Dehydroepiandrosterone in the Treatment of Neuropsychiatric Conditions

Owen M. Wolkowitz

Victor I. Reus

Nicole Maninger

Synthia Mellon

Dehydroepiandrosterone (DHEA) has been evaluated in the treatment of neuropsychiatric disorders almost from the time of its initial discovery and synthesis, with published reports appearing as early as 1952 (1,2). Several uncontrolled positive reports were published in the 1950s, but large-scale enthusiasm for this potential therapy languished until the late 1980s through the mid-1990s. At that time, an expanding body of preclinical data, plus the first adequately controlled clinical trial (3), fostered hope that DHEA might, conservatively, increase well-being and, optimistically, extend life, protect the brain, and retard the ravages of aging.

Popular book claims that DHEA is the next antiaging and miracle hormone as well as the widespread unregulated use of DHEA (due to the 1994 Dietary Supplement Health and Education Act in the United States) have concerned many medical investigators and practitioners, because much of the evidence for DHEA’s efficacy comes from preclinical studies, and such data may not readily extrapolate to humans (4). Further, much of the clinical data that exist have been derived from uncontrolled studies, and the full risk-benefit ratio of long-term DHEA use remains unknown (5,6). This chapter seeks to put the possible role of DHEA and its more stable sulfate ester, DHEA-Sulfate (DHEA-S), as psychotropic agents, into scientific context and to review clinical data with regard to the use of DHEA in neuropsychiatric illnesses. The main focus will be on clinical efficacy and safety in the treatment of depression, schizophrenia, and dementia. More extensive reviews of pertinent preclinical studies are available elsewhere (7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21).

DEHYDROEPIANDROSTERONE AS A NEUROSTEROID HORMONE

The mechanism by which DHEA and DHEA-S [jointly referred to here as “DHEA(S)”] operate, other than by serving as prohormones for testosterone and estradiol, is not fully understood. A list of possible mechanisms relevant to neuropsychiatric illness is provided in Table 6.1; additional discussions of DHEA(S)’s mechanisms are detailed elsewhere (18, 19, 20, 21, 22, 23, 24, 25). Traditional steroids such as cortisol, estrogen, and progesterone affect gene transcription by binding to specific cytoplasmic receptors, which then translocate into the nucleus, where they bind to steroid-binding elements on the DNA. No specific receptors have yet been identified for DHEA(S), although intriguing leads are now emerging (23,26). The possibility was raised that DHEA, at very high concentrations, is able to directly activate the estrogen receptor-beta (27) and, after oxidation, is able to bind to intracellular progesterone receptors (18). In some tissues, DHEA-S activates peroxisome proliferator-activated receptor alpha (PPAR alpha), an intracellular receptor belonging to the steroid receptor superfamily (28, 29, 30). Such activation may be related to DHEA-S’s ability to regulate nuclear factor κB
(NF-κB) expression and to exhibit antioxidant effects (28, 29, 30). In contrast to the glucocorticoid hormone cortisol, DHEA(S) affect neurons via an additional distinct mechanism, noncompetitive interactions at the GABA_A, NMDA, and sigma receptors (8,19, 20, 21,31,32). DHEA(S)’s actions at the GABA_A receptor are generally antagonistic in nature, with DHEA-S having more potent antagonistic effects than DHEA (33, 34, 35). Importantly, however, various metabolites of DHEA, such as androstanediol and androsterone, have GABA_A receptor agonistic effects. Therefore, the balance of GABA_A receptor agonism versus antagonism may depend on the local concentrations of DHEA versus DHEA-S versus DHEA metabolites.

TABLE 6.1 Possible Mechanisms of Neuropsychiatric Effects of DHEA(S)

GABA-A receptor antagonism (and perhaps weak agonism) (8,35,155)

