| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Neurology, Max Planck Institute of Psychiatry, 80804 Munich, Germany
Address all correspondence and requests for reprints to: Dr. Florian Then Bergh, National Institute of Neurological Diseases and Stroke, Laboratory of Molecular Biology, National Institutes of Health, 36 Convent Drive, Room 3C11, Bethesda, Maryland 20892-4092. E-mail: thenberf{at}ninds.nih.gov
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Twenty-one patients with definite relapsing-remitting MS (11 females, aged 33.9 ± 2.0 yr; Expanded Disability Status Scale score of neurological impairment, 2.0–6.5) in acute relapse were treated with placebo (n = 13) or 300 mg moclobemide (reversible monoamine oxidase A inhibitor; n = 8) for 75 days. All received oral fluocortolone from day 7 on, and the dose was tapered until day 29. Effects were evaluated using the combined dexamethasone-CRH test and clinically on days 1, 30, and 75.
At baseline, the HPA axis was mildly activated, comparably for treatment groups [area under the curve for cortisol (AUC-Cort), 213.8 ± 76.8 arbitrary units in the moclobemide group vs. 225.8 ± 65.1 in the steroid alone group; mean ± SEM]. In a group of healthy controls with comparable demographic characteristics, the AUC-Cort was 107.4 ± 14.1. Moclobemide cotreatment resulted in normalization of the HPA axis response, whereas the HPA system hyperresponse was maintained with steroids alone (AUC-Cort on day 30, 85.9 ± 22.8 vs.177.1 ± 68.5; on day 75, 111.0 ± 46.0 vs. 199.2 ± 64.6). The change in Expanded Disability Status Scale was comparable for both groups.
Although corticosteroids alone had no effect on the HPA response using the dexamethasone-CRH test, treatment with moclobemide combined with corticosteroids favors normalization of the HPA response in relapsing-remitting MS.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Although earlier reports on the regulation of the HPA system in MS yielded variable results (7, 8, 9), recent literature more consistently shows hyperactivity of the HPA axis under basal conditions (10), in dynamic testing (10, 11, 12, 13, 14), or postmortem (15, 16, 17). Using the combined dexamethasone-CRH test (Dex-CRH test), we previously showed hyperactivity of the HPA system in MS patients, which was significantly correlated to the clinical course of MS (12, 18). As yet, however, little is known about the time course of HPA axis activity and its relation to therapeutic interventions. It is conceivable, for example, that chronic hypersecretion of cortisol leads to a desensitization of immune cells toward the effects of corticosteroids, making steroid medication for acute relapse less effective. In the rat, susceptibility to experimental allergic encephalomyelitis, an animal model of MS, is related to the immunogenetic background and the antigen used for immunization (19). In addition, HPA system activity has also been shown to influence the course of experimental allergic encephalomyelitis (20, 21, 22, 23), although its role in determining disease susceptibility in different rat strains is controversial (24).
Neuroendocrine feedback has also been studied in patients with affective diseases in the search for the neurobiological basis of mood disorders. Hyperactivity of the HPA system is one of the most consistent findings in major depression (reviewed in Refs. 25 and 26). The Dex-CRH test in depressed patients yields results similar to those in MS (27, 28, 29). Interestingly, successful treatment with antidepressants is associated with reduction or normalization of HPA hyperactivity (28, 30, 31). Animal experiments have shown that antidepressants increase glucocorticoid receptor messenger ribonucleic acid in the hypothalamus (32), and that amitriptyline (33) and the reversible inhibitor of monoamine oxidase A, moclobemide (34), both increase the amount of glucocorticoid receptors in the rat hypothalamus. One of the mechanisms of action of antidepressants may therefore be the restoration of the disturbed feedback regulation of the HPA system.
We hypothesize that due to HPA axis activation and accompanying subclinical hypercortisolism, the target tissues (notably cells of the immune system) may adapt to steroid effects and be less susceptible to steroid medication in MS patients. Restoring normal HPA axis function by comedication (with an antidepressant) may therefore improve the therapeutic effect of such medication. In the first step we wanted to determine whether HPA system dysregulation was fixed or could be modified and to assess the tolerability of combined treatment. We therefore conducted a double blind, placebo-controlled trial to evaluate the effect of cotreatment with the antidepressant moclobemide as an adjunct to oral corticosteroids in MS.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
As the comedication we chose moclobemide over amitriptyline, because the latter has anticholinergic side-effects, adversely affecting bladder function, which is often impaired in MS. Patients were treated with moclobemide or placebo for 6 days before steroids were added; the interval was introduced to allow for increased synthesis of glucocorticoid receptors, according to previous data from animal experiments (34).
