Case 13-2005 — A 48-Year-Old Man with Weakness of the Limbs and Multiple Tumors of Spinal Nerves
http://www.100md.com
《新英格兰医药杂志》
Presentation of Case
A 48-year-old, right-handed man was evaluated in the neurology clinic of this hospital because of weakness of his arms and legs.
When the patient was 18 years of age, a mass on the right side of his neck developed; a biopsy was performed at another hospital, and he was told he had von Recklinghausen's neurofibromatosis. At 45 years of age, pain and weakness in his left leg developed. A radiograph of the left knee performed elsewhere showed what were described as tumors, and he was referred to an orthopedist at this hospital. Magnetic resonance imaging (MRI) showed an enhancing lesion, 2 cm in diameter, on the sciatic nerve, extending from the ischial tuberosity to the deep peroneal nerve. Physical therapy was administered for the next 10 months, but the pain persisted. A repeated MRI study showed that the thickening of the sciatic nerve involved both the peroneal and tibial branches, extended to the lowest level in the images, and had increased in diameter to 3.6 cm. Three months later, increased pain and a foot drop on the left developed; MRI scanning of the brain and the cervical, thoracic, and lumbar spine performed elsewhere showed no abnormalities of the brain, but there were bilateral enhancing lesions in all cervical neural foramina, with cord compression at C1–2, C2–3, and C4–5 and bilateral rounded nonenhancing lesions in the neural foramina of the thoracic and lumbar spine. Treatment with gabapentin was begun.
During the following year, the patient began using a four-point cane and an orthosis for his left foot, and weakness developed in the right leg. He fell on numerous occasions. He began to notice alternating pallor and redness in his feet. An eye examination documented no sign of Lisch nodules. Weakness developed in the left arm and subsequently in the right arm.
About one year after starting treatment with gabapentin, the patient was referred to the neurosurgical clinic at this hospital. An MRI of the brain showed no abnormalities. An MRI of the cervical spine showed bilateral enhancing tumors at each level. At the C1–2 level, the cord was compressed from both sides. At the C4–5 level, the cord was displaced into the right canal and indented. An MRI of the thoracic and lumbar spine showed expanded nerve roots bilaterally at every level without cord compression; moderate stenosis of the canal was seen at the L4–5 level. A computed tomographic (CT) scan of the abdominal area showed multiple tumors involving the pelvic and paraspinal nerves and multiple subcutaneous masses. The patient was then referred to the neurology clinic.
He characterized his primary problem as weakness, and he also had pain, primarily in his left leg, that was worse with activity. He had no problems with bowel or bladder control and no back pain, headaches, or vision problems. He did not drink or smoke and had been unemployed for four years. His medications were sertraline and gabapentin.
The patient's mother died at 87 years of age of lymphoma and his father at 47 years of age of smoking-related emphysema. His mother had had a lump on the back of her neck; he did not recall either parent having skin lesions. His brother was alive at 46 years of age with liver cancer and multiple sclerosis; his sister had died suddenly at the age of 39, and the patient did not know the cause. He did not recall either sibling having skin lesions. His sister's son was said to have neurofibromatosis. The patient was married and had one son, who was well.
He was alert and answered questions clearly. There were no dysmorphic features. The vital signs were normal. Both hands and feet were reddened and cold at the beginning of the visit; by the end of the visit, the patient's feet were deep red up to the middle of the tibia. Examination of the skin under ambient light and with a Wood's lamp revealed two café-au-lait macules greater than 1.5 cm in diameter, one on the left lower part of the abdomen and another on the inner aspect of the left buttock. A single, raised, purple cutaneous lesion, 1 cm in diameter, was present on the left side of the chest. There was no axial or inguinal freckling. Multiple subcutaneous tumors were visible throughout the body. The left leg was slightly thicker than the right. An examination of the head, neck, chest, and abdomen revealed no abnormalities.
On neurologic examination, the patient's gait was slow and shuffling, and he walked bent at the waist. The cranial-nerve function was normal. On motor examination, strength in both arms was 4/5 with bilateral pronator drift. Fine hand–motor coordination was intact. Motor examination of his legs and feet showed bilateral increased tone with contractures at both heel cords. The right foot was held in a turned-in position. The strength was 4/5 in the right leg and 3/5 in the left leg, with decreased muscle mass in the left leg and left foot drop. Brachioradialis reflexes were 2+ and symmetric in the upper extremities, but on the right side there was spread into the biceps. The Hoffmann's sign was positive on the left and negative on the right. Palmomental reflexes were negative bilaterally. Knee and ankle reflexes were 3+. The Babinski reflex on the right was unreactive and on the left turned upward. The results of routine laboratory tests were normal.
Two weeks later, laminectomy of C3–6, with resection of bilateral C2, right C3, and left C5 tumors, was performed, as was fusion of the cervical spine. On pathological examination, all tumors were neurofibromas. Three months later, the patient's pain had diminished and his gait had improved; he was able to walk up a full flight of stairs at home and used a single-point cane only when outside. A new right foot drop had developed. Muscle strength in all extremities was 5/5 except for the ankle flexors, which were 4/5, and he was unable to flex either great toe upward. Muscle tone was slightly increased in the lower extremities, but less so than it had been at the first visit.
Three months after the patient's evaluation, his sister's son was also evaluated in the neurology clinic. He was a 29-year-old, right-handed man who had had flat, pigmented lesions on his skin as a teenager. When he was 19 years of age, a subcutaneous nodule appeared on his right shoulder; over the years, more nodules developed, one of which, on his left lower back, became painful. The painful nodule was removed, and pathological examination disclosed a neurofibroma. He had had weakness of his left leg when climbing stairs for three months before his evaluation. He had no other neurologic or physical symptoms.
The nephew thought his mother had had what he described as "brown birthmarks." At 39 years of age, she had fallen backward and lost consciousness; a craniotomy was performed at another hospital, but she died three days later. An autopsy reported the cause of death as blunt trauma to the head, with epidural, subdural, and subarachnoid hemorrhage. No skin lesions or nerve tumors were noted. His maternal half-brother and half-sister each had one brown skin lesion. His father was well.
Examination of the nephew's skin under ambient light showed freckles in the inguinal regions but not in the axillae. Multiple café-au-lait macules were present, many greater than 1.5 cm in diameter. There were multiple palpable subcutaneous nodules, two of which were painful, on the left leg and left shoulder. There were no cutaneous nodules. The results of a detailed neurologic examination revealed no abnormalities.
Diagnostic tests were performed on both the patient and his nephew.
Differential Diagnosis
Dr. Bruce R. Korf: This clinical history poses two major questions. First, what accounts for this man's weakness, both before and after his surgery? Second, what is his underlying diagnosis and how is his condition related to that in his nephew? May we review the MRIs?
