Neurofibromatosis Type 1

What is Neurofibromatosis Type 1

Neurofibromatosis type 1 (NF1) is a neurocutaneous disorder with a prevalence of approximately 1 in 3500 (Huson & Hughes, 1994). Patients with NF1 have a wide variety of manifestations and a broad range of severity. One of the most common complications of NF1 in childhood is cognitive/neuropsychological dysfunction. These deficits are wide ranging and can be responsible for significant difficulties in daily activities, especially in the academic environment.

 

Knowledge regarding the cognitive profile of children with NF1 has increased dramatically over the past decade. Initially mental retardation was thought to be highly prevalent in NF1 patients, however large population studies have shown that there is only a slight increase in the incidence of mental retardation (North, 1999). Many studies have shown a slight reduction of general intellectual functioning, with IQ falling within the average to low average range (Ozonoff, 1999). Despite only minimal influence of the NF1 gene upon general intellectual functioning, more significant specific cognitive deficits have been reported in several areas, including perception, attention, executive functioning and language skills. These neuropsychological deficits result in a high prevalence of academic difficulties, and have important implications for schooling, career choice, and overall quality of life.

History of NF1

In 1882 Dr Frederick von Recklinghausen provided the first systematic description of neurofibromatosis. He published two case reports in a monograph entitled On Multiple Neurofibromas of the Skin and their Relationship to Multiple Neuromas. Von Recklinghausen correctly identified the pathology of the cutaneous and subcutaneous tumours and coined the term “neurofibroma”. From this time the condition became known as von Recklinghausen’s Disease.

It was not until the beginning of last century that Thomson (1900) discovered the genetic nature of the disorder, demonstrating that the disease was familial in 30/77 reported cases. In 1918 Preiser and Davenport proposed autosomal dominant inheritance. From the end of the nineteenth century, innumerable case reports of NF1 appeared in the medical literature, often focusing on the more severe manifestations and overestimating disease complications. The first large and reasonably accurate surveys of the disease appeared in the 1950s (Borberg, 1951; Crowe, Schull, & Neel, 1956). These early studies were limited however as they failed to distinguish between different forms of neurofibromatosis.

Interest in NF1 was sparked through John Hurt’s portrayal of Joseph Merrick in the film “The Elephant Man”. Joseph Merrick was a severely disfigured man who lived in Great Britain during the late 1800’s. He was exhibited in “freak shows” and carnivals throughout the country. People were morbidly fascinated by his grotesque appearance. In 1971, based on the medical knowledge available to him at the time, Ashley Montagu, suggested a connection with NF1 in his book, ‘The Elephant Man’. The psychological and emotional impact this erroneous connection had on NF1 families at the time of a new diagnosis was immense. The image of the physical deformities of Joseph Merrick was immediately brought to mind. At the same time, significant research into NF began to be conducted and support groups for affected individuals and their families were established. In 1986, Tibbles and Cohen identified the medical condition of Joseph Merrick as “Proteus Syndrome,” a genetically separate but similar condition to NF1. Unfortunately, the confusion of the two conditions persists. The process of re-educating the medical community, the media, and the general public continues.

Detailed research into the clinical manifestations and cognitive profile has increased markedly over the past 15 years. In 1987, definitive diagnostic criteria for NF1 and NF2 were proposed (National Institutes of Health Consensus Development Conference, 1988), enabling accurate and uniform clinical diagnoses of the two disorders. In 1990 the NF1 gene was cloned and its gene product, neurofibromin, was identified (Wallace, Marchuk, Anderson, Letcher, Odeh, Saulino, Fountain, Brereton, Nicholson, & Mitchell, 1990). A few years later the NF2 gene was cloned and its gene product, merlin, was identified (Rouleau, Merel, Lutchman, Sanson, Zucman, Marineau, Hoang-Xuan, Demczuk, Desmaze & Plougastel, 1993). NF2 is also an autosomal dominant disorder however it has been localized to chromosome 22, as opposed to NF1, which has been localized to chromosome 17. The genetic discrimination between these two disorders has further alleviated diagnostic confusion.

 

Clinical complications of Neurofibromatosis Type 1

Skin features of NF1

Café-Au-Lait Spots (CALS) are the most common manifestation of NF1. They are often present at birth or become visible during the first few months of life, and usually appear before the age of two. By the age of five, over 95% of NF1 children have CALS and these increase in size and number throughout life (Korf, 1992). The typical CALS have sharp, well defined borders and become darker in sun-exposed areas. Although CALS are present in the general population, it is only six or more CALS which indicate a strong likelihood of the presence of the disease.

