Brain Advance Access published online on August 6, 2008
Brain, doi:10.1093/brain/awn178
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Adult-onset Alexander disease: a series of eleven unrelated cases with review of the literature
1Biochemistry and Genetics Unit, IRCCS Foundation, C. Besta Neurological Institute, Milan, 2Department of Neurology and Centre for Experimental Neurological Therapy, S. Andrea Hospital, University of Rome, Rome, 3Neurooncology Unit, 4Movement Disorders Unit, 5Neuropathology Unit, 6Neuroradiology Unit, 7Cerebrovascular Diseases Unit, IRCCS Foundation, C. Besta Neurological Institute, Milan and 8Laboratory of Molecular Genetics, G. Gaslini Institute, Genoa, Italy
Correspondence to:
Davide Pareyson, MD, Biochemistry and Genetics Unit, IRCCS Foundation, C. Besta Neurological Institute, Via Celoria 11, 20133 Milan, Italy E-mail: dpareys{at}istituto-besta.it
 |
Summary |
|---|
 Top Summary IntroductionPatients and MethodsResultsDiscussionConcluding remarksReferences |
|---|
Alexander disease (AD) in its typical form is an infantile lethal leucodystrophy, characterized pathologically by Rosenthal fibre accumulation. Following the identification of glial fibrillary acidic protein (GFAP) gene as the causative gene, cases of adult-onset AD (AOAD) are being described with increasing frequency. AOAD has a different clinical and neuroradiological presentation with respect to early-onset AD, as abnormalities are mainly concentrated in the brainstem–spinal cord junction. We report detailed clinical and genetic data of 11 cases of AOAD, observed over a 4-year period, and a review of the previously reported 25 cases of genetically confirmed AOAD. In our series, onset occurred as late as age 62, and up to 71 in an affected deceased relative. Most cases appeared sporadic, but family history may be misleading. The most frequent symptoms were related to bulbar dysfunction—with dysarthria, dysphagia, dysphonia (seven patients)—, pyramidal involvement (seven patients) and cerebellar ataxia (seven patients). Four patients had palatal myoclonus. Sleep disorders were also observed (four cases). Bulbar symptoms, however, were infrequent at onset and two symptomatic patients had an almost pure pyramidal involvement. Two subjects were asymptomatic. Misdiagnosis at presentation was frequent and MRI was instrumental in suggesting the correct diagnosis by showing, in all cases, mild to severe atrophy of the medulla oblongata extending caudally to the cervical spinal cord. In ten patients, molecular studies revealed six novel missense mutations and three previously reported changes in GFAP. The last typical patient carried no definitely pathogenic mutation, but a missense variant (p.D157N), supposedly a rare polymorphism. Revision of the literature and the present series indicate that the clinical picture is not specific, but AOAD must be considered in patients of any age with lower brainstem signs. When present, palatal myoclonus is strongly suggestive. Pyramidal involvement, cerebellar ataxia and urinary disturbances are common. Less frequent findings include sleep disorders and dysautonomia. Fluctuations may occur. The course is variable, usually slowly progressive and less severe than the AD forms with earlier onset. AOAD is more common than previously thought and might even be the most common form of AD. The diagnosis is strongly suggested by MRI and confirmed by GFAP gene analysis.
Key Words: Alexander disease; GFAP; brainstem diseases; medulla oblongata atrophy; palatal myoclonus
Abbreviations: AD, Alexander disease; AOAD, adult-onset AD; FLAIR, fluid-attenuated inversion recovery; GFAP, glial fibrillary acidic protein
Received May 22, 2008. Revised June 24, 2008. Accepted July 10, 2008.
|
Introduction |
|---|
|
TopSummaryIntroductionPatients and MethodsResultsDiscussionConcluding remarksReferences |
|---|
Alexander disease (AD) is an infantile leucodystrophy first described in 1949 and characterized by macrocephaly, psychomotor regression, spasticity, ataxia and seizures leading to death in a few years (Li et al., 2005
; Quinlan et al., 2007). Rosenthal fibres, eosinophylic inclusions localized in astrocyte cytoplasm, are the pathologic hallmark of the disease and are mainly found in perivascular, periventricular and subpial spaces of cerebral hemispheres, cerebellum and brainstem (Quinlan et al., 2007). In typical infantile cases, MRI shows extensive white matter signal hyperintensities in T2-weighted images, more marked in the frontal regions; a rim of periventricular T2 hypointensity; involvement of basal ganglia, thalami and brainstem; contrast enhancement particularly of periventricular regions and brainstem (van der Knaap et al., 2001).
A few sporadic or familial cases of AD with later onset and predominant involvement of brainstem were pathologically diagnosed owing to Rosenthal fibre presence (Schwankhaus et al., 1995). The molecular basis of AD remained obscure until it was observed that the transgenic mouse overexpressing the glial fibrillary acidic protein (GFAP) develops Rosenthal fibres in great amount (Messing et al., 1998), making GFAP the candidate gene for AD. Indeed, heterozygous GFAP mutations were found in patients with AD, thus allowing a diagnosis of certainty in vivo (Brenner et al., 2001) and enlarging the phenotypic spectrum of the disease. GFAP mutations have been associated with the following: (i) typical infantile AD, with onset before age 2 years and rapid lethal course; (ii) the juvenile form of AD, characterized by onset between age 2 and 12 years, and running a slower course; (iii) adult-onset AD (AOAD), with symptoms beginning during adolescence or later (after age 12), predominant brainstem involvement and survival into adulthood (van der Knaap et al., 2001, 2005; Li et al., 2005). Patients with AOAD usually present with dysarthria, dysphonia, dysphagia, pyramidal signs and ataxia; palatal myoclonus is common (Li et al., 2005; Romano et al., 2007). Their MRI is characterized by atrophy and changes in signal intensity in the medulla oblongata and upper spinal cord, with inconstant supratentorial periventricular white matter abnormalities (Farina et al., 2008).
