|
|
| |||||||||||||||||||||||||||||||||||||||||
Background Adrenomyeloneuropathy is an X-linked recessive disorder characterized by myelopathy, peripheral neuropathy, and cerebral demyelination, which develop in association with the accumulation of very-long-chain fatty acids. The administration of oleic and erucic acids inhibits the synthesis of very-long-chain fatty acids. Recently such dietary treatment has been widely publicized as a possible cure for this disease.
Methods We conducted an open trial in 14 men with adrenomyeloneuropathy, 5 symptomatic heterozygous women, and 5 boys (mean age, 13 years) with preclinical adrenomyeloneuropathy. The patients ate a low-fat diet and received daily doses of glycerol trioleate oil (1.7 g per kilogram of body weight) and glycerol trierucate oil (0.3 g per kilogram). Clinical manifestations, cerebral and spinal cord magnetic resonance imaging (MRI) scans, nerve conduction, and brain-stem auditory and somatosensory evoked potentials were studied prospectively over 18 to 48 months. Plasma levels of very-long-chain fatty acids and the side effects of erucic acid were monitored monthly.
Results By week 10, plasma very-long-chain fatty acid levels declined nearly to normal. Nonetheless, over a mean follow-up of 33 months none of the 14 men with adrenomyeloneuropathy improved. In nine men there was functional deterioration, coincident in four with new cerebral lesions on MRI. In a single patient there was a reduction in cerebellar demyelination, but without clinical improvement. In one of the five asymptomatic boys signs of myelopathy developed. There were no changes in the symptomatic heterozygous women. There was some improvement in peroneal-nerve conduction, but no detectable clinical improvement. Conduction to the parietal cortex (T12-P37 interpeak latency) worsened in both the symptomatic men and the boys with preclinical adrenomyeloneuropathy. There was no change in other somatosensory evoked potentials or in brain-stem auditory evoked potentials. Asymptomatic thrombocytopenia (<100,000 cells per cubic millimeter) was noted in six patients.
Conclusions In this open trial we found no evidence of a clinically relevant benefit from dietary treatment with oleic and erucic acids ("Lorenzo's Oil") in patients with adrenomyeloneuropathy. .
Adrenomyeloneuropathy and cerebral adrenoleukodystrophy are associated with impaired degradation of saturated very-long-chain fatty acids, which accumulate in cerebral white matter and the adrenal cortex2. Very-long-chain fatty acids are derived both from the diet and from endogenous synthesis by a microsomal system that elongates long-chain fatty acids7. On the basis of the as yet unproved hypothesis that very-long-chain fatty acids are toxic to myelin sheaths,1 therapeutic approaches have been developed to reduce their concentrations. The simple restriction of dietary very-long-chain fatty acids led to no biologic or clinical improvement8. In 1986, oleic acid, an inhibitor of the elongation of fatty acids,9 was found to reduce the levels of very-long-chain fatty acids in cultured fibroblasts from patients with adrenoleukodystrophy10. Trials using glycerol trioleate oil and a restricted diet decreased the plasma level of very-long-chain fatty acids by 50 percent in most patients11,12. Small changes in sural- and peroneal-nerve conduction rates suggested a possible effect of glycerol trioleate oil,12,13 but no improvement was seen in clinical manifestations and on brain magnetic resonance imaging (MRI)12,13. Because of the apparently poor clinical results of glycerol trioleate oil alone, new trials using glycerol trierucate, a more potent inhibitor of the synthesis of very-long-chain fatty acids, were launched, allowing the normalization of plasma very-long-chain fatty acid levels within four to six weeks14. However, despite the impression given by the recent popular movie, Lorenzo's Oil,15 erucic acid did not prevent neurologic deterioration in patients with cerebral adrenoleukodystrophy14,16.
We studied the effects of glycerol trioleate and trierucate oils in men with adrenomyeloneuropathy, boys with preclinical adrenomyeloneuropathy, and symptomatic heterozygous women (i.e., subjects with the less severe forms of adrenoleukodystrophy).