Dr. John W. Henson: MRI scans were obtained through the spine, the lumbosacral plexus, and the brain before and after the administration of gadolinium. Parasagittal images revealed large masses along the courses of multiple, bilateral cervical nerve roots (Figure 1A). On axial images through the C2 vertebral body, these lesions had mild heterogeneity of signal intensity. A focus of T2-weighted hypointensity was present within the lateral aspect of the right C2 paraspinal mass lesion, forming a so-called "target sign" (Figure 1B). Heterogeneous gadolinium enhancement was seen. The tumors produced severe compression of the cervical spinal cord at the C2 level and extended laterally into the brachial plexus. There was mild enlargement of thoracic spinal nerve roots. Coronal fast spin-echo inversion-recovery sequences revealed masses along the courses of the lumbar nerve roots and lumbosacral plexus, with extensions along both sciatic nerves (Figure 1C).
Figure 1. Neuroimaging Studies Showing Multiple Peripheral Nerve Tumors.
A fast spin–echo (FSE) T2-weighted image in the sagittal plane from the cervical spine revealed the presence of multiple mass lesions (arrows) along the cervical spinal roots (Panel A). An axial FSE T2-weighted image at the level of the second cervical vertebral body (Panel B) showed the presence of cervical spinal cord compression by mass lesions along bilateral spinal nerve roots. The tumors were hyperintense on T2-weighted images and had a focal region of low signal consistent with the so-called "target sign" (arrow) and had heterogeneous enhancement after the administration of gadolinium. A coronal FSE inversion recovery image through the lumbosacral plexus (Panel C) showed plexiform mass lesions and diffuse enlargement of the sciatic nerves (arrows).
Multiple neurogenic tumors can be either neurofibromas or schwannomas.1,2 On imaging studies, neurofibromas usually show heterogeneous enhancement, whereas schwannomas usually enhance homogeneously. Approximately half of neurofibromas show the target sign, which is a central region of lower intensity surrounded by higher intensity on T2-weighted images. This is more common in plexiform neurofibromas but can rarely be present in schwannomas. Neurofibromas are more likely to produce diffuse or plexiform lesions than are schwannomas.
Imaging of the patient's brain revealed no sign of lesions that are considered typical of neurofibromatosis type 1, such as glioma of the optic pathway, foci of abnormal T2-weighted hyperintensity within the brain, plexiform neurofibromas of the ophthalmic division of the fifth cranial nerves, or dysplasia of the greater wing of the sphenoid bone. There was no evidence of nodular enhancement along the courses of the eighth cranial nerves to suggest the presence of vestibular schwannomas, as are seen in neurofibromatosis type 2.
The interpretation of the neuroradiologic images was that there were multiple neurofibromas, with a plexiform pattern of involvement.
Dr. Korf: This patient was given a diagnosis of von Recklinghausen's neurofibromatosis at 18 years of age. That was at least 30 years ago, before the formulation of diagnostic criteria for neurofibromatosis type 1 and neurofibromatosis type 2 in 1987.3 These two forms of neurofibromatosis were not always clearly distinguished before 1987, and we do not know whether any other signs of either type were present in this patient. His current symptoms of pain and weakness began three years before his evaluation here. Review of the first MRI obtained of his left leg reveals enlargement of the sciatic nerve, with an appearance typical of neurofibroma. A neurofibroma is a benign neoplasm involving various cellular elements of the nerve sheath. Neurofibromas may occur as focal growths along a nerve or at a nerve ending in the skin, or they can extend along a length of nerve and involve multiple fascicles of a larger nerve, forming a plexiform neurofibroma.4
Plexiform neurofibromas can be asymptomatic, but pain, weakness, and sensory dysfunction can occur, particularly if the nerve enlargement leads to compression. Subsequent MRI scanning revealed much more widespread disease, with multiple tumors of the cervical nerve roots, several of which were compressing the spinal cord. Weakness increased over a period of about a year and spread to include his right leg and arm.
A neurologic examination revealed signs of both upper- and lower-motor-neuron dysfunction. The foot drop on the left suggests a peroneal neuropathy. Increased tone, brisk deep-tendon reflexes, and a positive Hoffmann's sign and Babinski's sign suggest cord compression. Vasomotor dysfunction in the lower part of the legs may be further signs of peripheral-nerve dysfunction. The patient had marked improvement in strength after resection of several cervical tumors, but a newly developed foot drop on the right suggests peroneal-nerve dysfunction on that side. His pain and weakness are likely to be the combined results of spinal cord compression at multiple levels and peripheral-nerve dysfunction in the legs, probably due to compression of massively enlarged nerves.
Classification and Diagnosis of Neurofibromatosis
This patient has a disorder that leads to multiple tumors of the peripheral-nerve sheath. Some form of neurofibromatosis would seem most likely. There are two widely recognized forms of neurofibromatosis, type 1 and type 2 (Table 1). Within the past decade, an additional entity referred to as schwannomatosis5 has been added to this short list, but since the tumors in this patient were neurofibromas, not schwannomas, we can exclude this condition. Neurofibromatosis type 2 is also unlikely. Although neurofibromas can occur in persons with neurofibromatosis type 2, schwannomas are far more common. The hallmark lesion, the vestibular schwannoma, was not present on cranial MRI studies of this patient.
Table 1. Classification of the Forms of Neurofibromatosis.
The presence of multiple neurofibromas in this patient fulfills one criterion for neurofibromatosis type 1 (Table 2),3,6 but at least two are required to establish the diagnosis. He had two café-au-lait spots, not the requisite six, and did not have axillary or inguinal freckles, Lisch nodules (iris hamartomas that are found in most adults with neurofibromatosis type 1 and that are highly specific signs of the disorder), skeletal dysplasia, or optic glioma. There was no definitive diagnosis of neurofibromatosis type 1 in a first-degree relative.
Table 2. Diagnostic Criteria for Neurofibromatosis Type 1.
There are variant forms of neurofibromatosis (Table 3). Familial multiple café-au-lait spots7 and Watson's syndrome8 clearly do not fit this patient's history. Patients with segmental neurofibromatosis have symptoms confined to a particular region of the body.9 This localized form of the disease is attributable to somatic mosaicism for a neurofibromatosis type 1 mutation, at least in some cases. The disease in the patient under discussion was too widespread to be described as segmental.
Table 3. Variant Forms of Neurofibromatosis Type 1.
Ten patients10 have been described who had generalized neurofibromas but who did not display other features of neurofibromatosis type 1. In this group, both isolated and plexiform neurofibromas were seen, including spinal-nerve-root tumors, some of which caused cord compression. Some patients had dermal neurofibromas, whereas others had firm subcutaneous masses similar to those reported in this patient. It is unclear whether any of the patients in the report had as many tumors as the patient in this case.