GFAP mutations frequently occur de novo, particularly in infantile cases, while in AOAD both sporadic cases with de novo mutations and familial cases with autosomal dominant transmission have been described (Li et al., 2006). The mutations thus far reported involve highly conserved domains. The issue of genotype–phenotype relationship is a matter of debate, as some mutations have been associated with all the three forms of AD (Li et al., 2005; Quinlan et al., 2007).
We report a series of 11 AOAD patients observed over a course of 4 years. Diagnosis was based on the typical MRI appearance and confirmed by GFAP mutation analysis in 10 cases. Patients 1 and 2 had been reported in previous reports (Romano et al., 2007; Salmaggi et al., 2007). The neuroradiological features of the present series of patients have been analysed in a separate paper (Farina et al., 2008), with a brief summary of clinical data in a table. Here, we present in detail the clinical phenotypes along with genetic findings. We also review the clinical findings of previously reported AOAD cases with identified GFAP mutations.
|
Patients and Methods |
|---|
|
TopSummaryIntroductionPatients and MethodsResultsDiscussionConcluding remarksReferences |
|---|
Patients
Nine patients were recruited at the IRCCS Foundation, C. Besta Neurological Institute between 2004 and 2007 and two patients were referred from other hospitals for a neuroradiological consultation in the same period. All had MRI suggestive of AOAD, i.e. T2-weighted hyperintensities and atrophy in medulla oblongata and upper cervical spinal cord. Family and medical history were collected and clinical examination performed. The results of the various exams performed during the diagnostic workup were reviewed.
Methods
Neuroradiology
All the patients had at least one MRI study of the brain and cervical spinal cord; complete spinal cord examination was obtained in three patients.
The examinations of the brain included T1-weighted, proton density and T2-weighted images, and fluid-attenuated inversion recovery (FLAIR) sequences in the three orthogonal plains. Post-contrast examinations were obtained in all patients but one.
Six patients were studied with diffusion-weighted imaging (DWI) with generation of mean diffusivity maps. Six patients were studied with proton MR spectroscopy, by means of a short echo-time multivoxel point-resolved sequence, with a section positioned on the centra semiovalia. In three patients, a single voxel study was attempted on the medulla oblongata.
MRI examinations of the spinal cord were performed with sagittal T1- and T2-weighted, and with axial T2*-weighted sequences. Post-contrast sagittal T1-weighted images were also available.
Molecular studies
A mutational screening was performed on DNA extracted from peripheral blood lymphocytes of the patients using primers, amplification conditions and a protocol already reported (Caroli et al., 2007). In particular, the nine coding exons of the GFAP gene were analysed, along with intronic flanking segments, by means of direct DNA sequencing. Each nucleotide variant thus detected was tested in 50 unrelated healthy individuals (100 control chromosomes), a comparison that made it possible to distinguish between disease-causing mutations and neutral common variants.
Review of the literature
We reviewed the literature of all reported AOAD cases with genetic analysis of the GFAP gene. We decided to include only patients with genetic diagnostic confirmation and to exclude those with only neuropathological diagnosis, as Rosenthal fibres may be found in different disorders and diagnosis might be disputed (Howard et al., 1993).
Search strategy included the following: (i) PubMed search with ‘Alexander disease’, ‘GFAP’ or ‘adult-onset’ used as keywords for articles up to June 18, 2008; (ii) the mutation database of the Waisman Center at http://www.waisman.wisc.edu/alexander/; (iii) the ‘Human Intermediate Filament Database’ of the Centre for Molecular Medicine and the Bioinformatics Institute in Singapore at http://www.interfil.org/details.php?id=NM_002055.
|
Results |
|---|
|
TopSummaryIntroductionPatients and MethodsResultsDiscussionConcluding remarksReferences |
|---|
Clinical findings
The clinical findings of the 11 patients are reported in Table 1 and the instrumental findings in Table 2.
|
|
Patients’ age ranged between 26 and 64 years, and age at onset between 13 and 62 years. Disease duration ranged between 4 months and 13 years in the nine symptomatic patients. Family history was positive in four cases (Patients 2, 7, 9 and 11). The mother of Patient 2 had died of a similar disorder, and her 6-year-old son had some seizures and carries the same mutation (Salmaggi et al., 2007
). One son of Patient 7 had a diagnosis of cerebellar pilocytic astrocytoma, which might well be a manifestation of AD as a posterior fossa tumour-like lesion with Rosenthal fibres (van der Knaap et al., 2005), while the other had vocal cord palsy, which also might be related to the disease. We recommended re-evaluation for possible AD for both. The mother of Patient 9 had been diagnosed with multiple sclerosis and died at age 57 years due to complications related to the neurological disease. The father of Patient 11 had been affected by an undiagnosed neurological disease with tetraparesis, previously attributed to spondilogenic cervical myelopathy, and had a typical MRI picture; he developed dysphagia and died at age 77 after 6 years of disease. All the other cases (Patients 1, 3–6, 8 and 10) were seemingly sporadic. However, the asymptomatic mother of Patient 1, aged 52, was recently found to carry the same GFAP mutation as her son.