Methods
Patient Population
This open trial included 14 men with adrenomyeloneuropathy, 5 symptomatic heterozygous women, and 5 boys with preclinical adrenomyeloneuropathy. All the men with adrenomyeloneuropathy and all the heterozygous women had myelopathy2; 11 of the men and 2 of the women had mild neuropathy2. The five boys had no neurologic symptoms6. The mean follow-up was 32.7 months in the men with adrenomyeloneuropathy (range, 18 to 48), 25.8 months in the symptomatic heterozygous women (range, 18 to 38), and 29.6 months in the boys with preclinical adrenomyeloneuropathy (range, 18 to 36). The study was approved by the institutional review board of the Hopital St. Vincent de Paul, Paris. Informed consent was obtained from the patients or their parents.
Clinical Assessment
The patients underwent standard physical examination, electrocardiography, echocardiography, and chest radiography at entry, after 6 months, and then every 12 months, with measurement of electrolyte, liver-enzyme, vitamin A and E, lipid, and essential fatty acid concentrations in plasma17. Plasma very-long-chain fatty acids (C22:0, C22:1, C24:0, C24:1, C26:0, and C26:1) were measured twice a month during the first three months, and then monthly18. Platelet and leukocyte counts were obtained every three months.
At each visit, two neurologists performed standardized neurologic examinations. Clinical efficacy was assessed with the Expanded Disability Status Scale (EDSS)19 and the functional-system subscale of the Multiple Sclerosis Minimal Record of Disability and Ambulation Index20. The degree of perceived global change was rated independently by the neurologists and the patients according to the following scale: A, marked improvement; B, moderate improvement; C, no change; D, moderate worsening; and E, marked worsening.
Cerebral function was evaluated in the adults with the Mini-Mental State examination and in the boys with the revised Wechsler Intelligence Scale. Patients with abnormal results on these tests or abnormalities on brain MRI underwent a more extensive neuropsychological evaluation.
Neuroradiologic Evaluation
Cerebral and spinal cord MRI was performed with a 0.5-T unit (General Electric Medical Systems, Milwaukee),6 with identical sequential tomographic sagittal, coronal, and axial sections. The demyelinated areas were quantified,21,22 and lesions were graded as mild, moderate, or severe. The estimated accuracy of measurement was within 9.6 percent, and variation in a single observer was within 2.3 percent21,22.
Electrophysiologic Evaluation
We measured motor- and sensory-nerve conduction velocities, including F-wave latencies and bilateral H-reflexes; brain-stem auditory evoked potentials; somatosensory evoked potentials; and visual evoked potentials with pattern stimuli, as previously described6. For somatosensory evoked potentials, the signals were registered as follows: for the lower limbs, at the spinal process to the iliac crest (T12) and in parietal areas of the scalp (2 cm behind the reference electrode Cz) (P37); for the upper limbs, at Erb's point (N10), at the level of the spinal process of the second cervical vertebra (N11), and in parietal areas of the scalp (2 cm behind and 7 cm lateral to the reference electrode Cz) (P14 and N20). To verify that changes were not due to variability in measurement, we determined that variation in a single observer for four nerve-conduction attributes (the amplitude and nerve conduction velocity of the peroneal and sensory median nerves) ranged from 2 to 4.5 percent.
Diet Design
The patients followed a low-fat diet that provided 10 percent of calories as fat, with less than 10 to 15 mg of C26:0 fatty acid per day. They received a mean (±SD) of 1.7 ±0.6 g of glycerol trioleate oil per kilogram of body weight per day and 0.3 ±0.05 g of glycerol trierucate oil per kilogram per day. Glycerol trioleate and trierucate oils were kindly provided by the Societe Industrielle des Oleagineux, Saint-Laurent-Blangy, France.
To avoid a deficiency of essential fatty acids, safflower oil (10 to 15 ml) and fish oil (2-g capsules) containing eicosapentaenoic acid (18 percent) and docosahexaenoic acid (12 percent) were added daily to the diet.
Statistical Analysis
Comparisons were performed with the chi-square test for trend for the global assessments by physicians and patients and with the Wilcoxon rank-sum test for other variables, including scores on the EDSS and the ambulation index. All reported P values are paired and two-sided.