The other relevant variant of neurofibromatosis is familial spinal neurofibromatosis.11,12,13 Patients with this condition have neurofibromas that originate from spinal nerve roots, often at every level, similar to the findings in the patient under discussion. Cord compression occurs commonly. Most patients have multiple café-au-lait spots, some have iris Lisch nodules, and skin-fold freckling is variable. Most patients have disease with features that fulfill the diagnostic criteria for neurofibromatosis type 1. However, typical dermal neurofibromas are usually absent in these patients, even in adulthood, when this tumor is apt to be found in patients with classic neurofibromatosis type 1. Mutations in the NF1 gene have been identified in several families.11,12,14
The patient in this case fits somewhere between these two entities — he has spinal tumors similar to those seen in patients with familial spinal neurofibromatosis, but he lacks multiple café-au-lait spots. I am inclined to classify him with the patients who have familial spinal neurofibromatosis because of his spinal tumors. Some patients with spinal neurofibromatosis have been described as having peripheral tumors.11 I suspect that the two entities of familial spinal neurofibromatosis and generalized neurofibromatosis without other signs may represent a spectrum of manifestations of one disease. In most patients — but not all — with spinal neurofibromatosis, findings have shown linkage to the NF1 locus or there are NF1 gene mutations, whereas molecular studies of individual patients with isolated generalized neurofibromas have not been reported.
Familial Spinal Neurofibromatosis
Does this patient have a familial disorder? His nephew is said to have "many" café-au-lait macules and had inguinal but not axillary freckling; both are usually seen in adults with neurofibromatosis type 1. The nephew had multiple painful subcutaneous neurofibromas, like the patient, and no dermal tumors. We do not know if he had spinal tumors, but he had weakness of the left leg. I believe that the nephew had the same disorder as our patient and that he would prove to have spinal neurofibromas if MRI were performed.
Although neurofibromatosis is a highly variable disorder, penetrance is complete, so I would not expect to see a skipped generation. Did the patient's sister, the mother of the nephew, also have this disorder? She died suddenly, reportedly with evidence of blunt head trauma. She is also said to have had "brown spots." Thus, it is possible that she also had neurofibromatosis. If the patient and his sister both had neurofibromatosis, one parent should also have been affected. The patient's mother is said to have had a "neck mass." Either parent could have been affected without the disease ever being diagnosed, given the paucity of externally visible signs in this patient. Alternatively, one of the patient's parents could have had mosaicism for a mutation that could have been transmitted to two children without affecting the parent, as has been documented for at least one family with neurofibromatosis type 1.15
In respect to this patient and his nephew, it is possible that each has a form of neurofibromatosis due to a distinct and independent gene mutation. Multiple different NF1 gene mutations in different members of the same extended family have been reported.16 I believe that this is unlikely in this case, since both the patient and his nephew have a similar disorder with a paucity of cutaneous signs and many subcutaneous and deep neurofibromas.
Genetic Testing in Neurofibromatosis
The patient and his nephew underwent a "diagnostic test." I do not believe that much more would have been learned from pathological examination of tumors. It would be valuable to obtain an MRI of the spine in the nephew, but nothing would be learned from additional MRI studies of the patient. The diagnostic test was therefore probably a molecular genetic test of the NF1 gene, searching for a gene mutation.
Until recently, there was no practical clinical diagnostic test for neurofibromatosis type 1. The gene is large (60 exons) and mutations are widely scattered in the gene, so that almost no two families with neurofibromatosis type 1 share the same mutation.17 This disorder has a high mutation rate, and hundreds of distinct mutations have been reported. There are some recurrent mutations, but no "hot spots" to guide clinical testing. There are no major genotype–phenotype correlations, except that patients who lack the gene entirely tend to have a severe phenotype with early onset and large numbers of neurofibromas, developmental delay, and dysmorphic facial appearance.18
Mutation analysis for neurofibromatosis type 1 is now available for clinical testing, with the use of a set of complementary techniques to find the various types of mutation, including deletion of the entire gene, small insertions or deletions that cause a frame shift, single-base changes that lead to a stop codon or amino acid substitution, and mutations that alter RNA splicing.19 Most of the mutations lead to premature termination of peptide synthesis and hence obliterate the function of the gene product. The fact that the mutations result in loss of function is compatible with the notion that NF1 functions as a tumor-suppressor gene in the pathogenesis of neurofibromas, requiring loss-of-function mutations in both alleles to produce a tumor.
Mutation testing has been performed in several patients with familial spinal neurofibromatosis (Figure 2). Of nine mutations reported, four are amino acid substitutions, which is remarkable since not more than 10 percent of all NF1 mutations have been found to be missense changes. Three are splicing mutations and only two generate stop codons, one by frame shift and one by nonsense mutation.11,12,14,20
Figure 2. Location of NF1 Gene Mutations Reported in Familial Spinal Neurofibromatosis.
The GAP (GTPase activating protein)-related domain, where the mutations occur, is labeled according to the source where the mutation was described. IVS5–2AG, L920P, P2200A, and S2698X were described by Kluwe et al.11; 8042insA by Ars et al.12; L357P and IVS39+3AG by Messiaen et al.14; and IVS31–5AG and L2067P by Kaufmann et al.20
Diagnosis
This patient had a mixed upper-motor-neuron dysfunction due to cord compression at multiple levels and lower-motor-neuron disease due to nerve compression. I believe the underlying cause was the presence of multiple spinal and plexiform neurofibromas caused by a variant of neurofibromatosis type 1 similar to familial spinal neurofibromatosis. I believe that both the patient and his nephew have the same disorder, and that both will be shown to have an NF1 gene mutation, probably a splicing defect or missense mutation.
Dr. Nancy Lee Harris (Pathology): Dr. MacCollin, can you give us your clinical impressions?
Dr. Mia MacCollin (Neurology): At the time the patient was first evaluated in the neurofibromatosis clinic, we who saw him believed that he had the rare syndrome of multiple isolated neurofibromas as described by Blakley et al.10 We thought that the nephew's illness was due to the relatively common occurrence of a new mutation in the NF1 gene and that his disease was genetically unrelated to his uncle's rare syndrome.
Clinical Diagnosis
Multiple isolated neurofibromas.
Dr. Bruce R. Korf's Diagnosis
Neurofibromatosis type 1 variant, similar to familial spinal neurofibromatosis.
Pathological Discussion
Dr. Harris: Dr. Smith, could you describe your operative findings?
Dr. Edward R. Smith (Surgery, Children's Hospital, Boston): A multilevel cervical laminectomy was performed, which revealed very large extradural tumors at C2 that were touching at midline. These tumors were resected, as was another extradural mass at C3. The dura was opened, revealing a sizable C5 tumor that was completely displacing the spinal cord to one side; it was removed. At the completion of the operation, despite the presence of abundant space in the canal, the spinal cord retained the impressive deformation shown on the preoperative radiographs, with a thin, effaced high cervical cord. The dura was closed, and a multilevel fusion with a lateral mass screw construct was performed.
Dr. Anat Stemmer-Rachamimov: The tumors were hypocellular, with myxoid stroma, a heterogeneous cell population of round cells and wavy cells, and entrapped large nerve bundles (Figure 3A). Only scattered Schwann cells, which are the neoplastic cells, were positive for S100 protein (Figure 3B), and neurofilament immunostaining highlighted enlarged nerve bundles and dispersed axons within the tumors (Figure 3C). The pathological diagnosis was multiple plexiform spinal-nerve-root neurofibromas, with no malignant transformation. The nephew's specimen was also a neurofibroma.
Figure 3. Specimens from Spinal-Nerve-Root Tumors.