Two patients had no symptom of AD and came to clinical and neuroradiological observation because of apparently unrelated conditions: Patient 5 because of chronic headache, with mixed tension and migraine features, and behavioural abnormalities with analgesic drug abuse, and Patient 9 after a subarachnoid haemorrhage of undetermined origin with good recovery; both had normal neurological examination. Although facial pain syndromes have been reported in AOAD (Balbi et al., 2008), we deemed that headache was not linked to AOAD in Patient 5. Onset symptoms in the other nine patients were as follows: lower limb weakness likely due to pyramidal involvement in four patients, bulbar symptoms (dysarthria, dysphonia and dysphagia) in three, upper limb wasting and weakness in one and urinary sphincter disturbances in one.
At last examination, seven patients had abnormalities of ocular movements (nystagmus, saccadic pursuit, diplopia or mild upper lid ptosis), seven had evidence of lower brainstem involvement (dysarthria, dysphagia or dysphonia) and four had palatal myoclonus (which also involved the tongue in Patients 4 and 6, and the pharynx in Patient 4). In Patient 3, palatal myoclonus was absent at age 25 and developed during the follow-up period at age 26. Patient 7 had been operated on for vocal cord palsy in adduction position that impaired respiration.
All the nine symptomatic subjects had some degree of gait difficulties because of spastic paraparesis (five cases) and/or ataxia (seven cases). Muscle wasting or weakness was observed in upper (five patients) and lower limbs (six patients). Babinski sign was often present (seven cases). Deep tendon reflexes were more frequently increased (five cases) than decreased (two patients). Motor impairment was sometimes asymmetric (Patients 1, 2, 8 and 10). Sensation was usually preserved, and only two patients (Patients 1 and 11) had decreased vibration sense. Six patients had bladder dysfunction, i.e. urgency, incontinence or retention. Only Patient 3 had clear-cut dysautonomia, with decreased sweating, hot intolerance with episodes of hyperthermia and hypotension; Patient 7 had impotence and Patient 11 had mild abnormalities at cardiovascular tests. Four patients had sleep disorders, including sleep apnoeas (three cases) and restless leg syndrome (one case). Two patients had scoliosis, severe in Case 3. Patient 1 had had two episodes of tonic–clonic seizures at age 3; their relation with AD had remained uncertain. None of the other patients had seizures. Interestingly, Patient 4 was subsequently seen at another centre and an attempt to treat palatal myoclonus with sodium valproate induced reversible Parkinsonism (Sechi et al., 2008).
The disease ran a fluctuating course in two patients (Patients 1 and 8), particularly in the onset phase, whereas it was usually progressive in the others. Disease progression rate was variable, and two patients (Patients 3 and 7) needed aids for walking or for daily life activities, four (Patients 2, 4, 10 and 11) had moderate difficulties, whereas three (Patients 1, 6 and 8) were relatively mildly affected.
Cognition was usually well preserved and only one patient had borderline scores at the Milan Overall Dementia Assessment.
EMG showed mild chronic neurogenic abnormalities in six cases. Myoclonus rhythm ranged from 1.5–2 (Patient 4) to 2–3 Hz (Patient 6). At motor-evoked potential studies, central conduction time was increased in two patients and borderline in another one. Seven out of eight patients had abnormalities of somatosensory-evoked potentials consistent with delayed transmission along the central sensory pathways. Brainstem auditory-evoked potentials were abnormal in all the seven patients who underwent the study, including one asymptomatic subject, and showed abnormalities of all waves or limited to waves I and III. Visual-evoked potentials, performed in six patients, showed mildly to moderately prolonged P100 latencies in four cases. Three out of seven patients showed mild EEG abnormalities, i.e. low-frequency activity in frontal, central or temporal regions. Polysomnography was performed in five patients and showed mild to severe apnoeas in three cases (Patients 6, 7 and 11). Autonomic system evaluation (cardiovascular reflexes) was abnormal in Patients 3 and 11.
A lumbar puncture was performed in seven cases (Patients 1–4, 7, 8, 10) and revealed no CSF abnormalities. Muscle biopsy revealed slight predominance of Type I fibres in Patient 3, and mild neurogenic changes and few Cox-depleted fibres in Patient 8. Lactate serum levels were normal in Patients 1–4. Minor laboratory abnormalities included the following: increased serum IgA (Patient 7), mild increase in serum alpha-fetoprotein level (Patient 9), slightly increased serum CK (Patient 10) and mild decrease in serum testosterone level (Patient 7, who had impotence).