Results
Changes in Plasma Very-Long-Chain Fatty Acid Levels
In all the patients, plasma C26:0 concentrations decreased to normal or nearly normal values within 10 weeks of treatment (Figure 1). In 9 of the 14 men with adrenomyeloneuropathy, the 5 boys with preclinical adrenomyeloneuropathy, and the 5 heterozygous women, C26:0 values were completely normal at the end of the follow-up period. In most patients, plasma C26:0 levels returned to normal when the plasma level of erucic acid was maintained between 125 and 160 µmol per liter. All the patients complied with the diet.
|
In the men with adrenomyeloneuropathy the mean EDSS score increased by 1.0 point (P<0.02) and the ambulation-index score by 1.26 points (P<0.001) (Table 1). These changes reflected mainly the worsening myelopathy in four men (Patients 1, 2, 5, and 6) who became dependent on a wheelchair. The worsening scores on the EDSS or ambulation index were not correlated with the duration of disability before the diet began. There was no significant change in the symptomatic heterozygous women or the boys with preclinical adrenomyeloneuropathy, except Patient 21 (Table 1).
|
|
According to the physician's assessment, 9 of the 14 men had more manifestations of the disease (Table 1). None of the men improved, even moderately. There was no change in the global assessment of the boys with preclinical adrenomyeloneuropathy or the symptomatic heterozygous women. Self-assessment by the patients was concordant (Table 1). The two men in whom dementia developed (Patients 7 and 13) could not give self-assessments.
Neuroradiologic Changes
Twelve men with adrenomyeloneuropathy had abnormal brain MRI scans before the diet (Table 2). In the seven with demyelination restricted to the internal capsule, the lesions remained unchanged in five and worsened markedly in two. In Patient 4, lesions that were initially in the left internal capsule increased fivefold and extended markedly into the left frontal lobe (Figure 2). In Patient 1 demyelination appeared in the corpus callosum and occipital lobes after 24 months of the diet and doubled in size during the following 14 months.
|
In two symptomatic heterozygous women, demyelinating lesions of the internal capsule remained unchanged. MRI results remained normal in all the boys with preclinical adrenomyeloneuropathy. Spinal cord atrophy did not correlate with the severity of neurologic disability before the diet and did not change during follow-up.
Electrophysiologic Changes
The men with adrenomyeloneuropathy had an 8 percent improvement (95 percent confidence interval, 2.0 to 13.5 percent) in peroneal-nerve conduction at the end of the follow-up period, with no changes in average conduction velocities of the posterior tibial, sensory, and motor median nerves (Figure 3). The conduction velocity of the posterior tibial nerve improved by 19 percent (95 percent confidence interval, 11.6 to 26 percent) in six of the men and declined by 6.5 percent (95 percent confidence interval, 4.7 to 8.4 percent) in three others. Similarly, sensory-median-nerve conduction improved by 20 percent (95 percent confidence interval, 15.7 to 25 percent) in six men and declined by 15 percent (95 percent confidence interval, 11 to 18 percent) in three.
|
Auditory conduction in the brain stem (wave I-V interpeak latency) was delayed in all the patients before the diet, and it remained unchanged in the men with adrenomyeloneuropathy and the symptomatic heterozygous women at the end of the follow-up period (Figure 3). The average wave I-V interpeak latency worsened by 0.18 ±0.04 msec in the five boys with preclinical adrenomyeloneuropathy.
We observed no change in median-nerve (N10 latency), medullothalamic, or thalamoparietal (N11-P14 and P14-N20 interpeak latencies) conduction times in any of the patients (Figure 3).
The conduction velocity of the posterior tibial nerve (T12 latency) improved by 7 percent (95 percent confidence interval, 4.0 to 10 percent) in the men with adrenomyeloneuropathy (Figure 3). Only one heterozygous woman had abnormal T12 latency before the study diet, and this variable improved by 5 percent. T12 latency improved by 1 SD in the five boys with preclinical adrenomyeloneuropathy (P<0.05).
After stimulation of the posterior tibial nerve, conduction to the parietal cortex (T12-P37 interpeak latency) could be measured in seven of the men with adrenomyeloneuropathy (the others had no identifiable P37 peak); it worsened by 17 percent (95 percent confidence interval, 2 to 31 percent) (Figure 3). This variable remained unchanged in the heterozygous women. The boys with preclinical disease all had abnormal T12-P37 interpeak latency that increased by an average of 1 SD.
