Adult-Onset Leukoencephalopathy with Axonal Spheroids and Pigmented Glia Can Present as Frontotemporal Dementia SyndromeWong J.C.a · Chow T.W.b–d · Hazrati L.-N.a
aTanz Centre for Research in Neurodegenerative Diseases, bRotman Research Institute and Division of Neurology at Baycrest, and cDivision of Neurology, Department of Medicine, and dDivision of Geriatric Psychiatry, Department of Psychiatry, University of Toronto, Toronto, Ont., Canada Corresponding Author
Background/Aims: We review the characteristics of adult-onset leukoencephalopathy with axonal spheroids and pigmented glia(ALSP) and determine prevalence of behavioral variant of frontotemporal dementia (bvFTD) features in ALSP. Methods: Clinical and pathological information was abstracted from histopathologically confirmed ALSP cases identified by a systematic literature search. A new case of ALSP presenting as bvFTD was also described. Results: We retrieved 51 ALSP cases. Mean age of onset was 42.2 years. Mean disease duration was 6.2 years, with 24 cases lasting 4 years or fewer. Fourteen cases had 3 or more of the 6 key bvFTD features. White matter hyperintensities on T2-weighted MRI, motor symptoms, seizures and amnesia were common. Conclusion: ALSP can underlie FTD syndrome, as well as rapidly progressive dementia.
Copyright © 2011 S. Karger AG, Basel
Adult-onset leukoencephalopathy with axonal spheroids and pigmented glia (ALSP) is a rare inherited or sporadic dementia affecting the white matter of the central nervous system, although it may be less rare than previously expected. ALSP is now thought to encompass previously distinct entities of hereditary diffuse leukoencephalopathy with axonal spheroids (HDLS) and familial pigmentary orthochromatic leukodystrophy (POLD) [1,2]. ‘Adult-onset leukoencephalopathy with neuroaxonal spheroids’ and minor variations of this name have also been used interchangeably with ALSP or HDLS. A diagnosis that requires histopathological confirmation, ALSP has distinctive pathological features of both HDLS and POLD . Accordingly, we refer to ALSP as the disease entity encompassing HDLS, POLD and adult-onset leukoencephalopathy with neuroaxonal spheroids. ALSP can be distinguished from other adult-onset leukodystrophies and leukoencephalopathies, which have been discussed elsewhere [3,4].
An early case report of POLD was published in 1936 , and a HDLS familial kindred was described in 1984 . Since then, several authors have qualitatively reviewed pathological and clinical characteristics of ALSP , HDLS  and POLD . We encountered a case of rapidly progressive frontotemporal dementia (FTD) that had ALSP at autopsy, raising the question of how often the neuropsychiatric manifestation of ALSP meets criteria for behavioral variant of FTD (bvFTD). FTD is the third most common cause of cortical dementia after Alzheimer’s disease and dementia with Lewy bodies. FTD most commonly presents with progressive behavioral changes, which is known as bvFTD; the other variant presents with progressive language dysfunction with preservation of other cognitive domains . Diagnosis of bvFTD has been based on the consensus report by Neary et al.  in 1998, but recently in 2011, newer bvFTD criteria were developed with improved diagnostic sensitivity . We thus performed a systematic literature search for all histopathologically confirmed cases of ALSP from 1970 to 2011 to characterize clinical and pathological features of ALSP and determine the prevalence of bvFTD features in ALSP based on 1998 and 2011 criteria. We also describe our new illustrative case study of a patient with underlying ALSP which presented as bvFTD.
A National Institute of Health PubMed database search for case reports and case series in English from 1970 to 2011 was performed, using keywords ‘adult-onset leukoencephalopathy with axonal spheroids and pigmented glia’; ‘hereditary diffuse leukoencephalopathy with axonal spheroids’; ‘pigmentary orthochromatic leukodystrophy’ or ‘orthochromatic pigmented leukodystrophy’; ‘leukoencephalopathy OR leucoencephalopathy AND neuroaxonal spheroids’ or ‘leukodystrophy OR leucodystrophy AND neuroaxonal spheroids.’ ‘Neuroaxonal’ and ‘axonal’ were also interchangeably searched. Medical Subject Headings proved too general for the objectives of this study. Reference lists from identified articles were then searched manually for additional case reports. Duplicated publications in the search were excluded. Publications were included if they were case reports or series addressing ALSP in human subjects aged ≧15 years old. All of these criteria were necessary for inclusion. After application of these inclusion criteria for publications, the following inclusion criteria were applied to select individual cases from these publications: (1) histopathologically confirmed (biopsy or autopsy) ALSP, and (2) availability of a descriptive clinical history.
