ADRENAL INSUFFCIENCY DUE TO X-LINKED ADRENOLEUKODYSTROPHY

Hugo W. Moser, MD

Gregory Kaltsas, MDAssistant Professor of Medicine,Dept of Pathophysiology,Laikon University Hospital,Athens 115 27, Greece, email: GKaltsas@endo.gr

Revised 9 Feb 2009

INTRODUCTION

X-linked adrenoleukodystrophy (X-ALD) is a frequent but underrecognized cause of primary adrenocortical insufficiency secondary to peroxisomal dysfunction biochemically characterized by accumulation of saturated very long chain fatty acids. At least six phenotypes can be distinguished that vary in the age and severity of clinical presentation. Two of the most frequent forms encountered are childhood cerebral ALD and adrenomyeloneuropathy (AMN), which can coexist (ALD/AMN) (1-3). Variations in methionine metabolism may be associated with the phenotypic spectrum noted in this ALD/AMN complex (3). Diagnosis is relatively easy biochemically and prenatal testing is possible in affected families (4;5).

BIOCHEMISTRY AND GENETICS OF X-ALD

X-ALD is a genetically determined disorder that involves mainly the white matter and axons of the central nervous system (CNS), the adrenal cortex and the testis (4). It is associated with the accumulation of saturated very long chain fatty acids (VLCFA), particularly hexacosanoic (C26:0) and lignoceric (C24:0) acids, due to the impaired capacity to degrade these substances (5), a reaction that normally takes place in the peroxisome (5;6). The gene that is defective is referred to as ABCD1. It is located at Xq28. It codes for a peroxisomal membrane protein (ALDP) (7), a member of the ATP binding cassette (ABC) transport protein family that helps form the channel through which VLCFAs move into the peroxisome. (8;9). Transfection of X-ALD cell lines with normal ABCD1 restores their capacity to degrade to VLCFA (9), through mechanisms that have not yet been determined (10). Accumulation of abnormal VLCFA in affected organs is presumed to underlie the pathologic process of the adrenoleukodystophies (11). CNS pathology is marked by diverse immune responses involving cellular and humoral mechanisms as well as cytokines and complement (3) in addition, to oxidative damage (lipid peroxidation). In the adrenal gland, abnormal VLCFA may directly alter cellular function by inhibiting the effects of ACTH on the adrenocortical cells, or indirectly by initiating an autoimmune response. In almost all instances, adrenocortical failure occurs along with irreversible degenerative neurologic defects.

More than 400 different mutations have been identified in X-ALD patients (12) and are updated in the website http://www.x-ald.nl). The mode of inheritance of X-ALD is X-linked recessive whereas the phenotype does not correlate with the type of mutation. The lowest estimate birth incidence in the United States was estimated at 1 in 21,000 and 1 in 16,800 for hemizygotes and hemizygotes plus heterozygotes, respectively (13). X-ALD has been reported in all ethnic groups with approximately the same frequency.

CLINICAL MANIFESTATIONS OF X-ALD

The clinical manifestations of X-ALD are summarized in several recent reviews (4;14;15). The range of clinical expression varies widely. Tables 1a and 1b summarize the principal phenotypes. Approximately 40% of patients have the childhood, adolescent or adult cerebral forms, with the childhood form being the most severe. These phenotypes are rapidly progressive. They are associated with an inflammatory response (16), in which autoimmune mechanisms may play a role (17) and which is associated with characteristic brain MRI (18) and magnetic resonance spectroscopy (19-21). Histopathologically, there is inflammatory demyelination, resulting in confluent and bilaterally symmetric loss of myelin in the cerebral and cerebellar white matter (22). The parieto-occipital regions are usually affected first, with asymmetric progression of the lesions toward the frontal or temporal lobes. In general, arcuate fibers are spared, except in chronic cases. Axonal loss may be considerable, but myelin loss is usually greater. Lesions may sometimes involve the brainstem, especially the pons. The spinal cord is usually spared, except for bilateral corticospinal tract degeneration.

In contrast, adrenomyeloneuropathy (AMN) is a slowly progressive disorder that affects the long tracts of the spinal cord mainly and in which the inflammatory response is absent or mild (22;23). Affected individuals also develop a degenerative axonopathy that involves the ascending and descending tracts of the spinal cord, especially in fasciculus gracilis and the lateral corticospinal tracts. The histologic pattern is Wallerian degeneration (22). AMN patients may survive to the eighth decade. However, approximately 20-30% of AMN patients later develop progressive cerebral involvement in which the inflammatory response is present (24). Approximately 50% of female heterozygotes develop an AMN like syndrome, but milder, in middle or later life. Almost all male X-ALD patients develop some degree of neurologic or endocrine abnormality at some time in their life. The various phenotypes often co-occur within the same family. The nature of the mutation or the degree of elevation of plasma VLCFA levels is not predictive of phenotype.