Neuroradiologic results
Detailed neuroradiologic findings have been reported in a separate paper (Farina et al., 2008) and are summarized in Table 2 (Fig. 1). Mild to severe atrophy of the medulla oblongata extending caudally to the cervical spinal cord was present in all patients. In these structures, signal abnormalities were also observed either diffuse or with patchy distribution; in a few cases, the corticospinal tract and the medial lemniscus were selectively involved. The hilum of the dentate nucleus was hyperintense in T2-weighted images in eight cases. Minimal to moderate supratentorial periventricular white matter abnormalities were found in eight patients; they were absent in the three oldest patients. The irregular ventricular profile described as the ‘garland’ by van der Knaap et al. (2006) was present in only two cases. Small areas of post-contrast enhancement were observed in five patients. Except for one case, enhancement was absent in the oldest patients of the series.
|
Mean diffusivity was not altered except in the abnormal periventricular white matter. Increase in myo-inositol was the only abnormality found on spectroscopy, also restricted to the abnormal periventricular white matter.
Molecular studies
All the patients under analysis carried at least one heterozygous missense nucleotide change of the GFAP gene, as shown in Table 3.
|
GFAP mutations, previously reported in AD patients but not in healthy individuals, were found in Patients 3 and 6 (p.R416W and p.R70W, respectively). Patient 7 carried a missense mutation (p.E332K), already reported as part of a complex allele (p.[R330G; E332K]) and demonstrated to display a mild functional effect on the induction of GFAP aggregates (Balbi et al., 2008
; Bachetti et al., 2008). Finally, the p.S398Y missense change, listed as an unpublished mutation in the GFAP mutation database of the Waisman Center (http://www.waisman.wisc.edu/alexander/), was detected in Patient 9. Novel mutations, resulted absent in 100 control chromosomes, were found in Patients 1, 2, 4, 5, 8 and 10, with the former three individuals already reported to carry p.L359P, p.S393I and p.R70Q, respectively (Caroli et al., 2007; Romano et al., 2007; Salmaggi et al., 2007). The apparently unrelated Patients 5 and 8 both carried the p.E205K mutation.
Patient 11 did not show any apparently causative mutation in coding regions. In addition, mutations in the promoter region, in exon–intron boundaries and in the large 3’UTR were also excluded (Bachetti et al., manuscript in preparation). However, in this patient we could detect a p.D157N missense variant, an aminoacid change already reported even in unaffected parents (Li et al., 2005), in control subjects (Li et al., 2006), but not in our control chromosomes set. These observations suggest that the p.D157N nucleotide change could be a polymorphism, though a disease-related mutation with incomplete penetrance cannot be excluded. The p.D157N variant was also detected in Patient 4, who also showed the p.R70Q mutation. Unfortunately, the phase of the two mutations could not be determined, owing to unavailability of the father, deceased. The mother did not carry any mutation; therefore, each variant could have been either transmitted from the father or occurred de novo.
Patients 1, 5 and 8, already known to carry disease-causing mutations, also showed the p.P47P variant, which, like p.D157N, had already been detected in unaffected parents (Li et al., 2005), in control subjects (Li et al., 2006), but also in our 100 control chromosomes set with a frequency of 8%.
The mother of Patient 1 carried the same p.L359P mutation, but was still asymptomatic at age 52. Unfortunately, because of unavailability of most of the patients’ parents, de novo or inherited occurrence could not be assessed for the other nine disease-related mutations.
Review of the literature
A total of 18 articles were reviewed and a total of 25 previously reported cases with evidence of AOAD and GFAP gene mutations identified. Sixteen were familial (from seven families) and nine were sporadic; 13 were females and 12 males; four further subjects, three females and one male, aged 30–64 years, were asymptomatic although they carried a GFAP mutation. Clinical and genetic findings are summarized in Table 4.
|
|
Discussion |
|---|
|
TopSummaryIntroductionPatients and MethodsResultsDiscussionConcluding remarksReferences |
|---|
We diagnosed the adult form of AD in 11 patients over a 4-year period (2004–07) and diagnosis was genetically confirmed in 10 cases. Until 2002, diagnosis required brain biopsy or was autoptic. Following the identification of GFAP as the causative gene of AD, AOAD cases are being increasingly reported. Including our cases, 33 symptomatic patients with AOAD and proven GFAP mutations have been reported thus far, indicating that AOAD is relatively frequent and probably until recently underdiagnosed. During the same period, we observed at our Institute in Milan three typical infantile AD cases (G. Uziel, personal communication). In the same 4-year period, the dedicated molecular laboratory in Genoa identified GFAP mutations in nine early-onset cases (including two from our Institute) (Caroli et al., 2007
) versus 12 adult index cases (10 of this series) (Balbi et al., 2008). Although these data suggest that AOAD might be the most common form of AD, they should not be considered representative of the whole AD population and further series are awaited to assess the AD epidemiology.
Thus far, MRI is the best diagnostic tool, since it provides the demonstration of atrophy and signal changes of the medulla oblongata and upper spinal cord, highly characteristic of AOAD, thus enabling us to select the patients to be submitted to genetic analysis (Farina et al., 2008).
Clinical aspects
Diagnosis is difficult on clinical grounds alone. Previous diagnoses in our patients included multiple sclerosis, primary lateral sclerosis, amyotrophic lateral sclerosis, myasthenia gravis, multisystem atrophy, spinocerebellar ataxia and cervical myelopathy. In our series, clearly familial cases were a minority; however, family history is often reported to be normal because of misdiagnosis in affected relatives, including spondilogenic cervical myelopathy and multiple sclerosis. Reduced penetrance of GFAP mutations, as in the family of our Patient 1, is another explanation of this phenomenon.