The tumor is composed of spindled and round cells with entrapped axons (Panel A; hematoxylin and eosin). With immunostaining for S100 protein (Panel B), only scattered neoplastic Schwann cells are stained. The other cells are non-neoplastic stromal and nerve cells. Enlarged nerve bundles and dispersed axons within the tumors are highlighted with neurofilament immunostaining (Panel C). Multiple spinal neurofibromas at autopsy (Panel D, from another patient) show markedly enlarged spinal nerve roots, seen from the thoracic cavity, which are symmetrically involved by plexiform neurofibromas. (Autopsy photograph courtesy of Dr. Matthew Frosch, Department of Pathology, MGH.)
Patients with multiple spinal neurofibromatosis often have spinal tumors that are distributed symmetrically and involve many or all spinal roots (Figure 3D).11 In the absence of other clinical features of neurofibromatosis type 1, the pathological distinction between schwannomas and neurofibromas is important, because multiple spinal-root-tumors may be seen in schwannomatosis and neurofibromatosis type 2.21 Schwannomas are also tumors of Schwann cells, but they lack the heterogeneous cell population found in neurofibromas. This distinction may be difficult to make when schwannomas have prominent myxoid changes that mimic the histological appearance of neurofibromas. In these cases, it is helpful to find areas with classic schwannoma features, or to establish by S100 immunostaining or electron microscopy that only a subgroup of cells are Schwann cells.
Dr. Harris: Dr. MacCollin, will you tell us about the additional diagnostic testing?
Dr. MacCollin: A tissue sample from the patient was sent to the laboratory of Dr. Ludwine Messaien at the Center of Medical Genetics, Ghent University Hospital in Belgium. No alterations were found in the NF1 gene by protein-truncation testing. Direct sequencing of the gene revealed an unusual alteration in exon 22 (G3827A, leading to the substitution of glutamine for arginine at residue 1276). This change had previously been detected in a single sample from an unrelated 19-year-old patient with an unusually large load of cutaneous and plexiform neurofibromas. It had also been reported twice22 in a series of 301 sequence variants in the NF1 gene. It occurs at a CpG hypermutatable site that is also involved in the more typical change, C3826T, which causes a nonsense mutation at arginine 1276. The patient's nephew had the same mutation. Subsequent MRI examinations showed that the patient's nephew has multiple spinal nerve root plexiform neurofibromas, without cord compression. We have been unable to analyze the autopsy specimens from the patient's sister.
Anatomical Diagnosis
Neurofibromatosis type 1, affecting the patient and his nephew, with multiple spinal neurofibromas.
Dr. Korf reports having received lecture fees from Pharmacia.
Source Information
From the Department of Genetics, University of Alabama at Birmingham (B.R.K.); and the Departments of Radiology and Neurology (J.W.H.) and Pathology (A.S.-R.), Massachusetts General Hospital and Harvard Medical School — both in Boston.
References
Khong PL, Goh WH, Wong VC, Fung CW, Ooi GC. MR imaging of spinal tumors in children with neurofibromatosis 1. AJR Am J Roentgenol 2003;180:413-417.
Suh JS, Abenoza P, Galloway HR, Everson LI, Griffiths HJ. Peripheral (extracranial) nerve tumors: correlation of MR imaging and histologic findings. Radiology 1992;183:341-346.
Stumpf DA, Alksne JF, Annegers JF, et al. Neurofibromatosis: conference statement: National Institutes of Health Consensus Development Conference. Arch Neurol 1988;45:575-578.
Korf BR. Plexiform neurofibromas. Am J Med Genet 1999;89:31-37.
MacCollin M, Woodfin W, Kronn D, Short MP. Schwannomatosis: a clinical and pathologic study. Neurology 1996;46:1072-1079.
Gutmann DH, Aylsworth A, Carey JC, et al. The diagnostic evaluation and multidisciplinary management of neurofibromatosis 1 and neurofibromatosis 2. JAMA 1997;278:51-57.
Arnsmeier SL, Riccardi VM, Paller AS. Familial multiple cafe au lait spots. Arch Dermatol 1994;130:1425-1426.
Tassabehji M, Strachan T, Sharland M, et al. Tandem duplication within a neurofibromatosis type 1 (NF1) gene exon in a family with features of Watson syndrome and Noonan syndrome. Am J Hum Genet 1993;53:90-95.
Ruggieri M, Huson SM. The clinical and diagnostic implications of mosaicism in the neurofibromatoses. Neurology 2001;56:1433-1443.
Blakley P, Louis DN, Short MP, MacCollin M. A clinical study of patients with multiple isolated neurofibromas. J Med Genet 2001;38:485-488.
Kluwe L, Tatagiba M, Funsterer C, Mautner VF. NF1 mutations and clinical spectrum in patients with spinal neurofibromas. J Med Genet 2003;40:368-371.
Ars E, Kruyer H, Gaona A, et al. A clinical variant of neurofibromatosis type 1: familial spinal neurofibromatosis with a frameshift mutation in the NF1 gene. Am J Hum Genet 1998;62:834-841.
Pulst S-M, Riccardi VM, Fain P, Korenberg JR. Familial spinal neurofibromatosis: clinical and DNA linkage analysis. Neurology 1991;41:1923-1927.
Messiaen L, Riccardi V, Peltonen J, et al. Independent NF1 mutations in two large families with spinal neurofibromatosis. J Med Genet 2003;40:122-126.
Lázaro C, Ravella A, Gaona A, Volpini V, Estivill X. Neurofibromatosis type 1 due to germ-line mosaicism in a clinically normal father. N Engl J Med 1994;331:1403-1407.
Upadhyaya M, Majounie E, Thompson P, et al. Three different pathological lesions in the NF1 gene originating de novo in a family with neurofibromatosis type 1. Hum Genet 2003;112:12-17.
Castle B, Baser ME, Huson SM, Cooper DN, Upadhyaya M. Evaluation of genotype-phenotype correlations in neurofibromatosis type 1. J Med Genet 2003;40:e109-e109.
Leppig KA, Kaplan P, Viskochil D, Weaver M, Ortenberg J, Stephens K. Familial neurofibromatosis 1 microdeletions: cosegregation with distinct facial phenotype and early onset of cutaneous neurofibromata. Am J Med Genet 1997;73:197-204.
Messiaen LM, Callens T, Mortier G, et al. Exhaustive mutation analysis of the NF1 gene allows identification of 95% of mutations and reveals a high frequency of unusual splicing defects. Hum Mutat 2000;15:541-555.
Kaufmann D, Muller R, Bartelt B, et al. Spinal neurofibromatosis without cafe-au-lait macules in two families with null mutations of the NF1 gene. Am J Hum Genet 2001;69:1395-1400.
Evans DGR, Mason S, Huson SM, Ponder M, Harding AE, Strachan T. Spinal and cutaneous schwannomatosis is a variant form of type 2 neurofibromatosis: a clinical and molecular study. J Neurol Neurosurg Psychiatry 1997;62:361-366.