Metabolized to GABA-A receptor agonist neurosteroids such as androstanediol and androsterone (156, 157, 158, 159, 160)
Metabolized to testosterone and estrogen (152,156,161,162)
NMDA and sigma-receptor potentiating effects (163, 164, 165, 166)
Possible activation of the estrogen receptor-beta (at very high concentrations) and the progesterone receptor (after oxidation) (18,23,27,167,168)
Activation of intracellular peroxisome proliferator-activated receptor alpha (PPAR alpha) (in the liver) (23,28,29,169)
Inhibition of the transcription factor, nuclear factor-kappa B (NF-κB) (170)
Increases brain regional serotonin and dopamine activity (171,172)
Increases hippocampal primed burst potentiation and/or long-term potentiation and cholinergic function (101,173, 174, 175)
Antiglucocorticoid and/or cortisol-lowering activity (176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186)
Antioxidant effects; scavenges free radicals (29,187,188)
Protects against excitatory and corticosteroid neurotoxicity (189, 190, 191)
Antiapoptotic (or proapoptotic) effects (192,193)
Regulates neuronal and glial survival (dose-dependent effect) (189,191,193,194); but c.f. (195)
Increases production and release of amyloid precursor protein and secretion of nonamyloidogenic isoforms (which are neuroprotective) and decreases production and deposition of amyloid β protein (99,196)
Inhibits production of proinflammatory cytokines, IL-1α, IL-6, and TNF-α (77,197, 198, 199, 200)
Increases IGF-I levels and bioavailability (3,119,201,202)
Increases neurite length in developing rodent frontal cortex (203)

Numerals in parentheses after each mechanism refer to numbered references.

Important actions in the central nervous system were initially inferred from the fact that DHEA(S) are synthesized in situ in brain; indeed, they have been termed neurosteroids for this reason (17,31). In rats, formation or accumulation in the brain depends on in situ mechanisms unrelated to the peripheral endocrine gland system (19, 20, 21). It is unknown if this is also the case in humans. It could be different, since rodents, unlike humans, do not produce appreciable amounts of DHEA(S) in the periphery (4,36, 37, 38). The synthetic pathway of DHEA(S) and its relationship to other steroid hormones is detailed in Figure 6.1.

Figure 6.1. (Opposite page) • Biosynthetic Pathway of Dehydroepiandrosterone (DHEA) and DHEA-sulfate (DHEA-S). P450scc, mitochondrial cholesterol side-chain cleavage enzyme, mediates 20 hydroxylation, 22 hydroxylation, and scission of the c20-22 bond. 3β-HSD mediates both 3β hydroxysteroid dehydrogenase and Δ5-Δ4 isomerase activities; P450c11β, mitochondrial 11 hydroxylase, mediates 11 hydroxylation; P450c11AS, mitochondrial aldosterone synthase, mediates c11,18 hydroxylation and 18 oxidation. P450c21 mediates 21 hydroxylation of progesterone or 17 OH progesterone—this is the only steroidogenic P450 to date whose protein or mRNA has not been detected in the brain; 17βHSD, mediates c17β reduction or c17 oxidation by several different isoforms of the enzyme; 5α-reductase-reductase (types I and II) mediate the conversion of progesterone to dihydroprogesterone, 11 deoxycorticosterone (DOC) to 5-α-dihydroDOC and testosterone to dihydrotestosterone (DHT). 3-α-HSD mediates the conversion of dihydroprogesterone to allopregnanolone, of 5-α-dihydroDOC to allotetrahydroDOC (not shown), or of dihydrotestosterone to androstanediol (not shown). P450aro mediates the conversion of testosterone to estradiol or of androstenedione to estrone. Interconversion of free steroids such as pregnenolone and DHEA to their sulfated derivatives, mediated by sulfotransferase and sulfatase, are also shown. Certain neurosteroids synthesized in these pathways are generally GABA-A receptor agonists [e.g., allopregnanolone, tetrahydrodeoxycorticosterone (THDOC), androstanediol, androsterone] and others are generally antagonists or inverse agonists [e.g., pregnenolone, pregnenolone sulfate (PS), dehydroepiandrosterone (DHEA), DHEA sulfate (DHEA-S)].

In the DHEA clinical research literature, the question frequently arises with regard to the relevance of serum DHEA(S) levels for estimating brain DHEA(S) content. In one of the only studies to examine this, Guazzo et al. (39) found that cerebrospinal fluid (CSF) and serum levels of DHEA (r=0.65) and DHEA-S (r=0.88) were significantly correlated. They also found that glucocorticoid treatment, which suppresses the hypothalamus-pituitary-adrenal axis, lowered serum cortisol and DHEA, as well as CSF DHEA, but this treatment did not lower CSF cortisol or DHEA-S (39).

Endogenous Dehydroepiandrosterone and DHEA Sulfate Levels Correlate with Mood, Memory, and Functional Abilities in Humans

Correlational studies have provided indirect evidence of DHEA’s effects on mood, memory, and functional abilities in humans, but numerous caveats, outlined more fully elsewhere (24), are important to consider before ascribing causality in these relationships.