Lower brainstem involvement is the most characterizing clinical finding (it was reported in 23 out of 25 symptomatic cases in the literature and in seven of our nine symptomatic patients), with dysarthria, dysphonia, vocal cord palsy, rhinolalia, dysphagia and may suggest the diagnosis. Since palatal myoclonus is rarely observed in clinical practice, its occurrence is strongly suggestive of AOAD, where it has been reported in 41% of the patients. However, in only three of our patients, bulbar involvement was the opening symptom and two still had no bulbar sign at diagnosis. Pyramidal involvement is frequent and may be the heralding symptom: two of our patients presented with asymmetric spastic paraparesis or tetraparesis as the isolated or main AOAD manifestation, and only MRI made the diagnosis possible. Therefore, the disease spectrum extends to include patients without bulbar involvement, at least in the first phases of the disease.
In some patients, brainstem involvement is a source of sleep disorders, including snoring, apnoeas and REM behaviour disorders. They were clinically relevant in four of our patients, and have been reported in the literature in at least eight patients (Table 4) (Stumpf et al., 2003; Li et al., 2005; Howard et al., 2008). Upper eyelid ptosis and diplopia may also occur. Cerebellar signs and symptoms are rather frequent, with axial and limb ataxia, nystagmus or saccadic pursuit.
Cervical spinal cord involvement may sometimes extend to the anterior horns and originate wasting and weakness of the upper limbs, of the scapular girdle, of the neck muscles (Okamoto et al., 2002), with EMG evidence of chronic denervation.
Unlike the early onset forms of AD, cognitive impairment is absent and epilepsy is exceptional, confirming that supratentorial involvement is limited or absent in AOAD. Definite intellectual deficit was not observed in our patients and has not been reported with certainty in the literature. Only one of our patients had seizures during infancy (possibly unrelated to AD) and three patients in the literature had seizures during the course of the disease (Stumpf et al., 2003; van der Knaap et al., 2006).
In our series, sphincter disturbances were rather common and in one instance were the first symptom of the disease. However, we observed clear-cut dysautonomia in only one patient, who had decreased sweating and episodes of hypertermia during summer. In the literature, orthostatic hypotension, hypothermia episodes, sweating disturbances, pupillary abnormalities, constipation, impotence, variably occurred in 13 patients and were considered expression of mild to severe dysautonomia, possibly related to hypothalamic involvement (Okamoto et al., 2002; Kinoshita et al., 2003; Probst et al., 2003; Stumpf et al., 2003; Li et al., 2005; Sreedharan et al., 2007; Balbi et al., 2008; Howard et al., 2008). Parkinsonism, rigidity or bradykinesia, suggestive of basal ganglia involvement, have been reported in four patients (Li et al., 2005; Salvi et al., 2007; Howard et al., 2008; Sechi et al., 2008—Case 4 of the present series). Hormone abnormalities have also been reported in only a few patients in the literature, and included vitamin D deficiency, primary ovarian failure, elevated thyroid-stimulating hormone (Sreedharan et al., 2007; Howard et al., 2008). Only one of our patients, however, had low testosterone levels and our data do not suggest that endocrinopathies are a feature of AOAD. Vertebral anomalies, particularly scoliosis and kyphosis, have been reported in AOAD in nine cases (Okamoto et al., 2002; Stumpf et al., 2003; Thyagarajan et al., 2004) and were present in two of our patients. Dysmorphic features, such as high arched palate, short neck, have been exceptionally described (Stumpf et al., 2003). The intraventricular cerebral tumour found in an AOAD patient may be a coincidental finding (Hirayama et al., 2008).
The disease course is variable and unpredictable; some patients showed intermittent symptoms, particularly in the early phases of the disease, whereas others had a rapidly progressive disease leading to loss of autonomous gait and death in a few years. In most patients, however, the disease runs a slowly progressive course allowing a relatively autonomous life for many years. In our series, disease duration ranged from 0.3 to 13 years and was up to 22 years in living patients reported in the literature (Li et al., 2005; van der Knaap et al., 2005). Patients may eventually become chairbound because of spastic tetraparesis and ataxia. Dysphagia may require positioning of nasogastric tube or percutaneous gastrostomy. Dyspnea is also reported and might be due to different causes, namely vocal cord palsy as in our Patient 7, bulbar dysfunction or muscle weakness. Aspiration pneumonia and respiratory insufficiency of whatever origin are the usual cause of death, which occurred in seven AOAD reported patients, aged 14–67 years, after 1.5–20 years of disease (Namekawa et al., 2002; Probst et al., 2003; Stumpf et al., 2003; Thyagarajan et al., 2004; Li et al., 2005; van der Knaap et al., 2006). Autopsy, performed in six cases, and biopsy of the medulla oblongata in one showed Rosenthal fibres in supratentorial periventricular and subpial regions, in medulla oblongata (with microcystic cavitations and fibre loss in the pyramids), cerebellum and spinal cord (Namekawa et al., 2002; Probst et al., 2003; Stumpf et al., 2003; Li et al., 2005; van der Knaap et al., 2006).