 

Skinfold freckling represents the second most common feature of NF1. It is present in approximately 80% of patients, with the majority having multiple sites of freckling. This freckling generally appears before the age of five and is often present in the axilla, inguinal regions, and submammary regions in women. In obese individuals with NF1, freckling is often seen between skin folds.

Discrete neurofibromas are benign tumours arising from nerves that contain neural elements and an overgrowth of small blood vessels and fibrous connective tissue. They do not usually develop until preadolescence, and increase during adolescence and young adult years. Early neurofibromas may appear as “divots” in the skin, with reddening of the skin due to dilation or proliferation of overlying capillaries. Cutaneous neurofibromas occur within the dermis and epidermis and movement of the skin results in movement of the tumour.

In contrast, subcutaneous neurofibromas lie underneath the dermis and the skin moves over them. Subcutaneous neurofibromas are discrete, firm, generally have a spherical or oval shape and are often tender to palpation.

Over 50% of children with NF1 have neurofibromas, with the majority having more than one. The occurrence of neurofibromas increases as the patients get older, and over 95% of patients with NF1have neurofibromas in adulthood (Young et al., 2002). The earlier neurofibromas appear, the more likely the individual is to develop a high tumour load with age. Discrete neurofibromas are usually not premalignant lesions and rarely transform into malignant tumours (neurofibrosarcomas).

Plexiform neurofibromas arise from hamartamatous overgrowth of groups of deep or superficial nerves and are present in approximately 30% of patients with NF1. They appear as a soft-tissue mass under the skin, often following the course of a peripheral nerve. They may be superficial, deep, or nodular. Hyperpigmentation of the overlying skin may occur, as well as hypertrophy or deformity of the involved area. Plexiform neurofibromas may cause severe cosmetic disfigurement or compression or overgrowth of other structures. Large plexiform neurofibromas are congenital in origin and are usually obvious by the age of two years.

Plexiform neurofibromas have a potential for malignant transformation (Korf, 1999). Approximately 3% of lesions may develop into highly malignant peripheral nerve sheath tumours, accounting for the increase lifetime risk of malignancy in patients with NF1 (King, Debaun, Riccardi & Gutmann, 2000; Waggoner, Towbin, Gottesman & Gutmann, 2000).

Eye abnormalities in NF1

Lisch nodules are harmless iris hamartomas. They are rare in children under six, yet become increasingly frequent in older patients. Over 50% of NF1 children have Lisch nodules and there is an increasing frequency of these lesions in older patients, with over 98% of adults with NF1 having them. Lisch nodules do not portend other ocular manifestations of NF1 or visual compromise. They can be seen by a slit-lamp examination, and present as three-dimensional translucent masses punctuated by melanin containing cells. Optic pathway gliomas are the most common central nervous system tumours in patients with NF1 (Listernick, Louis, Packer & Gutmann, 1997). They are present in 15-20% of patients on neuroimaging, however only 30-50% of these tumours become symptomatic.

Histologically optic gliomas are low grade pilocytic astrocytomas and it is not known why some tumours progress rapidly to cause symptoms while others remain quiescent for many years. Common symptoms of optic pathway tumours are decreased visual acuity, visual field defects, proptosis, strabismus and hypothalamic dysfunction. A study of precocious puberty in NF1 found all cases were associated with optic pathway tumours with primary or secondary involvement of the hypothalamus (Habiby, Silverman, Listernick & Charrow, 1995). The first sign of precocious puberty in young children with NF1 may be an acceleration of linear growth. Thus advanced bone age is an indication for cranial imaging to investigate a possible hypothalamic tumour.

Listernick and colleagues (1994) examined the natural history of optic gliomas in children with NF1 and found that all symptomatic tumours were diagnosed before six years of age, and only 9% of children had evidence of tumour growth or deteriorating vision after diagnosis (Listernick, Charrow, Greenwald & Mets, 1994).

Bone abnormalities in NF1

NF1 patients may show abnormal bone development such as bone overgrowth. Overgrowth occurs when the upper or lower leg bones become thicker and longer on one side of the body, producing an inequality in leg lengths. Another orthopaedic problem seen in children with NF1 is congenital bowing of a long bone, most commonly the tibia. This bowing can be problematic, as fractures in this area tend to result in pseudoarthrosis (a new, false joint arising at the site of an ununited fracture) and fail to heal.

Scoliosis occurs in approximately 25% of children with NF1. The majority of cases of scoliosis in NF1 have a long C-shaped curve involving a significant portion of the spine (up to 10 segments), which is the type of scoliosis often seen in the general population. A more severe type of scoliosis is seen specifically in NF1 and involves a smaller portion of the spine (less than five vertebrae) causing a sharp, angular curve. This latter form may be associated with a localised area of vertebral dysplasia, possibly secondary to a paravertebral neurofibroma.