All but one man with adrenomyeloneuropathy (Patient 7) had normal visual evoked potentials at entry. P100 latency worsened in this patient and in Patient 13, whose demyelination increased in the parieto-occipital lobes. There was no change in visual evoked potentials in the other men, the heterozygous women, or the boys with preclinical disease.
Side Effects
The platelet count declined in 23 of the 24 patients,23 with no correlation with plasma levels of erucic, arachidonic, and docosahexaenoic acid. Twelve of the patients had platelet counts between 100,000 and 150,000 per cubic millimeter, and six had counts below 100,000 per cubic millimeter. The lowest value during the period of dietary treatment was 52,000 per cubic millimeter. None of the patients had abnormal bleeding or hematoma. Three patients had asymptomatic neutropenia (500 to 700 cells per cubic millimeter). Two men with adrenomyeloneuropathy had mild asymmetric cardiac septal hypertrophy after 12 months of the study diet, but ventricular function remained normal. None of the patients had abnormal electrocardiograms or cardiac enlargement. A 12 percent decrease (95 percent confidence interval, 5 to 20 percent) in the cholesterol level was observed in the three groups, with a 13 percent increase (95 percent confidence interval, 4 to 24 percent) in high-density lipoprotein cholesterol only in the men with adrenomyeloneuropathy.
Although all the patients received safflower and fish oil rich in n-3 fatty acids, a 24 percent (95 percent confidence interval, 17 to 31 percent) decrease in plasma arachidonic acid levels and a 30 percent (95 percent confidence interval, 16 to 42 percent) decrease in plasma docosahexaenoic acid levels were observed in all the patients, but with no symptoms of essential fatty acid deficiency.
Discussion
Early symptoms of adrenomyeloneuropathy usually appear when patients are between 24 and 28 years of age. These patients have myelopathy, often with mild peripheral neuropathy, and half of them have moderate cerebral involvement demonstrable by MRI and neuropsychological tests2,6,24. There are no data, however, that document the natural history of untreated adrenomyeloneuropathy. A review of the literature indicates that 35 percent of patients with adrenomyeloneuropathy have a marked progression of their disease within three years of diagnosis,3,4,6,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48 the duration of the present trial. At least 30 percent of women who are heterozygous for adrenoleukodystrophy have spastic paraparesis, usually milder and of later onset (at about the age of 40) than in men with adrenomyeloneuropathy2,5. Patients with preclinical adrenomyeloneuropathy may have sensory disturbances but no neurologic disability6. Although one cannot be certain that these patients will have full expression of adrenomyeloneuropathy, the electrophysiologic abnormalities are predictive of clinical disease.
The myelopathy characteristically includes degeneration of the corticospinal and gracile tracts, medial lemnisci, and spinocerebellar pathways in the medulla49. The glycerol trioleate and trierucate therapy did not protect the four men whose myelopathy worsened during the trial. In the other men with adrenomyeloneuropathy and the symptomatic heterozygous women, the clinical and MRI manifestations of myelopathy remained stable. A marked increase in T12-P37 interpeak latency in the patients with clinical and preclinical adrenomyeloneuropathy indicated a worsening.
Regarding peripheral neuropathy, the men with adrenomyeloneuropathy, the symptomatic heterozygous women, and the boys with preclinical disease had some improvement in peroneal-nerve conduction, and more limited improvement in posterior-tibial-nerve and sensory-median-nerve conduction, that was more pronounced than with glycerol trioleate alone12,13,24 but still far from restoring normal conduction or reversing the disability. Since it is not known to occur spontaneously, this slight improvement could be due to the glycerol trioleate and trierucate therapy.