The following data were abstracted from each case: (1) patient information: sex, age of disease onset, age of death, clinical diagnosis, pathological diagnosis, earliest symptom and family history of neurodegenerative disease; (2) pathological features of ALSP: features characteristic of HDLS and POLD, white matter changes and concomitant pathology; (3) presence of bvFTD criteria based on 1998  and 2011  criteria; (4) clinical signs or symptoms not included in bvFTD criteria: whether cognitive, motor, behavioral or psychiatric; and (5) imaging results: atrophy, T2-hyperintensities in MRI and abnormalities in single photon emission computed tomography (SPECT) and positron emission tomography (PET).
To fully illustrate one case for the reader, a retrospective chart review of clinical and pathological records of a patient from a tertiary academic center was performed. Informed consent was provided by the patient’s family to publish his case.
Results of the systematic literature search are summarized in figure 1. The literature search identified 51 individual cases that fulfilled the inclusion criteria, with absence of exclusion criteria, from 31 publications [2,3,4,6,7,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36]. Including the new case described in this study, a total of 52 individual ALSP cases were subsequently analyzed. Widening the range of keywords for the search made us aware that although ALSP is rare, there was a substantial number of ALSP cases already published.
|Fig. 1. Flow chart of systematic literature search for histopathologically-confirmed cases of ALSP from 1970 to 2011.|
Online supplementary Appendix A (see online at www. karger.com/doi/10.1159/000331422) lists specific information extracted from 31 publications and 1 case of ALSP. Overall, there were 30 women and 21 men (1 unrecorded). Pathological diagnoses were recorded as ALSP in 8 patients, HDLS in 12 patients, POLD in 17 patients, and adult-onset leukoencephalopathy with neuroaxonal spheroids in 15 patients. Eight diagnoses were confirmed by biopsy antemortem, while the rest were confirmed at autopsy. Mean age of onset was 42.2 ± 12.1 years, ranging from 15 to 78 years (n = 52). Initial symptoms of disease were highly varied and are listed in detail in online supplementary Appendix A. Patients could present with more than one initial symptom, which included personality or behavioral changes (n = 15), motor abnormalities (n = 15), depression or anxiety (n = 9), memory impairment (n = 9), speech abnormalities (n = 9) and seizure (n = 3).
Mean duration of progression was 6.2 ± 6.0 years, ranging from 2 months to 34 years (n = 45). Excluding the two extreme cases of 2 months and 34 years of progression, which were both reported by Axelsson et al. , mean duration was 5.7 ± 4.2 years (n = 43). Of the 45 patients who were followed until death, 24 had a disease duration of 4 years or less. Mean age of death was 48.6 ± 14.3 years, ranging from 17 to 89 years (n = 45). Of the 44 cases that did not have brain biopsies performed, 14 listed antemortem diagnoses, none of which accurately diagnosed ALSP: Pick’s disease, bvFTD or FTD (n = 4), Alzheimer’s disease (n = 2), corticobasal degeneration (n = 2), presenile dementia (n = 2), psychiatric disorder (n = 2), microvascular leukoencephalopathy (n = 1) and Binswanger leukoencephalopathy (n = 1).
Twenty-five patients had at least one family member with histopathologically confirmed ALSP, belonging to 10 unique kindreds; 24 had family histories of neurodegenerative diseases, which are listed in online supplementary Appendix A.
