Cognitive Evaluation of Neurologically Asymptomatic Boys With X-linked Adrenoleukodystrophy

Background  Various studies have demonstrated abnormal neuropsychological function in boys with the childhood cerebral phenotype of X-linked adrenoleukodystrophy. Not much is known about the cognitive function of neurologically asymptomatic boys with X-linked adrenoleukodystrophy who have normal brain magnetic resonance imaging results.

Objective  To describe the cognitive profile of 52 neurologically asymptomatic boys with X-linked adrenoleukodystrophy (mean ± SD age, 6.7 ± 3.6 years).

Methods  Neuropsychological tests included evaluation of IQ (full-scale IQ, verbal IQ, and performance IQ), 5 major cognitive domains (language, visuospatial skills, perception, visuomotor or graphomotor skills, memory, and attention or executive function), adaptive skills, and academic achievement. Standardized z scores relative to age-appropriate published norms were generated. Association between age and cognitive performance was evaluated using nonparametric Spearman rank correlation and robust median regression adjusting for full-scale IQ and socioeconomic status.

Results  All but 4 patients had normal cognitive function. There was a negative correlation between age and visual perception as well as age and visuomotor skills after adjustment for full-scale IQ and socioeconomic status.

Conclusions  This study provides, to our knowledge, the first evidence of overall normal cognitive function in neurologically and radiologically normal boys with X-linked adrenoleukodystrophy, indicating no evidence of neurodevelopmental abnormalities despite the inherent ABCD1 mutation. Subtle deterioration with age was observed in some functional domains. This suggests that prevention and timely institution of therapy can potentially preserve cognitive function seen in patients with the cerebral X-linked adrenoleukodystrophy phenotype. X-linked adrenoleukodystrophy should be considered a candidate disorder for neonatal screening.

X-linked adrenoleukodystrophy (X-ALD) is a neurodegenerative disorder affecting the nervous system, adrenal cortex, and testes due to a defect in ABCD1, which codes for a peroxisomal membrane protein, a member of the ATP-binding cassette transporter superfamily.^1-3 The accumulation of saturated very long-chain fatty acids is the principal biochemical abnormality in X-ALD, and demonstration of abnormally high very long-chain fatty acid levels in plasma is the most frequently used diagnostic assay.⁴

X-linked adrenoleukodystrophy is expressed in a variety of clinical phenotypes with and without cerebral demyelination. The childhood cerebral phenotype of X-ALD usually manifests in the first 10 years of life.¹ Psychological and behavioral abnormalities have considerable prognostic significance, and in many cases, they initially resemble attention-deficit/hyperactivity disorder. Abnormalities in performance IQ (PIQ) and executive function have been demonstrated in patients with childhood cerebral adrenoleukodystrophy, most of whom had extensive posterior white matter changes.⁵ The corpus callosum, parieto-occipital lobes, and temporal lobes⁶ are most commonly involved in X-ALD. The abnormalities of these regions lead to abnormal processing of visual and auditory input. It is not known whether there are inherent cognitive abnormalities in patients with X-ALD owing to the nature of the biochemical and genetic defect when inflammatory demyelination is not demonstrable. To our knowledge, cognitive function in patients with X-ALD without any obvious white matter changes has not been evaluated previously. One study⁵ that included 8 neurologically asymptomatic boys suggested normal cognitive profiles with subtle abnormalities in verbal fluency; however, only 1 of those 8 boys had normal brain magnetic resonance imaging (MRI) results. In this study, we evaluated the neuropsychological profile of 52 neurologically asymptomatic boys with X-ALD who had normal conventional brain MRI results.

This study included 52 neurologically asymptomatic boys with normal brain MRI results and confirmed diagnosis of X-ALD. Their mean ± SD age was 6.7 ± 3.6 years, with a median of 6.2 years and a range from 2.1 to 14.6 years. The MRI results were evaluated using the previously published Loes scoring methodology.⁶ An MRI with a Loes score of 0.5 or less was classified as normal. One patient who was receiving medications for attention-deficit/hyperactivity disorder at the time of evaluation was excluded from the analyses.