INCIDENCE OF ADRENAL DEFICIENCY IN X-ALD

Correlation between the severity of adrenal and neurological involvement. The incidence of adrenal insufficiency in the various phenotypes is shown in table 1. While many patients have both neurologic involvement and adrenal insufficiency, a considerable nubmer have only or the other. The patients with the "Addison only" phenotype by definition are free of demonstrable neurologic involvement, although many in this category will later become neurologically involved. ALD is the cause for up to 20 percent of boys with idiopathic Addison disease. Biochemical evidence of adrenal insufficiency can be present for up to two years before the development of clinical signs (25). However, in 20 of 41 AMN patients studied by Brennemann et al (26) the ACTH stimulation test gave normal results. The incidence of adrenal insufficiency in the patients with the childhood cerebral forms of ALD (79%) appears to be higher than in the AMN patients.

Addison disease is rare in women heterozygous for X-ALD (1% or less), and considerably less frequent than the AMN-like syndrome, which develops in approximately 50% of women in middle age or later. Even though it is rare for heterozygous women to show clinically evident adrenal insufficiency or abnormalities in plasma ACTH level or ACTH stimulation test, postmortem studies have shown abnormalities that resemble those in affected males (27). When more subtle tests of adrenal function, such as the response to ovine corticotropin-releasing-hormone, were performed, subnormal responses were demonstrated in five of eight women whose ACTH stimulation tests were normal (28).

X-ALD appears to be a more frequent cause of Addison's Disease in males than is generally recognized. Lauretti et al (29) found that 5 of 14 male patients, ages 12-45 years, previously diagnosed as having primary adrencortical insufficiency, had abnormally high plasma VLCFA levels and had X-ALD. Jorge et reported X-ALD in ten of 37 patients with idiopathic Addison Disease (27%), and found that the incidence was highest (5 out of 5) in those patients in whom adrenal insufficiency became evident before 7.5 years of age (30). These results have important clinical implications. The diagnosis of X-ALD has profound implications for prognosis, therapy and genetic counseling. It is therefore important that screening for X-ALD be carried out in all male patients with idiopathic Addison Disease. The need to do so is particularly great in patients in whom the adrenal insufficiency manifested before 7.5 yeas of age.

Table 2 shows the patterns of adrenal dysfunction in patients with various X-ALD phenotypes. Elevated ACTH levels and impaired cortisol response to ACTH administration are the most frequent finding, but impaired aldosterone response was also observed in one third of the patients.

DIAGNOSIS OF X-ALD

The plasma assay for VLCFA is the most frequently used diagnostic assay. It is reliable for the identification of affected males (31) VLCFA levels are already increased on the day of birth and in untreated patients remain approximately the same throughout life. The assay can be used to identify asymptomatic patients by screening members of the extended family (11). While plasma levels of VLCFA are increased in many women who are heterozygous for X-ALD, false negative results occur in approximately 15 to 20% of obligate heterozygotes (31). Mutation analysis is the most accurate method for the definitive identification of heterozygotes (4). Studies of chorion villus biopsies or amniocytes permit the prenatal identification of affected male fetuses (32).

The diagnosis of X-ALD should be suspected in:

1. Boys with progressive behavioral, cognitive or neurologic disturbances beginning at 3 years of age or later. The initial manifestations resemble those of attention deficit disorder, and even in X-ALD patients may appear to respond to medications such as ritalin. In X-ALD evidence of dementia, spatial disorientation, more serious behavioral disturbances, and difficulty in hearing supervene.

2. Males with Addison's disease in which etiology has not been defined. Since the plasma VLCFA assay is non-invasive, and the practical and genetic implications of the diagnosis of X-ALD are great, the VLCFA assay be part of the routine initial evaluation of male patients with Addison Disease.

3. Men and women with progressive spastic paraparesis. AMN is often misdiagnosed as multiple sclerosis. The diagnosis of X-ALD should be considered even when there is no clinical or biochemical evidence of adrenal insufficiency. In a series from Germany (26) adrenal function was normal in 20 of 41 men with AMN, and adrenal insufficiency occurs in less than one percent of women with and AMN-like syndrome.

4. Patients in whom adrenal insufficiency occurs in combination with neurologic disability (Table 3).

5. Patients who are at genetic risk of having X-ALD on the basis of pedigree. Because X-ALD is X-linked recessive, a large number of relatives in the nuclear and extended family are at genetic risk. In one family we screened 174 family members. Bezman et al (13) reported that extended family screening led to the identification of 504 affected males, half of whom were asymptomatic, and of 1,270 heterozygotes. Detection of asymptomatic patients is particularly important, since therapeutic interventions have the greatest chance of success when clinical manifestations are still mild.

Testing typically includes three VLCFA parameters: the level of hexacosanoic acid (C26:0), and the ratio of hexacosanoic acid to tetracosanoic acid (C26:0/C24:0), and to docosanoic acid (C26:0/C22:0) (31). VLCFA levels are also elevated in some other peroxisomal disorders. The diagnosis should be confirmed by molecular genetic testing of the ABCD1 gene locus (36).