Table 1 shows the prevalence of bvFTD features in ALSP patients according to the 2011 and 1998 criteria [9,10]. All cases had progressive deterioration, fulfilling the first criteria for bvFTD according to 2011 criteria ; 14 cases had 3 or more of 6 clinically discriminating features, fulfilling the criteria for possible bvFTD [3,6,7,10,11,12,18,20,23,27,28,33,36]; 9 cases had 3 features [6,7,18,20,23,27,33,36], 4 cases had 4 features [3,11,12,28] and our case had 5 features. Of these 14 cases, 6 went on to fulfill criteria for probable bvFTD with imaging consistent with bvFTD (frontal and/or temporal atrophy on CT or MRI, or frontal hypoperfusion or hypometabolism on SPECT or PET) and significant functional decline [7,10,12,23,28,33]. These cases also fulfilled the requirement for possible or probable bvFTD diagnoses that other non-degenerative nervous system, medical or psychiatric disorders could not better account for the disease, and the requirement for probable bvFTD diagnosis that biomarkers indicative of another neurodegenerative process were absent . None of the ALSP cases had frontotemporal lobar degeneration pathology or a known pathogenic mutation, thus none had a diagnosis of definite bvFTD .
|Table 1. Number of 52 ALSP cases meeting clinical criteria for bvFTD, using 2011 and 1998 consensus clinical criteria|
From a total of five 1998 core features , the mean number of core features exhibited was 2.0 ± 1.6 (n = 52); 3 cases had all 5 core features [11,12,23] and 10 cases had 4 core features [3,4,6,7,27,30,33,36]. The most common core feature was insidious onset and gradual progression. Mean number of supportive features exhibited was 3.7 ± 2.5, ranging from 0 to 11 out of a total of 18 supportive features (n = 52) . Common supportive diagnostic features included altered speech output, primitive reflexes, incontinence, frontal or anterior temporal abnormality on imaging, and mutism. Notably, 50 patients had onset of disease before age 65 years.
Table 2 demonstrates the prevalence of clinical signs and symptoms in ALSP patients that have not been covered by the bvFTD criteria. Motor findings of pyramidal signs, postural or gait disturbance, seizures, and dysarthria or mutism were most frequently reported. Mean age of seizure onset was 45.3 ± 14.8 years, ranging from 21 to 83 years (n = 22). There were no signs of epileptiform activity on electroencephalogram (EEG) investigations, but slow or δ-wave activity seen across all dementias was observed.
|Table 2. Non-bvFTD clinical characteristics of 52 ALSP cases|
Table 3 demonstrates the prevalence of specific gross and microscopic pathological features. Almost all cases featured myelin loss. The majority of cases also had gliosis and axonal spheroids. Macrophages with pigmented granules were frequently found, as were iron-positive macrophages. Some cases had macrophages with lipid granules or intracytoplasmic ‘fingerprint’ patterns. White matter was most often affected in the frontal lobe and corpus callosum. Sparing of subcortical U-fibers and ventricular dilation were common findings. Only two cases had concomitant pathology of senile plaques and amyloid angiopathy with Braak neurofibrillary tangle stage III-IV in one case and stage I in the other case .
|Table 3. Pathological features of the 52 ALSP cases|
Imaging findings are summarized in table 4. CT or MRI often showed brain atrophy regionally or diffusely. White matter hyperintensities on T2-weighted MRI were most frequently observed diffusely. Of note, despite histopathological descriptions of subcortical U-fiber sparing, 2 cases had reported white matter hyperintensities in this tract [14,30]. On PET, generalized reduced metabolism was exhibited predominantly in the frontal lobes of one case  and parietal lobes of another .
|Table 4. Imaging features of the 52 ALSP cases|
A 60-year-old right-handed Caucasian man with 17 years of formal education presented to a specialty memory clinic after 2 years of cognitive decline that had accelerated in the past year. The history of emotional blunting, poor interpersonal conduct, amotivational syndrome, new preference for sweets, and distractibili- ty – against relative preservation of spatial orientation and short-term memory – raised the highest suspicion of bvFTD. At presentation, he required prompting for activities of daily living, yet maintained fluent speech. His family denied dysarthria, dysphagia, new weakness or falls. He had no stroke risk factors, no significant neurological or psychiatric past medical history, and no relevant family history. Mental status testing showed abnormal executive functions of sustained attention, set-shifting and self-monitoring. Elemental neurological examination was normal.