At the other end of the disease spectrum, two of our patients had no symptoms definitely attributable to AOAD, and were diagnosed by chance on MRI performed for possibly unrelated clinical problems; one further asymptomatic mutation carrier was uncovered during genetic family investigation. Four other asymptomatic patients with GFAP mutations have been reported in the literature (Okamoto et al., 2002; Stumpf et al., 2003; Shiihara et al., 2004; Balbi et al., 2008). It is impossible to predict when they will develop the disease, if ever. One patient had no definite sign at age 64 (Balbi et al., 2008). AOAD may have its onset very late, as late as age 58 years in a case reported by Howard et al. (2008) and 62 in a patient briefly mentioned by Brockmann and co-authors (2003); in the latter case, however, clinical information was scanty and significance of the GFAP mutation uncertain. In our series, we have the patients with the oldest age at onset, namely Patient 10, who carried the p.D128N mutation, aged 62 at onset and the father of Patient 11, who had the typical MRI picture, aged 71 at symptom appearance, thus further extending the age of disease onset in AD. Three of our patients had children. We offered neurological visit and genetic counselling for them. Genetic counselling is complicated in AOAD, as age of onset is highly variable and penetrance is incomplete. For at-risk subjects aged 18 and older asking for presymptomatic diagnosis, we would recommend following the international guidelines for late-onset autosomal dominant neurological disorders such as Huntington's chorea (Goizet et al., 2002).
Neuroradiologic aspects
Two interesting observations can be made on the MRI evaluation of our series of patients (Farina et al., 2008). With progression of age of onset, progressive restriction of signal abnormalities is observed on MRI. Supratentorial periventricular abnormalities tended not to appear in patients with age over 40 years at onset of the disease. A similar phenomenon was found regarding post-contrast enhancement. These two aspects suggest that the pathological process in AOAD becomes progressively restricted to the brainstem–spinal cord junction and that the influence of Rosenthal fibres accumulation in the endfeet of perivascular astrocytes on the blood–brain barrier is also decreased. The attempt to correlate the clinical symptoms and neurophysiological findings with MRI involvement of specific nuclei or fibre tracts was frustrating. The only association was between palatal myoclonus and signal abnormalities involving both the inferior olives and the hilum of the dentate nuclei, which were definite in three patients (Patients 4, 6 and 11) and borderline for the olives in Patient 3. However, two patients who had abnormalities in the olivary and dentate nuclei did not present myoclonus.
The clinical and neuroradiological differential diagnoses have been discussed in detail in previous papers (Romano et al., 2007; Farina et al., 2008). A diagnosis that should also be considered is adult-onset polyglucosan body disease, characterized clinically by cognitive impairment, pyramidal involvement, sensory-motor polyneuropathy, sphincter dysfunction, sometimes with cerebellar ataxia and extrapyramidal signs, and neuroradiologically by white matter changes and atrophy of medulla oblongata and spinal cord (Klein et al., 2004).
Molecular aspects
Ten of our patients had nine different missense mutations in highly conserved domains of the GFAP protein, with six novel and three previously reported changes, in addition to known common variants (Table 3). Thus far, there are 20 GFAP mutations associated with AOAD, including a complex allele with two mutations; all are missense mutations (Tables 3 and
4). GFAP is the intermediate filament peculiar to the astrocytes, where it contributes to cytoskeletal formation and maintenance, cell communication, and may affect functioning of the blood–brain barrier.
Mutant GFAP proteins, supposedly changing their own capabilities to form filaments, precipitate in aggregates together with heat-shock proteins and 
-B-crystallin, thus forming Rosenthal fibres and deranging astrocytes functions (Quinlan et al., 2007). This is in keeping with the hypothesis that mutant GFAP causes AD through a gain-of-function/dominant-negative mechanism and, consequently, with the autosomal dominant transmission of the disease. However, no genotype–phenotype correlation has been established, and some mutations have been associated with all three forms of AD (Quinlan et al., 2007; Yoshida et al., 2007). In particular, it is not clear why in some patients GFAP mutations lead to late onset and prevalent localization of pathological, neuroradiological and clinical changes in the lower brainstem, while in other patients AD onset occurs early and supratentorial abnormalities dominate the picture. The explanation for these different phenotypes may reside in the affected mutation sites and their effects on the GFAP protein, as confirmed for a few mutations observed in adults, which displayed an effect milder than severe mutations associated with infantile AD (Bachetti et al., 2008). The p.E205K mutation could be associated with late onset of disease, as it was carried by both the symptomatic Patient 8, with onset at age 50 years, and the 30-year-old asymptomatic Patient 5. In addition, other genetic or still unknown environmental factors have been hypothesized to influence the phenotype (Li et al., 2005). These modifiers might explain the variable disease expressivity and the reduced penetrance of some variants, which may also be detected in healthy parents. It is worth noting that there are only very few instances of coexistence of AOAD and earlier onset AD in the same family, therefore in a homogeneous genetic background (Stumpf et al., 2003; Hirayama et al., 2008).