DEPRESSION

In one of the largest cross-sectional population-based studies, Barrett-Connor et al. (40) assessed depression ratings in relation to plasma levels of several steroid hormones (total and bioavailable estradiol and testosterone, estrone, androstenedione, cortisol, DHEA, and DHEA-S) in 699 nonestrogen using, community dwelling, postmenopausal women (aged 50 to 90 years). They found that only DHEA-S levels were significantly and inversely correlated with ratings of depressed mood; this association was independent of age, physical activity, and weight change. Further, women with categorical diagnoses of depression had significantly lower DHEA-S levels compared to age-matched nondepressed women (40). Similarly, Scott et al. (41), found low DHEA-S, but normal DHEA, levels in depressed patients. Schmidt et al. (42) also found that women with the first onset of major or minor depression during perimenopause showed low morning plasma DHEA and DHEA-S (but not cortisol)
levels. Dysthymic patients have also been shown to have low DHEA-S levels (43). In another study, partially or completely remitted depressed patients had DHEA levels intermediate between currently depressed patients and controls, and, in the currently depressed patients, morning DHEA levels were inversely related to depression ratings, suggesting a partial normalization with clinical recovery (44). Low DHEA levels have also been reported in child and adolescent patients with depression (45).

The remaining literature examining serum, salivary, or urinary DHEA(S) levels in depression is inconsistent, with some reports of increased (46, 47, 48, 49), decreased (50), or unaltered (51, 52, 53, 54, 55) levels. Assies et al. (56), examined diurnal salivary levels of DHEA-S and cortisol in a small group of medicated but still depressed patients with unipolar depression, and found that patients, compared to controls, had elevated DHEA-S levels but normal cortisol levels. In a particularly detailed and well-conducted study, Heuser et al. (48) studied 24-hour plasma DHEA levels (every 30 minutes) in severely depressed patients and healthy controls. They found that depression was significantly associated with increased diurnal minimum and mean DHEA plasma levels, but there was no change in the diurnal maximum plasma levels and the diurnal amplitude of DHEA. They also found that the elevations in plasma DHEA levels paralleled elevations in plasma cortisol in that sample. In another study, elevations in DHEA-S levels in depressed patients were positively correlated with depression severity ratings (46). In contrast to the findings of Michael et al. (44), cited above, others have reported that elevated DHEA-S levels in depression decrease with remission (57). Several groups have found that DHEA-to-cortisol ratios more accurately discriminate depressed from nondepressed individuals (44,58,59), with lower morning ratios seen in the depressed individuals (44,58).

It is apparent that the relationship between DHEA(S) and depression is complex, with reports both of significant positive and negative relationships between DHEA(S) levels and depression. There is, at present, no parsimonious way of reconciling the diverse findings, but certain demographic variables, such as gender, age, age at onset, comorbid psychiatric and medical diagnoses, acuteness versus chronicity of stress, medication status, as well as timing of the sample collection, seem relevant.

SCHIZOPHRENIA

In some reports, patients with schizophrenia had low serum levels of DHEA (60, 61, 62, 63), but DHEA-S levels were elevated (64). In a small study comparing 13 acutely exacerbated paranoid schizophrenics with matched controls, low DHEA and high DHEA-S levels were observed in the schizophrenic group, but these differences were not statistically significant (65). However, in general agreement with a “diagnostic cut-off” of DHEA (<470 ng per dL), as proposed by Erb et al. (63), Brophy et al. (65) determined the mean serum DHEA level in their sample of schizophrenic patients to be 405 ng per dL compared to 506 ng per dL in their sample of controls. In contrast, a more recent study in medicated male schizophrenic patients found elevated serum levels of DHEA but decreased levels of DHEA-S (66). In another study, Harris et al. (67) noted that morning serum DHEA levels and/or DHEA-to-cortisol ratios were directly correlated with aspects of memory performance and were inversely correlated with ratings of psychosis and parkinsonian movements in chronic, medicated, institutionalized schizophrenics. These findings cumulatively raise the possibility that low DHEA levels (or low DHEA-to-cortisol ratios) identify a particularly impaired subgroup of chronic schizophrenic patients.

Only gold members can continue reading. Log In or Register to continue