Fahsold R, Hoffmeyer S, Mischung C, et al. Minor lesion mutational spectrum of the entire NF1 gene does not explain its high mutability but points to a functional domain upstream of the GAP-related domain. Am J Hum Genet 2000;66:790-818.(Bruce R. Korf, M.D., Ph.D)
A 48-year-old, right-handed man was evaluated in the neurology clinic of this hospital because of weakness of his arms and legs.
When the patient was 18 years of age, a mass on the right side of his neck developed; a biopsy was performed at another hospital, and he was told he had von Recklinghausen's neurofibromatosis. At 45 years of age, pain and weakness in his left leg developed. A radiograph of the left knee performed elsewhere showed what were described as tumors, and he was referred to an orthopedist at this hospital. Magnetic resonance imaging (MRI) showed an enhancing lesion, 2 cm in diameter, on the sciatic nerve, extending from the ischial tuberosity to the deep peroneal nerve. Physical therapy was administered for the next 10 months, but the pain persisted. A repeated MRI study showed that the thickening of the sciatic nerve involved both the peroneal and tibial branches, extended to the lowest level in the images, and had increased in diameter to 3.6 cm. Three months later, increased pain and a foot drop on the left developed; MRI scanning of the brain and the cervical, thoracic, and lumbar spine performed elsewhere showed no abnormalities of the brain, but there were bilateral enhancing lesions in all cervical neural foramina, with cord compression at C1–2, C2–3, and C4–5 and bilateral rounded nonenhancing lesions in the neural foramina of the thoracic and lumbar spine. Treatment with gabapentin was begun.
During the following year, the patient began using a four-point cane and an orthosis for his left foot, and weakness developed in the right leg. He fell on numerous occasions. He began to notice alternating pallor and redness in his feet. An eye examination documented no sign of Lisch nodules. Weakness developed in the left arm and subsequently in the right arm.
About one year after starting treatment with gabapentin, the patient was referred to the neurosurgical clinic at this hospital. An MRI of the brain showed no abnormalities. An MRI of the cervical spine showed bilateral enhancing tumors at each level. At the C1–2 level, the cord was compressed from both sides. At the C4–5 level, the cord was displaced into the right canal and indented. An MRI of the thoracic and lumbar spine showed expanded nerve roots bilaterally at every level without cord compression; moderate stenosis of the canal was seen at the L4–5 level. A computed tomographic (CT) scan of the abdominal area showed multiple tumors involving the pelvic and paraspinal nerves and multiple subcutaneous masses. The patient was then referred to the neurology clinic.
He characterized his primary problem as weakness, and he also had pain, primarily in his left leg, that was worse with activity. He had no problems with bowel or bladder control and no back pain, headaches, or vision problems. He did not drink or smoke and had been unemployed for four years. His medications were sertraline and gabapentin.
The patient's mother died at 87 years of age of lymphoma and his father at 47 years of age of smoking-related emphysema. His mother had had a lump on the back of her neck; he did not recall either parent having skin lesions. His brother was alive at 46 years of age with liver cancer and multiple sclerosis; his sister had died suddenly at the age of 39, and the patient did not know the cause. He did not recall either sibling having skin lesions. His sister's son was said to have neurofibromatosis. The patient was married and had one son, who was well.
He was alert and answered questions clearly. There were no dysmorphic features. The vital signs were normal. Both hands and feet were reddened and cold at the beginning of the visit; by the end of the visit, the patient's feet were deep red up to the middle of the tibia. Examination of the skin under ambient light and with a Wood's lamp revealed two café-au-lait macules greater than 1.5 cm in diameter, one on the left lower part of the abdomen and another on the inner aspect of the left buttock. A single, raised, purple cutaneous lesion, 1 cm in diameter, was present on the left side of the chest. There was no axial or inguinal freckling. Multiple subcutaneous tumors were visible throughout the body. The left leg was slightly thicker than the right. An examination of the head, neck, chest, and abdomen revealed no abnormalities.
On neurologic examination, the patient's gait was slow and shuffling, and he walked bent at the waist. The cranial-nerve function was normal. On motor examination, strength in both arms was 4/5 with bilateral pronator drift. Fine hand–motor coordination was intact. Motor examination of his legs and feet showed bilateral increased tone with contractures at both heel cords. The right foot was held in a turned-in position. The strength was 4/5 in the right leg and 3/5 in the left leg, with decreased muscle mass in the left leg and left foot drop. Brachioradialis reflexes were 2+ and symmetric in the upper extremities, but on the right side there was spread into the biceps. The Hoffmann's sign was positive on the left and negative on the right. Palmomental reflexes were negative bilaterally. Knee and ankle reflexes were 3+. The Babinski reflex on the right was unreactive and on the left turned upward. The results of routine laboratory tests were normal.
Two weeks later, laminectomy of C3–6, with resection of bilateral C2, right C3, and left C5 tumors, was performed, as was fusion of the cervical spine. On pathological examination, all tumors were neurofibromas. Three months later, the patient's pain had diminished and his gait had improved; he was able to walk up a full flight of stairs at home and used a single-point cane only when outside. A new right foot drop had developed. Muscle strength in all extremities was 5/5 except for the ankle flexors, which were 4/5, and he was unable to flex either great toe upward. Muscle tone was slightly increased in the lower extremities, but less so than it had been at the first visit.
Three months after the patient's evaluation, his sister's son was also evaluated in the neurology clinic. He was a 29-year-old, right-handed man who had had flat, pigmented lesions on his skin as a teenager. When he was 19 years of age, a subcutaneous nodule appeared on his right shoulder; over the years, more nodules developed, one of which, on his left lower back, became painful. The painful nodule was removed, and pathological examination disclosed a neurofibroma. He had had weakness of his left leg when climbing stairs for three months before his evaluation. He had no other neurologic or physical symptoms.
The nephew thought his mother had had what he described as "brown birthmarks." At 39 years of age, she had fallen backward and lost consciousness; a craniotomy was performed at another hospital, but she died three days later. An autopsy reported the cause of death as blunt trauma to the head, with epidural, subdural, and subarachnoid hemorrhage. No skin lesions or nerve tumors were noted. His maternal half-brother and half-sister each had one brown skin lesion. His father was well.
Examination of the nephew's skin under ambient light showed freckles in the inguinal regions but not in the axillae. Multiple café-au-lait macules were present, many greater than 1.5 cm in diameter. There were multiple palpable subcutaneous nodules, two of which were painful, on the left leg and left shoulder. There were no cutaneous nodules. The results of a detailed neurologic examination revealed no abnormalities.
Diagnostic tests were performed on both the patient and his nephew.
Differential Diagnosis
Dr. Bruce R. Korf: This clinical history poses two major questions. First, what accounts for this man's weakness, both before and after his surgery? Second, what is his underlying diagnosis and how is his condition related to that in his nephew? May we review the MRIs?