Neuropsychological function was scored with methods described in a previous article from our group⁷ based on the general procedures developed by Shapiro and Klein.⁸ Testing of IQ was done using the age-appropriate Stanford-Binet Intelligence Scale–Fourth Edition (in 14 patients) and Wechsler Intelligence Scales (26 patients received the Wechsler Intelligence Scale for Children–Third Edition, 11 received the Wechsler Preschool and Primary Scale of Intelligence–Revised, and 1 received the Wechsler Intelligence Scale for Children–Revised).

The instruments chosen to assess neuropsychological function across the broad age range were mostly based on a battery selected by Shapiro and Klein,⁸ whose selection criteria focused not only on evaluating all of the major domains of neuropsychological function but also tried to combine the availability of good normative data over the widest applicable age range. Table 1 lists the testing instruments and respective ages at which each was used. The obtained scores were transformed into age-adjusted z scores using the published norms.

The effect of socioeconomic status (SES) on intellectual function, especially verbal and language skills, is well established. We used the Index of Social Position by Hollingshead¹⁷ to assess the SES of the families of patients included in the study to analyze its effects on cognitive performance of our study population.

Test results were standardized to z scores using age-appropriate published norms. A z score of −2 or less represented 2 SD or more below the normal mean and was considered significantly abnormal. Robust nonparametric median regression adjusting for full-scale IQ (FSIQ) and SES was used to evaluate the association between age and performance on neuropsychological testing (the absolute Index of Social Position score was used for adjustment instead of the SES category). A P value of .05 or less was considered statistically significant.

Table 2 summarizes the evaluation for the entire cohort. The IQ subsets (FSIQ, verbal IQ [VIQ], and PIQ), global z score, and performance on the 5 cognitive function domains were within normal limits for 49 patients and were abnormal in 3. Figures 1, 2, 3, and 4 demonstrate performance on IQ subsets and major cognitive function domains.

Four patients showed significant abnormality in 1 or more domains. Patient 1 was aged 5.2 years, with significant abnormalities in FSIQ and VIQ; his age-appropriate z score for FSIQ was –2.13, and his VIQ z score was –2.73. He also showed significant abnormalities in the language domain (z score = –2.59) and Vineland Adaptive Behavior Scales (adaptive behavior composite z score = –2.67; communication domain z score = –3.00; daily living skills subscale z score = –2.73; and socialization domain z score = –2.47). His PIQ and performance on other functional domains were within normal limits. Patient 2 was aged 12.8 years. He showed significant abnormality in the executive function domain (z score = –2.04), his PIQ placed in the low average range, and his VIQ and FSIQ were placed within normal limits. His performance in all of the other functional domains was within normal limits. Patient 3 was aged 2.6 years; he showed significantly poor performance in the visual perception domain (z score = –3.40). He showed no abnormalities on VIQ, which was the only other evaluation the patient could complete (VIQ z score = 0.50). He did not complete testing for PIQ (therefore, the FSIQ could not be completed), visual perception, memory, and executive domains. Patient 4 was aged 2.4 years and had significant abnormality in the Vineland Adaptive Behavior Scales domain-testing activities of daily living (z score = –2.00). This patient had no abnormalities in any other tested cognitive domain.

There was no association between age and FSIQ, PIQ, or VIQ after adjusting for SES. There was a significant negative association between age and visual perception as well as age and visuomotor skills after adjustment for FSIQ and SES (for visual perception, β = −0.15, P = .02; for visuomotor skills, β = −0.12, P = .005). There was no association between attention or executive function, memory, and language z scores with age after adjustment for FSIQ and SES.

This study demonstrates an overall normal cognitive profile in asymptomatic boys with X-ALD who have no detectable abnormalities on conventional brain MRI. Forty-eight patients demonstrated normal-for-age performance on neuropsychological testing. Four patients had moderate deficits in verbal skills, the visuomotor domain, and adaptive behavior. Such variations may also occur in a normal population. It is not certain whether they can be attributed to X-ALD. Thus, for the tested cohort as a whole, we did not detect cognitive deficiencies using standard testing procedures, suggesting a normal neurodevelopmental process despite the inherent defect of ALDP (a protein encoded by the mutant gene in X-ALD).

There was an inverse correlation between age and performance in visuomotor and visual perception domains. It may be that some decline in cognitive performance might surface later with increasing age; however, definitive implications of this apparent association cannot be made in a cross-sectional study. Despite the weak negative age association, the standardized z scores remained within normal limits. Alternatively, the negative association could be owing to the better ability to characterize cognitive performance for older children because of the greater sensitivity of the testing instrument; thus, the weak negative association could be an artifact of the wider standard deviations observed in the younger subset mainly owing to inherent inadequacies of the evaluation tools.