Macrocephaly and short stature and are frequent manifestations of NF1. Approximately 40-50% of children with NF1 have head sizes that are well above average for their age and height (absolute head circumference >98th percentile). Head growth tends to follow a curve parallel but greater than normal growth rate. In most cases there are no associated neurological problems, and the increased head circumference is due to increased brain volume rather than ventricular dilation or abnormalities of the bony calvarium. This increased brain volume has been shown to be associated specifically with lateral volume expansion of the cerebral hemispheres (DiMario, Ramsby, & Burleson, 1999), an increase in gray-matter (Moore, Slopis, Jackson, De Winter and Leeds, 2000), as well as increases in white-matter volume (Said, Yeh, Greenwood, Whitt, Tupler & Krishnan, 1996).

Short stature (<10th percentile) is common in patients with NF1. One study found that 25.5% of prepubescent children had short stature and there was a significant gradual reduction of their relative height for age (standard scores) during puberty (Carmi, Shohat, Metzker and Dickerman, 1999). In addition, short stature was also more common among patients with familial NF1 particularly if their father was affected and amongst those with CNS pathology (Carmi et al., 1999).

Central nervous system involvement in NF1

Headache is a common complication of NF1 and usually occurs in the absence of intracranial lesions or raised intracranial pressure. Since there is an increased risk of development of CNS tumours in patients with NF1, many patients with headaches require neuroimaging, although the risks for such problems are low.

Seizures are present in approximately 5% of patients with NF1. The types of seizures are variable and there is no one type of seizure which is typical in NF1 patients (Kulkantrakorn & Geller, 1998). Vivarelli and colleagues found that 7% of their patients had epilepsy, with the majority of their patients having partial rather than generalised seizures. Over half of their patients had seizures secondary to brain lesions. In most patients the seizures were able to be controlled, but around 29% were drug resistant. All patients with uncontrolled seizures had severe mental retardation (Vivarelli, Grosso, Calabrese, Faretani, Di Bartolo, Morgese & Balestri, 2003).

Other brain abnormalities have been identified, particularly on magnetic resonance imaging (MRI). The most common lesions, often referred to as T2-hyperintensities, are seen on T2 weighted MRI images as areas of hyperintensity (prolongation of T2). These lesions are usually isointense on T1 weighted images, they exert no mass effect, there is no surrounding oedema, and do not enhance with contrast (Sevick, Barkovich, Edwards, Koch, Berg & Lempert, 1992). They are not associated with focal neurological deficits or the presence of macrocephaly. The T2-hyperintensities are seen in 60-70% of children with NF1 (North 1997), and are also referred to as ‘hamartomas’ or unidentified bright objects (UBOs). They most commonly occur in the basal ganglia, brain stem, thalamus, optic tracts, and cerebellum (Van Es, North, McHugh, & de Silva, 1996). Pathological studies suggest that these hyperintense areas on MRI may represent dysmyelination or increased water content in the brain. Structural brain abnormalities have also been identified including larger brain volumes (Said, Yeh, Greenwood, Whitt, Tupler & Krishnan, 1996), increased area of the corpus callosum (Kayl, Moore, Slopis, Jackson & Leeds, 2000; Moore, Slopis, Jackson, De Winter & Leeds, 2000), and abnormal grey- white-matter ratios (Moore et al., 2000).

Cognitive dysfunction in NF1

  • Based upon Hyman, SL., Shores E.A., & North, K.N. (2005). The Nature and Frequency of Cognitive Deficits in Children with Neurofibromatosis Type 1. Neurology, 65, 1037-1044. (Part of Dr Hyman’s PhD Thesis)

Cognitive deficits represent the most common complication in children with NF1, affecting up to 80% of children (Hyman et al., 2005). School performance is often a major concern of parents as it can cause significant morbidity in terms of educational opportunities, career choice and self-esteem. There does not however appear to be a unique profile of learning disabilities as both nonverbal and verbal learning problems are both common (Ozonoff, 1999). Children with NF1 are often easily distractible, poorly organised, and have difficulties with visual perception (Eldridge, Denckla, Bien, Myers, Kiaser-Kupfer, Pikus, Schlesinger, Parry, Dambrosia, Zasloff & Mulvihill, 1989; North, Joy, Yuille, Cocks & Hutchins, 1995). Language problems as well as problems with motor coordination are also quite common (Mazzocco, Turner, Denckla, Hofman, Scanlon & Vellutino, 1995). Mental retardation (Full Scale IQ < 70) is only slightly more common in patients with NF1 than in the general population (North, 1999).