All individuals with confirmed ALD/AMN complex, including symptomatic female heterozygotes, should undergo neuroimaging to determine if cerebral involvement is present, and testing of adrenal function. Adrenal function should be evaluated by measurement of plasma ACTH concentration and the rise in plasma cortisol concentration following ACTH stimulation. Cranial MRI is always abnormal in symptomatic males, demonstrating demyelination in cerebral white matter. The occipitoparietal region is typically affected and lesions are usually bilateral, but can be limited to only one side (37). The progression of the disease correlates with the presence of contrast enhancement on T1-weighted MR images (38). Proton MR spectroscopy detects white matter abnormalities that may not be apparent on conventional MR imaging and may predict disease progression (21;39).

PATHOLOGY OF THE ADRENAL GLAND IN X-ALD

In patients with X-ALD the adrenocortical cells, particularly those in the inner fasciculata-reticularis, become ballooned and striated due to the accumulation of lamellae, lamellar-lipid profiles, and fine lipid clefts (40) (Figure 1). Histochemical and biochemical studies make it likely that these lammellae consist of cholesterol esterified with saturated very long chain fatty acids (41;42). The inclusions and strikingly elevated levels of very long chain fatty acids are also demonstrable in the fetal adrenal gland (43). The striated material appears to lead to cell dysfunction atrophy and death of the cells. Whitcomb et al carried out studies of the cortisol response to stimulation with increasing concentrations of ACTH in cultured adrenal cells, and found that this response was impaired when the culture medium contained very long chain fatty acids (Figure 2). Ultimately primary atrophy of the adrenal cortex ensues. Inflammatory cells are rarely involved.

THERAPY

For those patients with X-ALD who have impaired adrenal function glucocorticoid replacement therapy is mandatory and life-saving. In patients with normal adrenal function at time of first contact, adrenal function should be monitored yearly. Glucocorticoid replacement requirements are generally the same as in other forms of primary adrenal insufficiency whereas some patients may not require mineralocorticoid replacement. While there is one report of substantial improvement of neurologic function when replacement therapy was administered to a patient with AMN (44) the general impression is that adrenal replacement therapy does not alter neurologic progression.

Therapy of the neurologic aspects of X-ALD is a major challenge. Hematopoietic stem cell transplantation (HSCT), is emerging as the treatment of choice for individuals with early stages of cerebral involvement in ALD (14;45) and can lead to long-term stabilization and occasionally improvement (14;46).Stem cells can be harvested from peripheral blood, bone marrow, and umbilical cord blood. Although the mechanism of this effect is still unclear, bone marrow cells do express the ABCD1 gene and plasma very long chain fatty acid levels are reduced after bone marrow transplantation. It has been shown that bone marrow-derived cells do enter the brain and that part of the brain microglial cells are bone marrow-derived (47)). It may also diminish the brain inflammatory response. Brain MRI abnormalities precede symptoms in patients with the cerebral forms of X-ALD (48). Follow-up for 5 to 10 years after bone marrow transplantation in 12 boys with childhood cerebral ALD revealed that MRI abnormalities reversed in two, improved in one, and did not change in one (46). Eight participants stabilized and remained unchanged following an initial period of continued demyelination. The plasma VLFA concentration decreased by 55 percent and remained slightly above the normal range. In another study that evaluated the outcome of HSCT of 126 boys with cerebral ALD, a 92 percent five-year survival was found in a 25-member subgroup in which transplant was performed in the early stage of the disease (49). A later retrospective report identified a nontransplanted 30-member subgroup of patients with early-stage cerebral ALD who were matched by neurologic disability and MRI severity scores with a transplanted 19-member subgroup with early-stage cerebral ALD (50). The five-year survival was significantly worse for the nontransplanted patients compared with the transplanted group (54 versus 95 percent).

Current strategy is to monitor asymptomatic patients by MRI at 6 months to yearly intervals and consider HSCT when the MRI abnormality is advancing and clinical disability is mild. It is not recommended for patients who already have advanced cerebral involvement, because it has not reversed these severe deficits and in some instances may have accelerated disease progression. Bone marrow transplantation carries a high risk. It is not recommended for patients without cerebral involvement because of the high risk and because up to 50% of untreated X-ALD patients never develop cerebral involvement. It has not been tested systematically in AMN because of concern that the risk-benefit ratio may not be favorable, and it is uncertain whether the procedure will affect the non-inflammatory distal axonopathy which is the main pathological feature in AMN (22) and differs from that in the cerebral forms of the disease.

Other therapeutic modalities include dietary therapies with restriction of fat intake and the administration of a mixture of glyceryl trioleate and glyceryl trierucate, also referred to as Lorenzo's Oil (51). This therapy normalizes plasma very long chain fatty acid levels within four weeks, but its therapeutic effects in patients who are already symptomatic has been disappointing (52). An international trial to determine whether administration of the oil to neurologically asymptomatic patients prevents or diminishes subsequent neurological disability is in progress. Lorenzo's Oil therapy does not improve adrenal function (15;52). Other therapies that are under consideration of being tested are 4-phenylbutyrate (53), Lovastatin (54;55), and gene therapy (56).