There are numerous studies that have demonstrated neuropsychological abnormalities in patients with the cerebral phenotype of X-ALD; however, it is not clear whether these patients would have normal cognitive development in the absence of these cerebral abnormalities. All of the patients in our cohort had normal conventional MRI results. However, this does not rule out the presence of more subtle abnormalities. Studies that use magnetic resonance spectroscopy and diffusion-tensor imaging may show abnormalities in the regions that appear normal on conventional MRI. Asymptomatic boys with X-ALD and other phenotypes of X-ALD with normal brain MRI results and no evidence of obvious inflammatory demyelination have been shown to have reduced N-acetylaspartate levels, suggesting subtle neuroaxonal abnormalities that are not detectable on conventional MRI.^18-20 The observed subtle decrease (or failure to grow) in performance within some cognitive functional domains may be owing to axonal structural abnormalities not visible on conventional MRI that might be demonstrable with the use of specialized imaging modalities such magnetic resonance spectroscopy and diffusion-tensor imaging, which were not included in this analysis. Studies combining cognitive evaluation with specialized magnetic resonance methodologies such as spectroscopy may serve as sensitive markers and aid in predicting disease progression.²¹ However, even if the abnormalities were present in our cohort, they do not appear to have an effect on standard tests of cognitive function.

The presence of normal cognitive function in most boys with X-ALD whose conventional MRI results are normal, combined with the encouraging results of early therapeutic interventions, suggests that X-ALD should be viewed as a candidate disorder for neonatal screening. Recent studies from our group provide evidence that dietary therapy with Lorenzo's oil has a preventive effect in asymptomatic patients with normal MRI results²² and that patients with X-ALD with impaired adrenal reserve can be identified and treated before they develop overt Addison disease.²³ The follow-up study by Peters et al²⁴ shows a favorable outcome (92% 5-year survival rate) of hematopoietic cell transplantation in patients with X-ALD who receive the transplant when the inflammatory brain disease is still in its early stages. The 3-pronged management approach discussed earlier can help improve the prognosis of X-ALD. Furthermore, the risk of developing cerebral disease is not constant with increasing age. Prior to age 7 years, the risk of cerebral disease is approximately 40%. It diminishes progressively after that age, and it is less than 10% after age 15 years.²⁵ Thus, early identification of at-risk subjects is of great importance. Our demonstration that cognitive function is intact in young asymptomatic patients indicates that there is a substantial therapeutic “window of opportunity” and strengthens the rationale for neonatal screening.

Correspondence: Hugo W. Moser, MD, Kennedy Krieger Institute, 707 N Broadway, Baltimore, MD 21205 (moser@kennedykrieger.org).

Accepted for Publication: September 1, 2005.

Author Contributions:Study concept and design: Raymond and H. W. Moser. Acquisition of data: Cox, Raymond, Mahmood, and A. B. Moser. Analysis and interpretation of data: Cox, Dubey, Mahmood, and H.W. Moser. Drafting of the manuscript: Raymond, Mahmood, and H.W. Moser. Critical revision of the manuscript for important intellectual content: Cox, Dubey, Raymond, Mahmood, A. B. Moser, and H. W. Moser. Statistical analysis: Dubey and Mahmood. Obtained funding: Raymond and H. W. Moser. Administrative, technical, and material support: Cox, Raymond, Mahmood, A. B. Moser, and H. W. Moser. Study supervision: Cox, Raymond, and H. W. Moser.

Funding/Support: This study was supported by the Johns Hopkins University School of Medicine General Clinical Research Center grant M01-RR00052 from the National Center for Research Resources/National Institutes of Health, Baltimore, Md, and grants HD10981 and HD39276 from the National Institutes of Health, Bethesda, Md.

Acknowledgment: We thank Dr Martha Denckla for her review of the manuscript and her valuable suggestions. We also thank the patients and families for their support and cooperation, as well as our research nurses Pauline Green and Kim Hollandsworth, program coordinator Willie Foreman, and secretary Annette Snitcher for their caring personal commitment to the study.