Dr. John W. Henson: MRI scans were obtained through the spine, the lumbosacral plexus, and the brain before and after the administration of gadolinium. Parasagittal images revealed large masses along the courses of multiple, bilateral cervical nerve roots (Figure 1A). On axial images through the C2 vertebral body, these lesions had mild heterogeneity of signal intensity. A focus of T2-weighted hypointensity was present within the lateral aspect of the right C2 paraspinal mass lesion, forming a so-called "target sign" (Figure 1B). Heterogeneous gadolinium enhancement was seen. The tumors produced severe compression of the cervical spinal cord at the C2 level and extended laterally into the brachial plexus. There was mild enlargement of thoracic spinal nerve roots. Coronal fast spin-echo inversion-recovery sequences revealed masses along the courses of the lumbar nerve roots and lumbosacral plexus, with extensions along both sciatic nerves (Figure 1C).
Figure 1. Neuroimaging Studies Showing Multiple Peripheral Nerve Tumors.
A fast spin–echo (FSE) T2-weighted image in the sagittal plane from the cervical spine revealed the presence of multiple mass lesions (arrows) along the cervical spinal roots (Panel A). An axial FSE T2-weighted image at the level of the second cervical vertebral body (Panel B) showed the presence of cervical spinal cord compression by mass lesions along bilateral spinal nerve roots. The tumors were hyperintense on T2-weighted images and had a focal region of low signal consistent with the so-called "target sign" (arrow) and had heterogeneous enhancement after the administration of gadolinium. A coronal FSE inversion recovery image through the lumbosacral plexus (Panel C) showed plexiform mass lesions and diffuse enlargement of the sciatic nerves (arrows).
Multiple neurogenic tumors can be either neurofibromas or schwannomas.1,2 On imaging studies, neurofibromas usually show heterogeneous enhancement, whereas schwannomas usually enhance homogeneously. Approximately half of neurofibromas show the target sign, which is a central region of lower intensity surrounded by higher intensity on T2-weighted images. This is more common in plexiform neurofibromas but can rarely be present in schwannomas. Neurofibromas are more likely to produce diffuse or plexiform lesions than are schwannomas.
Imaging of the patient's brain revealed no sign of lesions that are considered typical of neurofibromatosis type 1, such as glioma of the optic pathway, foci of abnormal T2-weighted hyperintensity within the brain, plexiform neurofibromas of the ophthalmic division of the fifth cranial nerves, or dysplasia of the greater wing of the sphenoid bone. There was no evidence of nodular enhancement along the courses of the eighth cranial nerves to suggest the presence of vestibular schwannomas, as are seen in neurofibromatosis type 2.
The interpretation of the neuroradiologic images was that there were multiple neurofibromas, with a plexiform pattern of involvement.
Dr. Korf: This patient was given a diagnosis of von Recklinghausen's neurofibromatosis at 18 years of age. That was at least 30 years ago, before the formulation of diagnostic criteria for neurofibromatosis type 1 and neurofibromatosis type 2 in 1987.3 These two forms of neurofibromatosis were not always clearly distinguished before 1987, and we do not know whether any other signs of either type were present in this patient. His current symptoms of pain and weakness began three years before his evaluation here. Review of the first MRI obtained of his left leg reveals enlargement of the sciatic nerve, with an appearance typical of neurofibroma. A neurofibroma is a benign neoplasm involving various cellular elements of the nerve sheath. Neurofibromas may occur as focal growths along a nerve or at a nerve ending in the skin, or they can extend along a length of nerve and involve multiple fascicles of a larger nerve, forming a plexiform neurofibroma.4
Plexiform neurofibromas can be asymptomatic, but pain, weakness, and sensory dysfunction can occur, particularly if the nerve enlargement leads to compression. Subsequent MRI scanning revealed much more widespread disease, with multiple tumors of the cervical nerve roots, several of which were compressing the spinal cord. Weakness increased over a period of about a year and spread to include his right leg and arm.
A neurologic examination revealed signs of both upper- and lower-motor-neuron dysfunction. The foot drop on the left suggests a peroneal neuropathy. Increased tone, brisk deep-tendon reflexes, and a positive Hoffmann's sign and Babinski's sign suggest cord compression. Vasomotor dysfunction in the lower part of the legs may be further signs of peripheral-nerve dysfunction. The patient had marked improvement in strength after resection of several cervical tumors, but a newly developed foot drop on the right suggests peroneal-nerve dysfunction on that side. His pain and weakness are likely to be the combined results of spinal cord compression at multiple levels and peripheral-nerve dysfunction in the legs, probably due to compression of massively enlarged nerves.
Classification and Diagnosis of Neurofibromatosis
This patient has a disorder that leads to multiple tumors of the peripheral-nerve sheath. Some form of neurofibromatosis would seem most likely. There are two widely recognized forms of neurofibromatosis, type 1 and type 2 (Table 1). Within the past decade, an additional entity referred to as schwannomatosis5 has been added to this short list, but since the tumors in this patient were neurofibromas, not schwannomas, we can exclude this condition. Neurofibromatosis type 2 is also unlikely. Although neurofibromas can occur in persons with neurofibromatosis type 2, schwannomas are far more common. The hallmark lesion, the vestibular schwannoma, was not present on cranial MRI studies of this patient.
Table 1. Classification of the Forms of Neurofibromatosis.
The presence of multiple neurofibromas in this patient fulfills one criterion for neurofibromatosis type 1 (Table 2),3,6 but at least two are required to establish the diagnosis. He had two café-au-lait spots, not the requisite six, and did not have axillary or inguinal freckles, Lisch nodules (iris hamartomas that are found in most adults with neurofibromatosis type 1 and that are highly specific signs of the disorder), skeletal dysplasia, or optic glioma. There was no definitive diagnosis of neurofibromatosis type 1 in a first-degree relative.
Table 2. Diagnostic Criteria for Neurofibromatosis Type 1.
There are variant forms of neurofibromatosis (Table 3). Familial multiple café-au-lait spots7 and Watson's syndrome8 clearly do not fit this patient's history. Patients with segmental neurofibromatosis have symptoms confined to a particular region of the body.9 This localized form of the disease is attributable to somatic mosaicism for a neurofibromatosis type 1 mutation, at least in some cases. The disease in the patient under discussion was too widespread to be described as segmental.
Table 3. Variant Forms of Neurofibromatosis Type 1.
Ten patients10 have been described who had generalized neurofibromas but who did not display other features of neurofibromatosis type 1. In this group, both isolated and plexiform neurofibromas were seen, including spinal-nerve-root tumors, some of which caused cord compression. Some patients had dermal neurofibromas, whereas others had firm subcutaneous masses similar to those reported in this patient. It is unclear whether any of the patients in the report had as many tumors as the patient in this case.
The other relevant variant of neurofibromatosis is familial spinal neurofibromatosis.11,12,13 Patients with this condition have neurofibromas that originate from spinal nerve roots, often at every level, similar to the findings in the patient under discussion. Cord compression occurs commonly. Most patients have multiple café-au-lait spots, some have iris Lisch nodules, and skin-fold freckling is variable. Most patients have disease with features that fulfill the diagnostic criteria for neurofibromatosis type 1. However, typical dermal neurofibromas are usually absent in these patients, even in adulthood, when this tumor is apt to be found in patients with classic neurofibromatosis type 1. Mutations in the NF1 gene have been identified in several families.11,12,14
The patient in this case fits somewhere between these two entities — he has spinal tumors similar to those seen in patients with familial spinal neurofibromatosis, but he lacks multiple café-au-lait spots. I am inclined to classify him with the patients who have familial spinal neurofibromatosis because of his spinal tumors. Some patients with spinal neurofibromatosis have been described as having peripheral tumors.11 I suspect that the two entities of familial spinal neurofibromatosis and generalized neurofibromatosis without other signs may represent a spectrum of manifestations of one disease. In most patients — but not all — with spinal neurofibromatosis, findings have shown linkage to the NF1 locus or there are NF1 gene mutations, whereas molecular studies of individual patients with isolated generalized neurofibromas have not been reported.
