 61 of 334 (18.2%) NF2 patients were younger than 15 years according to a research based on the UK database of NF2 [7]. Another research reported that 25 of 80 (31.2%) individuals diagnosed as NF2 were before the age of 18 years old [4]. Almost half of these patients are sporadic cases without family history of NF2. Symptoms in pediatric NF2 patients are usually caused by non-VS CNS tumors. Ophthalmic changes are rather typical to pediatric NF2 patients due to early onset cataracts, retinal hamartoma or optic sheath meningiomas [1]. Some patients present with facial palsy secondary to facial nerve schwannomas [4]. Only 20–32% of pediatric NF2 patients have hearing problems at the initial presentation [4,5].

Phenotypes of NF2

There are two major phenotypes of NF2 previously described. Patients with severe clinical course are classified as the “Wishart type” who usually present symptoms before the age of 25 years with rapid progression of bilateral VS and/or other CNS tumors. The other phenotype is the “Gardner type” that appears later in life following a benign course with small number of CNS tumors. However, it is not easy to perfectly fit every patient to either type because of the substantial variation of clinical courses [20].

Another phenotype, rather minor, is “congenital NF2” which is defined as “NF2 with onset before the age of 1 year” [18,20]. Natural history is described differently from that of typical NF2 patients where small bilateral VSs (<1 cm) are detected as early as the first months of life via incidental MRI findings. Small bilateral VSs stay stable or asymptomatic for one to two decades until sudden and rapid (<12 months) progression at the age of 11–15 years. Congenital NF2 is also associated with other CNS tumors (e.g., meningiomas, ependymomas) and exhibits large numbers of skin NF2 plaques in atypical locations (e.g., face, hands, legs and knees) which revert to normal skin appearance at the time of VS progression. Asymptomatic diffuse high signal lesions in the periventricular regions surrounded with calcification are known as typical signs in MRI findings.

Mosaic NF2 is an alternative form of NF2 caused by postzygotic somatic mutations of the NF2 gene [20]. Patients with mosaic NF2 usually present mild symptoms with unilateral VS associated with ipsilateral meningiomas, or multiple schwannomas localized to one part of the peripheral nervous system. A recent study of 1055 de novo NF2 patients estimated that approximately 60% of these cases were mosaic patients [8]. Mosaic NF2 is diagnosed according to the latest criteria (Table 1) via genetic testing (Fig. 3).

Fig. 3.

Flow chart of genetic testing for NF2 and SWN [16]. SMARCB1 : SWI/SNF-related, matrix-associated, actin-dependent regulator of chromatin, subfamily b, member 1, LZTR1 : leucine zipper-like transcription regulator 1, PV : pathogenic variant, NF2 : neurofibromatosis type 2, VAF : variant allele fraction, SWN : schwannomatosis, LOH : loss of heterozygosity.

Survival rates

At least two studies have reported survival rates of NF2 [10,14]. Evans et al. [10] estimated a median survival time of approximately 15 years after diagnosis. More than 40% of NF2 patients were expected to pass away by the age of 50 years, whereas all patients would die by 70 years. According to Otsuka et al. [14], the overall survival rates at 5, 10, and 20 years in NF2 patients were 85%, 67%, and 38%, respectively. However, early ages at symptom onset showed poor survival. The survival rates at 5, 10, and 20 years in patients younger than 25 years were 80%, 60%, and 28%, respectively, whereas the rates were 100%, 87%, and 62% in patients older than 25 years, respectively. More than 60% had died before reaching 44 years. They also reported that the size of VS at diagnosis showed correlation to survival rates. Small VS (≤2 cm) showed better survival than those with medium-sized tumors (>2 cm and ≤4 cm), although patient’s sex, positive family history of NF2, and presence of other CNS tumors or dermal abnormalities did not affect their survival. The most frequent causes of death are tumor burden, peri-operative complications, and malignancy from NF2 related tumors [20].

NF2 tumor suppressor gene

The NF2 gene was identified in 1993 revealing 17 exons on chromosome 22q12 which encoded for a protein called merlin (moesin-ezrin-radixin-like protein) or schwannomin [2,17,23]. Merlin is a protein product similar to the ERM (ezrin, radixin, moesin) protein family members and regulates cell proliferation and functions in cell-to-cell adhesion and transmembrane signaling. Following the Knudson’s two-hit model of tumorigenesis, schwannomas and meningiomas occur when both alleles of the NF2 gene are inactivated due to germline mutations, somatic mutations, or both of them happened one by one [2,12]. Pathogenic NF2 alterations penetrate nearly 100%, which means that individuals who carry the same pathogenic variant on the NF2 gene as their affected parent will develop the same clinical disorders in the future [16].

Correlation between NF2 gene alterations and clinical phenotypes

Molecular genetic studies have indicated that the type of mutation in the NF2 gene might be related to its clinical severity [14]. Generally, frameshift deletions/insertions and nonsense mutations that create truncated proteins are the most frequent germline mutations that cause the severe conditions, whereas missense mutations and splice-site mutations demonstrate milder clinical courses [3]. The Wishart type, which is typically associated with truncating alterations, show poor outcomes followed by younger age at diagnosis, higher incidence of meningiomas, spinal tumors, and non-VSs [21]. In contrast, the Gardner type has lower risk of mortality with fewer meningiomas than the Wishart phenotype due to its association with missense mutations or splice-site mutations [2,22].

Genetic testing for NF2 and other SWN

The flow of genetic testing for NF2 and SWN is shown in Fig. 3. First, pathogenic variants in NF2, SMARCB1 and LZTR1 are tested from unaffected tissues such as blood or saliva. Results with heterozygous pathogenic variants in each gene with clinical features are classified as full NF2-, SMARCB1-, or LZTR1-related SWN. Mosaic NF2-, SMARCB1-, or LZTR1- related SWN will be diagnosed when the pathogenic variants in each gene exhibit a variant allele fraction clearly less than 50% with clinical features. If there were no pathogenic variants in NF2, SMARCB1 or LZTR1, additional testing from two or more anatomically unrelated tumors is required. The individual will be classified as mosaic NF2-, SMARCB1-, or LZTR1-related SWN when the tumors shared pathogenic variants in NF2, SMARCB1 or LZTR1. Another testing for loss of heterozygosity of chromosome 22q is considered when tumors do not share pathogenic variants in these genes. The result finally classifies the patient either to 22q-related SWN or “SWN-not elsewhere classified”.

Genetic testing is a certain method to make diagnosis of NF2, although it is not necessary if the patient meets the characteristic clinical features. It plays a supporting role when young individuals have not exhibited any symptoms yet even though with high suspicion of NF2 according to the presence of first-, second- or third-degree relatives that are diagnosed with NF2. Presymptomatic genetic testing gives benefit to the family that the parents can survey the chance of NF2 in their children and perhaps to provide early care intervention [2]. It is also useful for confirmation of NF2 when the individual has no clinical features of NF2 but incidentally identified a pathogenic variant of NF2 in the blood via genetic sequencing for another disease [22]. However, this situation also gives caution to us not to easily diagnose NF2 or other SWN with the presence of pathogenic variants alone. In response to the increasing number of panel testing and wide analysis for underlying genetic conditions among various diseases nowadays, pathogenic variants of NF2, SMARCB1 or LZTR1 have been frequently identified in some asymptomatic individuals who had no suspicion of NF2 and other SWN [11,16]. The diagnosis of NF2 should be prudently done with consideration of the patient’s clinical features and background altogether.