The lack of effect of dietary therapy on demyelinating lesions in our patients with adrenomyeloneuropathy is consistent with observations in patients with cerebral adrenoleukodystrophy, even when demyelination is moderate14,16,24. We have confirmed the lack of benefit in a separate group of 35 children with cerebral adrenoleukodystrophy. Although the present data in patients with adrenomyeloneuropathy with cerebral lesions leave little likelihood of a positive effect of glycerol trioleate and trierucate oils, we cannot completely exclude the possibility that progression would have been more frequent or more rapid without the diet. Given the rapid extension of cerebral lesions in more than one third of the patients in our study, we consider any real benefit unlikely. On the other hand, the patients with clinical and preclinical adrenomyeloneuropathy and the symptomatic heterozygous women who had no cerebral lesions before the trial remained normal during the period of observation. Because of the limited frequency of cerebral demyelination in these forms of adrenoleukodystrophy, as well as its long and variable evolution,2,5,6,24 additional data are needed to determine whether a diet of glycerol trioleate and trierucate oils can prevent the development of the lesions.
Our trial documented various degrees of progression in preexisting myelopathy and cerebral lesions in a large fraction of patients with clinical or preclinical adrenomyeloneuropathy and heterozygous women. In the absence of an untreated control group, however, we could not determine whether glycerol trioleate and trierucate oils slowed the rate of deterioration. There are several obstacles to placebo-controlled trials in this disease. First, because of wide publicity, the public's high expectation of benefit from glycerol trioleate and trierucate has made the administration of a placebo ethically impossible. Second, adrenomyeloneuropathy affects only 1 in 45,000 people in Europe and the United States,1,2 which makes a controlled trial almost impossible. In fact, given the variability of the disease's manifestations and its long and variable course, such a trial would have to run for 5 to 10 years to allow the detection of any moderate therapeutic effect that could have escaped the present open trial. We do not think that a possible limited benefit from glycerol trioleate and trierucate merits the organization of large multicenter trials in patients with advanced myelopathy or cerebral lesions. In these patients, bone marrow transplantation is a promising therapy despite the associated morbidity50.
To a limited extent, our data support a positive effect of glycerol trioleate and trierucate on peripheral-nerve conduction. These changes did not modify the clinical disabilities of patients with advanced adrenomyeloneuropathy, but they may be encouraging as a way to prevent the peripheral neuropathy. For patients in whom peripheral neuropathy predominates, early and prolonged therapy with glycerol trioleate and trierucate oils may be useful. Proof of a clinically important benefit from this dietary treatment requires a large, controlled trial. Since no clear benefits were evident in our small group of partients with preclinical adrenomyeloneuropathy, there should be no practical or ethical obstacle to the use of a placebo control.
The expectation that Lorenzo's oil would produce clinical benefits in patients with cerebral adrenoleukodystrophy or adrenomyeloneuropathy stemmed from its effects on the synthesis of very-long-chain fatty acids. Despite the near-normalization of plasma levels of very-long-chain fatty acids, we saw no improvement in preexisting myelopathy and cerebral lesions in our patients. No animal model of adrenoleukodystrophy is available, and therefore neither the manipulation of very-long-chain fatty acid levels nor the mechanisms of demyelination can be studied experimentally. The recent identification of the adrenoleukodystrophy gene1 may lead to a "gene knock-out" murine model that could help in the search for therapies, including gene transfer to hematopoietic cells.
Source Information
From INSERM Unite 342 (P.A., F.R., N.C.) and the Departments of Radiology (C.A.), Neurology (I.J., A.L.), and Pediatric Endocrinology (C.J., P.-F.B.), Hopital St. Vincent de Paul; and the Departments of Neurophysiology (M.-C.L.-R., F.B.), and Biochemistry (C.W.), Hopital St. Antoine -- both in Paris.
Address reprint requests to Dr. Aubourg at INSERM U342, Hopital St. Vincent de Paul, 82 Ave. Denfert-Rochereau, 75014 Paris, France.
References
| |||||||||||||||||||||||||||||||||||||||||
Related Letters:
More on Lorenzo's Oil
Odone A., Odone M., Aubourg P., Rizzo W. B.
Extract |
Full Text
N Engl J Med 1994;
330:1904-1905, Jun 30, 1994.
Correspondence
Lorenzo's Oil and Lymphocytopenia
Unkrig C. J., Schroder R., Scharf R. E., Aubourg P.
Extract |
Full Text
N Engl J Med 1994;
330:577, Feb 24, 1994.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