Familial Spinal Neurofibromatosis
Does this patient have a familial disorder? His nephew is said to have "many" café-au-lait macules and had inguinal but not axillary freckling; both are usually seen in adults with neurofibromatosis type 1. The nephew had multiple painful subcutaneous neurofibromas, like the patient, and no dermal tumors. We do not know if he had spinal tumors, but he had weakness of the left leg. I believe that the nephew had the same disorder as our patient and that he would prove to have spinal neurofibromas if MRI were performed.
Although neurofibromatosis is a highly variable disorder, penetrance is complete, so I would not expect to see a skipped generation. Did the patient's sister, the mother of the nephew, also have this disorder? She died suddenly, reportedly with evidence of blunt head trauma. She is also said to have had "brown spots." Thus, it is possible that she also had neurofibromatosis. If the patient and his sister both had neurofibromatosis, one parent should also have been affected. The patient's mother is said to have had a "neck mass." Either parent could have been affected without the disease ever being diagnosed, given the paucity of externally visible signs in this patient. Alternatively, one of the patient's parents could have had mosaicism for a mutation that could have been transmitted to two children without affecting the parent, as has been documented for at least one family with neurofibromatosis type 1.15
In respect to this patient and his nephew, it is possible that each has a form of neurofibromatosis due to a distinct and independent gene mutation. Multiple different NF1 gene mutations in different members of the same extended family have been reported.16 I believe that this is unlikely in this case, since both the patient and his nephew have a similar disorder with a paucity of cutaneous signs and many subcutaneous and deep neurofibromas.
Genetic Testing in Neurofibromatosis
The patient and his nephew underwent a "diagnostic test." I do not believe that much more would have been learned from pathological examination of tumors. It would be valuable to obtain an MRI of the spine in the nephew, but nothing would be learned from additional MRI studies of the patient. The diagnostic test was therefore probably a molecular genetic test of the NF1 gene, searching for a gene mutation.
Until recently, there was no practical clinical diagnostic test for neurofibromatosis type 1. The gene is large (60 exons) and mutations are widely scattered in the gene, so that almost no two families with neurofibromatosis type 1 share the same mutation.17 This disorder has a high mutation rate, and hundreds of distinct mutations have been reported. There are some recurrent mutations, but no "hot spots" to guide clinical testing. There are no major genotype–phenotype correlations, except that patients who lack the gene entirely tend to have a severe phenotype with early onset and large numbers of neurofibromas, developmental delay, and dysmorphic facial appearance.18
Mutation analysis for neurofibromatosis type 1 is now available for clinical testing, with the use of a set of complementary techniques to find the various types of mutation, including deletion of the entire gene, small insertions or deletions that cause a frame shift, single-base changes that lead to a stop codon or amino acid substitution, and mutations that alter RNA splicing.19 Most of the mutations lead to premature termination of peptide synthesis and hence obliterate the function of the gene product. The fact that the mutations result in loss of function is compatible with the notion that NF1 functions as a tumor-suppressor gene in the pathogenesis of neurofibromas, requiring loss-of-function mutations in both alleles to produce a tumor.
Mutation testing has been performed in several patients with familial spinal neurofibromatosis (Figure 2). Of nine mutations reported, four are amino acid substitutions, which is remarkable since not more than 10 percent of all NF1 mutations have been found to be missense changes. Three are splicing mutations and only two generate stop codons, one by frame shift and one by nonsense mutation.11,12,14,20
Figure 2. Location of NF1 Gene Mutations Reported in Familial Spinal Neurofibromatosis.
The GAP (GTPase activating protein)-related domain, where the mutations occur, is labeled according to the source where the mutation was described. IVS5–2AG, L920P, P2200A, and S2698X were described by Kluwe et al.11; 8042insA by Ars et al.12; L357P and IVS39+3AG by Messiaen et al.14; and IVS31–5AG and L2067P by Kaufmann et al.20
Diagnosis
This patient had a mixed upper-motor-neuron dysfunction due to cord compression at multiple levels and lower-motor-neuron disease due to nerve compression. I believe the underlying cause was the presence of multiple spinal and plexiform neurofibromas caused by a variant of neurofibromatosis type 1 similar to familial spinal neurofibromatosis. I believe that both the patient and his nephew have the same disorder, and that both will be shown to have an NF1 gene mutation, probably a splicing defect or missense mutation.
Dr. Nancy Lee Harris (Pathology): Dr. MacCollin, can you give us your clinical impressions?
Dr. Mia MacCollin (Neurology): At the time the patient was first evaluated in the neurofibromatosis clinic, we who saw him believed that he had the rare syndrome of multiple isolated neurofibromas as described by Blakley et al.10 We thought that the nephew's illness was due to the relatively common occurrence of a new mutation in the NF1 gene and that his disease was genetically unrelated to his uncle's rare syndrome.
Clinical Diagnosis
Multiple isolated neurofibromas.
Dr. Bruce R. Korf's Diagnosis
Neurofibromatosis type 1 variant, similar to familial spinal neurofibromatosis.
Pathological Discussion
Dr. Harris: Dr. Smith, could you describe your operative findings?
Dr. Edward R. Smith (Surgery, Children's Hospital, Boston): A multilevel cervical laminectomy was performed, which revealed very large extradural tumors at C2 that were touching at midline. These tumors were resected, as was another extradural mass at C3. The dura was opened, revealing a sizable C5 tumor that was completely displacing the spinal cord to one side; it was removed. At the completion of the operation, despite the presence of abundant space in the canal, the spinal cord retained the impressive deformation shown on the preoperative radiographs, with a thin, effaced high cervical cord. The dura was closed, and a multilevel fusion with a lateral mass screw construct was performed.
Dr. Anat Stemmer-Rachamimov: The tumors were hypocellular, with myxoid stroma, a heterogeneous cell population of round cells and wavy cells, and entrapped large nerve bundles (Figure 3A). Only scattered Schwann cells, which are the neoplastic cells, were positive for S100 protein (Figure 3B), and neurofilament immunostaining highlighted enlarged nerve bundles and dispersed axons within the tumors (Figure 3C). The pathological diagnosis was multiple plexiform spinal-nerve-root neurofibromas, with no malignant transformation. The nephew's specimen was also a neurofibroma.
Figure 3. Specimens from Spinal-Nerve-Root Tumors.