Brain MRI on previous work-up for presumed depression showed symmetrical cerebral atrophy with frontal predominance (fig. 2a). There were hyperintensities in deep white matter of bilateral frontal lobes, left temporal lobe, subcortical regions and splenium of the corpus callosum. On a SPECT scan the same year, perfusion in the right anterior frontal lobe was decreased. Brain MRI repeated 6 months later showed increased volume of the previously identified hyperintensities (fig. 2b). Frontal cerebral atrophy had also worsened and progressed dorsally. Hippocampal atrophy was not apparent.
|Fig. 2. Typical imaging, gross pathology and histopathology samples of ALSP in one patient. a Axial T2-weighted FLAIR MRI of the patient’s brain reveals frontal lobe hyperintensities bilaterally in deep white matter. b Axial T2-weighted FLAIR MRI 6 months later reveals increased hyperintensities in the same region, and more pronounced and dorsal frontal atrophy. c Coronal section shows intact subcortical U-fibers at the arrow, along with extensive demyelination in the frontal lobe. d Coronal section shows white matter loss limited to the temporal lobes, again with preservation of subcortical U-fibers. e HE and Luxol Fast Blue stains show macrophages with pigment (arrows) and large reactive astrocytes. f HE and Luxol Fast Blue stains show an axonal spheroid (arrow) and large reactive astrocytes. g Macrophages with Periodic acid-Schiff stain-positive pigment (arrows). h Transmission electron microscopy shows an intracytoplasmic ‘fingerprint’ pattern, which is characteristic of pigmentary orthochromatic leukodystrophy macrophages.|
Two and a half years into the illness, he lost spontaneous speech. There was no parkinsonism, loss of vertical saccades or gait abnormality. Brain MRI at that time showed further progression of all white matter changes, including involvement of the entire inferior corpus callosum. The patient developed incontinence and falls, then dysphagia in the third year of illness.
Within the next year, he manifested a generalized tonic-clonic seizure. Cerebrospinal fluid analysis did not support viral encephalitis, multiple sclerosis, Lyme’s disease or active inflammatory processes. This seizure heralded a rapid decline into nursing home admission for palliative care. Motor weakness rapidly worsened to the point of immobility and full dependence for nutrition. The patient died before completing a fourth year of illness. The family deferred an electromyogram study, but requested an autopsy.
The postmortem brain weighed 1,600 g. Cortical atrophy was pronounced at bilateral frontal tips, but not in other regions. Frontal coronal sections revealed bilateral atrophy of deep white matter, corpus callosum and internal capsule (fig. 2c–d). Subcortical U-fibers were intact. White matter atrophy was observed in temporal and subcortical regions and cerebellar peduncles. Occipital white matter was relatively spared. Ventricles appeared enlarged with a preserved but flattened head of the caudate nucleus.
Microscopic examination confirmed myelinated white matter loss with sparing of subcortical U-fibers. There were gliosis and reactive astrocytosis (fig. 2e), axonal spheroids (fig. 2f) and macrophages with pigmentation (fig. 2g) in white matter regions. Microcalcifications and vacuolations were observed in some sections. Electron microscopy revealed intracytoplasmic ‘fingerprint’ patterns in white matter regions (fig. 2h).
Analysis of clinical features from 52 histopathologically confirmed cases of ALSP revealed a high frequency of clinical bvFTD features. Four patients were diagnosed antemortem with a FTD syndrome. Using 2011 bvFTD criteria, 26.9% of ALSP cases could have been diagnosed with possible bvFTD and 11.5% could have been diagnosed with probable bvFTD . Furthermore, undocumented bvFTD features may have existed in some cases included in our review. For example, psychiatric symptoms of emotional lability, depression, irritability or aggression could have contributed to features typical of bvFTD – such as decline in interpersonal or personal conduct and loss of sympathy or empathy – that were not explicitly documented.
There were key distinguishing features of ALSP that are not commonly seen in bvFTD. First, the prominent white matter hyperintensities on T2-weighted MRI of ALSP cases were not consistent with ‘routine’ bvFTD imaging findings. In bvFTD, distinct atrophy is expected in frontal and temporal lobes with limited white matter damage [3,37]. Secondly, many ALSP patients had seizures. Interestingly, despite the prevalence of seizures, no patients showed epileptiform activity on EEG investigations. Abnormal EEG results are also unsupportive of bvFTD, and most ALSP patients who underwent EEG investigations had abnormal slow wave activity. Thirdly, the mean age of onset in ALSP patients (42.2 years, ranging from 15 to 78 years) was younger than that in FTD patients (60.4 years, ranging from 30 to 82 years) . Finally, ALSP progressed more rapidly (mean 74.4 months) than FTD (mean 89.1 months) . Common pathologies underlying rapidly progressive dementia include Creutzfeldt-Jakob disease, frontotemporal lobar degeneration, tauopathies, diffuse Lewy body disease and Alzheimer’s disease . While ALSP is a comparably rare pathology, it nevertheless should be considered in the differential diagnosis of rapidly progressive dementia.