Mental retardation is present in 6-7% of children with NF1. While this rate is approximately three times the rate of the general population, it is still relatively low compared to other genetic disorders affecting the central nervous system such as Fragile-X and Turners Syndrome. General intellectual functioning is significantly lower in children with NF1 compared to both the normal population and their siblings. IQ is however still within the lower end of the average range.

An important and unexpected finding in the study was that males with NF1 were more likely to have significantly weaker verbal than nonverbal skills compared to females (with nonverbal skills in the average range and verbal skills in low average range).

Children with NF1 have significantly lower reading, spelling, and mathematical ability than their siblings. When using an IQ-achievement discrepancy for defining specific learning disabilities (SLD), SLD was present in 20% of children with NF1 compared to 8% of controls (2.5 times more common in NF1). One of the most intriguing findings of the current study is the strong gender effects for SLD: 37% of males had an SLD compared to only 5% of females. This places males at a much greater risk for SLD in the NF1 population, and suggests that females are at no greater risk of SLD than their unaffected siblings. These gender effects are related to the high frequency of weaker verbal reasoning skills in males, and it appears that males with these significantly weaker verbal skills are at higher risk for specific learning disabilities.

Children with NF1 had a shortened attention span, as well as difficulties with sustained and switching attention when compared to their siblings. Interestingly they did not differ from their siblings on tasks of selective or divided attention. Although children with NF1 have a reduced attention span when compared to their siblings, they were able to mentally manipulate the information that they could encode within their working memory at the same level as their siblings. Approximately 63% of NF1 children have problems sustaining their attention.

There is a high comorbidity of NF1 with ADHD, with over 38% of the NF1 children fulfilling the diagnostic criteria for one of the ADHD subtypes. Compared to their siblings, ADHD was three times more common in the NF1 children. ADHD had similar frequencies in both NF1 males and females, unlike the general population in which there appears to be a predominance of males of approximately 3:1. The majority of NF1 children had the combined subtype (24.7%), which was double the prevalence of the inattentive subtype (12.3%). Hyperactivity alone was very rare (only 1/81). Interestingly, one study found that stimulant medication lead to an improvement in attention, anxiety-depression and social competence of children with co-morbid NF1 and ADHD.

Of those children with NF1 and SLD, approximately 45% have co-morbid ADHD. Of those with ADHD, approximately 26% have co-morbid SLD. Thus, there is a slightly increased risk of developing an SLD if ADHD is present (26% compared to 20%). NF1 children with an SLD are again slightly more likely to have ADHD than those without an SLD (45% compared to 37%). These co-morbidity figures are somewhat lower than those reported in the general population with some estimates as high as 70%.

Children with NF1 had deficits in both planning and abstract concept formation when compared to controls. However, they were not significantly impaired on a task of verbal fluency. The co-morbidity of ADHD and these executive deficits was examined, and there was no increase in frequency of executive deficits in those children with NF1/ADHD.

Visuospatial deficits are one of the most commonly and consistently reported neuropsychological deficits in NF1. Children with NF1 performed well below their siblings on a test of visuospatial ability. Performance on a perceptual task involving matching shapes and mental rotation was also significantly poorer in NF1 children than their unaffected siblings.

One of the most unusual findings of the present study was that children with NF1 did not differ from either the normal population or their siblings on memory tasks. Spared functioning was found for immediate memory, delayed memory and recognition memory in both the visual and verbal modalities. Interestingly, memory scores were actually much higher than IQ scores for the children with NF1.

The study demonstrates broad deficits in both receptive and expressive language skills in NF1 children. Children with NF1 had greater difficulties defining words, comprehending simple passages, and describing pictures. They also had difficulties with higher level verbal reasoning skills. Although studies have suggested that language deficits are just as common as visuospatial deficits in children with NF1, this study suggests that the language deficits are not of the same magnitude. Although language skills were not spared, they were no more deficient than IQ would predict, whereas visuospatial deficits were more common, more severe, and lower than IQ would predict.

Both fine motor skills and motor speed were reduced in NF1 children when compared to their siblings. Fine motor coordination deficits were present in approximately 20% of NF1 children. Slowing of motor speed was found in 30% of NF1 children.

Emotional and psychosocial problems

Children with NF1 have more problems with anxiety, depression and internalising behaviours, as well as having poorer social skills than their siblings (Dilts, Carey, Kircher, Hofman, Creel, Ward, Clark & Leonard, 1996). Johnson and colleagues also found a higher risk of suicide in patients with NF1, with 16% of children with NF1 having thoughts of suicide compared to 6% of unaffected siblings and 3% of unrelated controls (Johnson, Saal, Lovell, & Schorry, 1997). Samuelsson and Riccardi (1989) reported that 33% of adults with NF1 had been diagnosed with a mental illness, with the majority being diagnosed with anxiety and depression.