The tumor is composed of spindled and round cells with entrapped axons (Panel A; hematoxylin and eosin). With immunostaining for S100 protein (Panel B), only scattered neoplastic Schwann cells are stained. The other cells are non-neoplastic stromal and nerve cells. Enlarged nerve bundles and dispersed axons within the tumors are highlighted with neurofilament immunostaining (Panel C). Multiple spinal neurofibromas at autopsy (Panel D, from another patient) show markedly enlarged spinal nerve roots, seen from the thoracic cavity, which are symmetrically involved by plexiform neurofibromas. (Autopsy photograph courtesy of Dr. Matthew Frosch, Department of Pathology, MGH.)
Patients with multiple spinal neurofibromatosis often have spinal tumors that are distributed symmetrically and involve many or all spinal roots (Figure 3D).11 In the absence of other clinical features of neurofibromatosis type 1, the pathological distinction between schwannomas and neurofibromas is important, because multiple spinal-root-tumors may be seen in schwannomatosis and neurofibromatosis type 2.21 Schwannomas are also tumors of Schwann cells, but they lack the heterogeneous cell population found in neurofibromas. This distinction may be difficult to make when schwannomas have prominent myxoid changes that mimic the histological appearance of neurofibromas. In these cases, it is helpful to find areas with classic schwannoma features, or to establish by S100 immunostaining or electron microscopy that only a subgroup of cells are Schwann cells.
Dr. Harris: Dr. MacCollin, will you tell us about the additional diagnostic testing?
Dr. MacCollin: A tissue sample from the patient was sent to the laboratory of Dr. Ludwine Messaien at the Center of Medical Genetics, Ghent University Hospital in Belgium. No alterations were found in the NF1 gene by protein-truncation testing. Direct sequencing of the gene revealed an unusual alteration in exon 22 (G3827A, leading to the substitution of glutamine for arginine at residue 1276). This change had previously been detected in a single sample from an unrelated 19-year-old patient with an unusually large load of cutaneous and plexiform neurofibromas. It had also been reported twice22 in a series of 301 sequence variants in the NF1 gene. It occurs at a CpG hypermutatable site that is also involved in the more typical change, C3826T, which causes a nonsense mutation at arginine 1276. The patient's nephew had the same mutation. Subsequent MRI examinations showed that the patient's nephew has multiple spinal nerve root plexiform neurofibromas, without cord compression. We have been unable to analyze the autopsy specimens from the patient's sister.
Anatomical Diagnosis
Neurofibromatosis type 1, affecting the patient and his nephew, with multiple spinal neurofibromas.
Dr. Korf reports having received lecture fees from Pharmacia.
Source Information
From the Department of Genetics, University of Alabama at Birmingham (B.R.K.); and the Departments of Radiology and Neurology (J.W.H.) and Pathology (A.S.-R.), Massachusetts General Hospital and Harvard Medical School — both in Boston.
References
Khong PL, Goh WH, Wong VC, Fung CW, Ooi GC. MR imaging of spinal tumors in children with neurofibromatosis 1. AJR Am J Roentgenol 2003;180:413-417.
Suh JS, Abenoza P, Galloway HR, Everson LI, Griffiths HJ. Peripheral (extracranial) nerve tumors: correlation of MR imaging and histologic findings. Radiology 1992;183:341-346.
Stumpf DA, Alksne JF, Annegers JF, et al. Neurofibromatosis: conference statement: National Institutes of Health Consensus Development Conference. Arch Neurol 1988;45:575-578.
Korf BR. Plexiform neurofibromas. Am J Med Genet 1999;89:31-37.
MacCollin M, Woodfin W, Kronn D, Short MP. Schwannomatosis: a clinical and pathologic study. Neurology 1996;46:1072-1079.
Gutmann DH, Aylsworth A, Carey JC, et al. The diagnostic evaluation and multidisciplinary management of neurofibromatosis 1 and neurofibromatosis 2. JAMA 1997;278:51-57.
Arnsmeier SL, Riccardi VM, Paller AS. Familial multiple cafe au lait spots. Arch Dermatol 1994;130:1425-1426.
Tassabehji M, Strachan T, Sharland M, et al. Tandem duplication within a neurofibromatosis type 1 (NF1) gene exon in a family with features of Watson syndrome and Noonan syndrome. Am J Hum Genet 1993;53:90-95.
Ruggieri M, Huson SM. The clinical and diagnostic implications of mosaicism in the neurofibromatoses. Neurology 2001;56:1433-1443.
Blakley P, Louis DN, Short MP, MacCollin M. A clinical study of patients with multiple isolated neurofibromas. J Med Genet 2001;38:485-488.
Kluwe L, Tatagiba M, Funsterer C, Mautner VF. NF1 mutations and clinical spectrum in patients with spinal neurofibromas. J Med Genet 2003;40:368-371.
Ars E, Kruyer H, Gaona A, et al. A clinical variant of neurofibromatosis type 1: familial spinal neurofibromatosis with a frameshift mutation in the NF1 gene. Am J Hum Genet 1998;62:834-841.
Pulst S-M, Riccardi VM, Fain P, Korenberg JR. Familial spinal neurofibromatosis: clinical and DNA linkage analysis. Neurology 1991;41:1923-1927.
Messiaen L, Riccardi V, Peltonen J, et al. Independent NF1 mutations in two large families with spinal neurofibromatosis. J Med Genet 2003;40:122-126.
Lázaro C, Ravella A, Gaona A, Volpini V, Estivill X. Neurofibromatosis type 1 due to germ-line mosaicism in a clinically normal father. N Engl J Med 1994;331:1403-1407.
Upadhyaya M, Majounie E, Thompson P, et al. Three different pathological lesions in the NF1 gene originating de novo in a family with neurofibromatosis type 1. Hum Genet 2003;112:12-17.
Castle B, Baser ME, Huson SM, Cooper DN, Upadhyaya M. Evaluation of genotype-phenotype correlations in neurofibromatosis type 1. J Med Genet 2003;40:e109-e109.
Leppig KA, Kaplan P, Viskochil D, Weaver M, Ortenberg J, Stephens K. Familial neurofibromatosis 1 microdeletions: cosegregation with distinct facial phenotype and early onset of cutaneous neurofibromata. Am J Med Genet 1997;73:197-204.
Messiaen LM, Callens T, Mortier G, et al. Exhaustive mutation analysis of the NF1 gene allows identification of 95% of mutations and reveals a high frequency of unusual splicing defects. Hum Mutat 2000;15:541-555.
Kaufmann D, Muller R, Bartelt B, et al. Spinal neurofibromatosis without cafe-au-lait macules in two families with null mutations of the NF1 gene. Am J Hum Genet 2001;69:1395-1400.
Evans DGR, Mason S, Huson SM, Ponder M, Harding AE, Strachan T. Spinal and cutaneous schwannomatosis is a variant form of type 2 neurofibromatosis: a clinical and molecular study. J Neurol Neurosurg Psychiatry 1997;62:361-366.
Fahsold R, Hoffmeyer S, Mischung C, et al. Minor lesion mutational spectrum of the entire NF1 gene does not explain its high mutability but points to a functional domain upstream of the GAP-related domain. Am J Hum Genet 2000;66:790-818.(Bruce R. Korf, M.D., Ph.D)