ALSP is pathologically distinguished by myelin loss, gliosis, neuroaxonal spheroids, macrophages with pigmented granules, sparing of subcortical U-fibers and varied but frontal-predominant white matter damage. Some cases had macrophages with intracytoplasmic ‘fingerprint’ patterns. These patterns have been observed in previous case reports and are suggestive of ceroid [12,15,18]. Since oxidative stress or decreased lysosomal proteolytic activity can lead to ceroid or lipofuscin accumulation in human glial cells , these patterns raise the possibility that oxidative damage plays a role in the pathogenesis of ALSP. Macrophages with granular inclusions may also be associated with solvent inhalation, but none of the ALSP cases had a recorded history of solvent exposure .
The past decade featured an increasing number of ALSP cases diagnosed antemortem by brain biopsy, whereas previous diagnoses had largely relied on autopsies. Historically, other diagnostic techniques in ALSP also evolved. Since the 1970s, CT imaging has revealed regional or generalized atrophy in brains of ALSP patients. EEG tests have also been used for investigations, but have not been definitively diagnostic. MRI became commonplace since the late 1990s; T2-weighted MRI has been especially important in revealing white matter changes. Functional imaging such as SPECT or PET may provide more opportunities for antemortem diagnosis of ALSP in the future.
Systematic literature reviews are more commonly performed for randomized control or cohort studies, but we pursued a systematic literature review for case reports, which had several limitations. First, case selection in this review was based on fulfillment of inclusion criteria, and not on any additional parameters for study quality. Some case reports of ALSP were excluded due to lack of clinical information about an individual case [42,43]. Secondly, while we used a quantitative approach in order to provide an accurate description of ALSP, definitive conclusions could not be drawn about disease etiologies or correlations, due to the heterogeneity of authors’ qualitative descriptions in case reports. Thirdly, although we selected cases with adequate clinical descriptions, the majority of cases did not explicitly record signs or symptoms of bvFTD. Thus, omission of recorded features of bvFTD did not necessarily mean that they were not present. Furthermore, our assessment of fulfillment of bvFTD criteria was based solely on published clinical histories; optimally, full neuropsychological testing would have supported the bvFTD diagnosis. Finally, we limited this systematic literature search to English articles, but there are also French [5,44,45,46,47,48], German [49,50] and Japanese  reports that describe ALSP.
Based on the results of this systematic literature review on ALSP presenting as bvFTD, we recommend that ALSP should be considered in the differential diagnosis when white matter hyperintensities on T2-weighted MRI are noted in an FTD workup, or dementia is rapidly progressive. An increased autopsy rate for clinically diagnosed bvFTD can help elucidate its pathological basis. An increased biopsy rate can help rule out treatable conditions (e.g. vasculitis); avoid medications with potentially harmful side effects for diseases with no known treatment, such as ALSP; and provide definitive antemortem diagnoses for patients and their families.
This work was funded by the University of Toronto Dean’s Fund for New Faculty (No. 457494), an endowment to the Sam and Ida Ross Memory Clinic; Women of Baycrest (T.W.C.); Alzheimer Society of Ontario (L.-N.H.); and an American Academy of Neurology Medical Student Summer Research Scholarship (J.C.W.).
Lili-Naz Hazrati, MD, PhD, FRCPC
Tanz Center for Research in Neurodegenerative Diseases
Tanz Neuroscience Building, University of Toronto
6 Queen’s Park Crescent West, Toronto, ON M5S 3H2 (Canada)
Tel. +1 416 978 1877, E-Mail email@example.com
Accepted: July 29, 2011
Published online: October 5, 2011
Number of Print Pages : 9
Number of Figures : 2, Number of Tables : 4, Number of References : 51
Additional supplementary material is available online - Number of Parts : 1
Dementia and Geriatric Cognitive Disorders
Vol. 32, No. 2, Year 2011 (Cover Date: October 2011)
Journal Editor: Chan-Palay V. (Boston, Mass.)
ISSN: 1420-8008 (Print), eISSN: 1421-9824 (Online)
For additional information: http://www.karger.com/DEM