We found no GFAP mutation in Patient 11, except for the p.D157N polymorphic variant. A possible pathogenic role of this variant cannot be excluded with certainty. Indeed, a mild functional effect could be consistent not only with its association with a late onset of the disease, but also with its low penetrance within families and the rare frequency in healthy individuals (Li et al., 2005, 2006). Nonetheless, we have included this patient in the current series because of positive family history, clinical presentation fully compatible with AOAD, including palatal myoclonus, and MRI features in both the patient and his affected father highly suggestive of AOAD. Few otherwise typical AOAD patients who carried no mutations in the GFAP gene are on the record (see http://www.waisman.wisc.edu/alexander/). It is therefore possible that AOAD is genetically heterogeneous and mutations of some other gene(s), perhaps coding for a protein strictly related with GFAP, may account for a small percentage of AOAD patients without GFAP mutations.
|
Concluding remarks |
|---|
|
TopSummaryIntroductionPatients and MethodsResultsDiscussionConcluding remarksReferences |
|---|
AOAD is more common than previously thought and might even be the most common form of AD. The diagnosis is strongly suggested by MRI and confirmed by GFAP gene analysis. The clinical picture is not specific, but AOAD must be considered in patients of any age with lower brainstem signs. When present, palatal myoclonus is strongly suggestive of AOAD. Pyramidal involvement, cerebellar ataxia, urinary disturbances and sleep disorders are common. Infrequent findings include scoliosis and dysautonomia. Fluctuations may occur. The course is variable, usually slowly progressive and less severe than the AD forms with earlier onset. Unfortunately, no therapy is available for any AD type. Hopefully, better understanding of pathogenic mechanisms will make it possible to define therapeutic strategies.
|
Acknowledgements |
|---|
We are gratefully indebted to Dr Angelo Sghirlanzoni and Dr Licia Grazzi (IRCCS Foundation, C. Besta Neurological Institute, Milan, Italy), and Dr Paola Gambaro (Clinica Neurologica L. Sacco Hospital, Milan, Italy) who followed Patients 4–6. We are also grateful to Dr Graziella Uziel (IRCCS Foundation, C. Besta Neurological Institute, Milan, Italy) for helpful discussion regarding pediatric AD patients.
|
References |
|---|
|
TopSummaryIntroductionPatients and MethodsResultsDiscussionConcluding remarksReferences |
|---|
Bachetti T, Caroli F, Bocca P, Prigione I, Balbi P, Biancheri R, et al. Mild functional effects of a novel GFAP mutant allele identified in a familial case of adult-onset Alexander disease. Eur J Hum Genet (2008) 16:462–70.[CrossRef][Web of Science][Medline]
Balbi P, Seri M, Ceccherini I, Uggetti C, Casale R, Fundarò C, et al. Adult-onset Alexander disease: report on a family. J Neurol (2008) 255:24–30.[CrossRef][Web of Science][Medline]
Brenner M, Johnson AB, Boespflug-Tanguy O, Rodriguez D, Goldman JE, Messing A. Mutations in GFAP, encoding glial fibrillary acidic protein, are associated with Alexander disease. Nat Genet (2001) 27:117–20.[CrossRef][Web of Science][Medline]
Brockmann K, Meins M, Taubert A, Trappe R, Grond M, Hanefeld F. A novel GFAP mutation and disseminated white matter lesions: adult Alexander disease? Eur Neurol (2003) 50:100–5.[CrossRef][Web of Science][Medline]
Caroli F, Biancheri R, Seri M, Rossi A, Pessagno A, Bugiani M, et al. GFAP mutations and polymorphisms in 13 unrelated Italian patients affected by Alexander disease. Clin Genet (2007) 72:427–33.[Web of Science][Medline]
Farina L, Pareyson D, Minati L, Ceccherini I, Chiapparini L, Romano S, et al. Can MR imaging diagnose adult-onset Alexander disease? AJNR Am J Neuroradiol (2008) 29:1190–6.
Goizet C, Lesca G, Dürr A. French Group for Presymptomatic Testing in Neurogenetic Disorders. Presymptomatic testing in Huntington's disease and autosomal dominant cerebellar ataxias. Neurology (2002) 59:1330–6.
Hirayama T, Fukae J, Noda K, Fujishima K, Yamamoto T, Mori K, et al. Adult-onset Alexander disease with palatal myoclonus and intraventricular tumour. Eur J Neurol (2008) 15:e16–7.[CrossRef][Web of Science][Medline]
Howard RS, Greenwood R, Gawler J, Scaravilli F, Marsden CD, Harding AE. A familial disorder associated with palatal myoclonus, other brainstem signs, tetraparesis, ataxia and Rosenthal fibre formation. J Neurol Neurosurg Psychiatry (1993) 56:977–81.
Howard KL, Hall DA, Moon M, Agarwal P, Newman E, Brenner M. Adult-onset Alexander disease with progressive ataxia and palatal tremor. Mov Disord (2008) 23:118–22.[CrossRef][Web of Science][Medline]
Kinoshita T, Imaizumi T, Miura Y, Fujimoto H, Ayabe M, Shoji H, et al. A case of adult-onset Alexander disease with Arg416Trp human glial fibrillary acidic protein gene mutation. Neurosci Lett (2003) 350:169–72.[CrossRef][Web of Science][Medline]
Klein CJ, Boes CJ, Chapin JE, Lynch CD, Campeau NG, Dyck PJ, et al. Adult polyglucosan body disease: case description of an expanding genetic and clinical syndrome. Muscle Nerve (2004) 29:323–8.[CrossRef][Web of Science][Medline]
Li R, Johnson AB, Salomons G, Goldman JE, Naidu S, Quinlan R, et al. Glial fibrillary acidic protein mutations in infantile, juvenile, and adult forms of Alexander disease. Ann Neurol (2005) 57:310–26.[CrossRef][Web of Science][Medline]
Li R, Johnson AB, Salomons GS, van der Knaap MS, Rodriguez D, Boespflug-Tanguy O, et al. Propensity for paternal inheritance of de novo mutations in Alexander disease. Hum Genet (2006) 119:137–44.[CrossRef][Web of Science][Medline]
Messing A, Head MW, Galles K, Galbreath EJ, Goldman JE, Brenner M. Fatal encephalopathy with astrocyte inclusions in GFAP transgenic mice. Am J Pathol (1998) 152:391–8.[Abstract]
Namekawa M, Takiyama Y, Aoki Y, Takayashiki N, Sakoe K, Shimazaki H, et al. Identification of GFAP gene mutation in hereditary adult-onset Alexander's disease. Ann Neurol (2002) 52:779–85.[CrossRef][Web of Science][Medline]
Ohnari K, Yamano M, Uozumi T, Hashimoto T, Tsuji S, Nakagawa M. An adult form of Alexander disease: a novel mutation in glial fibrillary acidic protein. J Neurol (2007) 254:1390–4.[CrossRef][Web of Science][Medline]
Okamoto Y, Mitsuyama H, Jonosono M, Hirata K, Arimura K, Osame M, et al. Autosomal dominant palatal myoclonus and spinal cord atrophy. J Neurol Sci (2002) 195:71–6.[CrossRef][Web of Science][Medline]
Probst EN, Hagel C, Weisz V, Nagel S, Wittkugel O, Zeumer H, et al. Atypical focal MRI lesions in a case of juvenile Alexander's disease. Ann Neurol (2003) 53:118–20.[CrossRef][Web of Science][Medline]
Quinlan RA, Brenner M, Goldman JE, Messing A. GFAP and its role in Alexander disease. Exp Cell Res (2007) 313:2077–87.[CrossRef][Web of Science][Medline]
Romano S, Salvetti M, Ceccherini I, De Simone T, Savoiardo M. Brainstem signs with progressing atrophy of the medulla oblongata and upper cervical spinal cord. Lancet Neurol (2007) 6:562–70.[CrossRef][Web of Science][Medline]
Salmaggi A, Botturi A, Lamperti E, Grisoli M, Fischetto R, Ceccherini I, et al. A novel mutation in the GFAP gene in a familial adult onset Alexander disease. J Neurol (2007) 254:1278–80.[CrossRef][Web of Science][Medline]
Salvi F, Aoki Y, Della Nave R, Vella A, Pastorelli F, Scaglione C, et al. Adult Alexander's disease without leukoencephalopathy. Ann Neurol (2007) 58:813–4.[CrossRef]
Schwankhaus JD, Parisi JE, Gulledge WR, Chin L, Currier RD. Hereditary adult-onset Alexander's disease with palatal myoclonus, spastic paraparesis, and cererbellar ataxia. Neurology (1995) 45:2266–71.
Sechi GP, Conti M, Sau GF, Cocco GA. Valproate-induced parkinsonism, glial cells and Alexander's disease. Prog Neuropsychopharmacol Biol Psychiatry (2008) 32:1351–2.[CrossRef][Medline]
Shiihara T, Sawaishi Y, Adachi M, Kato M, Hayasaka K. Asymptomatic hereditary Alexander's disease caused by a novel mutation in GFAP. J Neurol Sci (2004) 225:125–7.[CrossRef][Web of Science][Medline]
Sreedharan J, Shaw CE, Jarosz J, Samuel M. Alexander disease with hypothermia, microcoria, and psychiatric and endocrine disturbances. Neurology (2007) 68:1322–3.
Stumpf E, Masson H, Duquette A, Berthelet F, McNabb J, Lortie A, et al. Adult Alexander disease with autosomal dominant transmission: a distinct entity caused by mutation in the glial fibrillary acid protein gene. Arch Neurol (2003) 60:1307–12.
Thyagarajan D, Chataway T, Li R, Gai WP, Brenner M. Dominantly-inherited adult-onset leukodystrophy with palatal tremor caused by a mutation in the glial fibrillary acidic protein gene. Mov Disord (2004) 19:1244–8.[CrossRef][Web of Science][Medline]
van der Knaap MS, Naidu S, Breiter SN, Blaser S, Stroink H, Springer S, et al. Alexander disease: diagnosis with MR imaging. AJNR Am J Neuroradiol (2001) 22:541–52.
van der Knaap MS, Ramesh V, Schiffmann R, Blaser S, Kyllerman M, Gholkar A, et al. Alexander disease: ventricular garlands and abnormalities of the medulla and spinal cord. Neurology (2006) 66:494–8.
van der Knaap MS, Salomons GS, Li R, Franzoni E, Gutiérrez-Solana LG, Smit LM, et al. Unusual variants of Alexander's disease. Ann Neurol (2005) 57:327–38.[CrossRef][Web of Science][Medline]
Yoshida T, Tomozawa Y, Arisato T, Okamoto Y, Hirano H, Nakagawa M. The functional alteration of mutant GFAP depends on the location of the domain: morphological and functional studies using astrocytoma-derived cells. J Hum Genet (2007) 52:362–9.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||