The trial included patients with relapsing-remitting or chronic
progressive MS. In this report we present data for relapsing-remitting
MS only; the study is ongoing in chronic progressive patients (see
below). The trial was designed as a double blind, placebo-controlled,
clinical trial. It was approved by the ethics committee of the Max
Planck Institute of Psychiatry and was conducted according to the
Declaration of Helsinki. For an overview of the trial design, see Fig. 1
.
|
Upon inclusion in the trial, patients were examined clinically [day 0; including scoring neurological disability according to Kurtzke’s Expanded Disability Status Scale (EDSS) (37)], and depressive symptoms were scored using the Hamilton depression (HAMD) score (38). Neuroendocrinological assessment was performed using the combined Dex-CRH test on day 1 (see below).
From day 2 until the end of the trial, patients received either 300 mg moclobemide daily or placebo in identical capsules. All patients received oral fluocortolone starting on day 7 at 100 mg daily; the dose was tapered over 22 days (100, 80, 60, and 40 mg for 3 days each, and 20 and 10 mg for 5 days each, respectively), together with ranitidine (300 mg daily).
On day 30, 1 day after the last dose of fluocortolone (10 mg), the clinical evaluation and Dex-CRH test were repeated. Patients continued to take the study medication (moclobemide or placebo) and were again examined clinically and by the Dex-CRH test on day 75, i.e. 6 weeks after cessation of steroid treatment.
Subjects
Figure 2 gives an overview of the
method of subject selection. Screening for this trial included all
patients admitted for corticosteroid treatment (n = 138). Of
these, 98 were excluded because they met exclusion criteria; in most
cases, corticosteroid treatment could not be delayed according to study
protocol on clinical grounds, or prior antidepressant medication could
not be excluded with sufficient certainty. The remaining 40 patients
(who had relapsing-remitting or chronic-progressive MS) were
consecutively assigned to the study medication. This procedure followed
a randomized, blinded treatment assignment, which had been provided
before the start of the trial by the institute’s biometrician.
|
gives the demographic and
clinical characteristics of the patients according to treatment
arm.
|
gives demographic characteristics of the
control subjects. Neuroendocrinological assessment
The Dex-CRH test was performed as previously described (12). Patients were pretreated with 1.5 mg dexamethasone, orally, at 2300 h the night before the test. On the day of the test, an iv cannula was inserted at 1430 h and kept patent by normal saline infusion. Blood was taken at 15-min intervals between 1500 and 1630 h for determination of plasma concentrations of cortisol and ACTH. At 1502 h, 100 µg synthetic human CRH (CLINALFA, Läufelingen, Switzerland) was injected as an iv bolus.
Determination of plasma hormone concentrations
Blood was drawn into prechilled tubes containing ethylenediamine tetraacetate and Trasylol and centrifuged, the plasma was taken off, frozen, and stored at -80 C until measurement. Cortisol and ACTH concentrations were determined using commercial RIAs (ImmuChem Cortisol, ICN Biomedicals, Inc., Costa Mesa, CA; RIA-ACTH, Nichols Institute Diagnostics, San Juan Capistrano, CA), with an interassay coefficient of variation of less than 8% and an intraassay coefficient of variation of less than 4%.
Data analysis
For statistical analysis, the following curve indicators of the
plasma hormone concentrations in the Dex-CRH test were used: maximum
concentration (Max-Cort and Max-ACTH), difference between baseline and
maximum after human CRH injection (
-Cort and -ACTH), and the area
under the time course curve according to the trapezoid rule (AUC-Cort
and AUC-ACTH). In an exploratory manner, the EDSS score was analyzed as
an indicator of the clinical effect of the treatments.
Group differences with respect to the clinical and neuroendocrine
parameters were statistically tested for significance by two-factorial
multivariate analysis of covariance. Group was a between-subjects
factor with three levels (MS patients with placebo, MS patients with
moclobemide, and controls), time was a within-subject factor with three
levels (baseline, day 30, and day 75), and age and sex were the
covariates. If significant main or interaction effects were found,
univariate F tests were performed to identify the variables with
significant contributions to these effects. For these variables, tests
with contrasts were subsequently carried out to locate the group or
time pairs with significant differences. Values were also compared with
the results obtained previously in 27 healthy controls. As the nominal
level of significance, 
= 0.05 was accepted. A
posteriori tests (univariate F tests and tests with contrast) were
performed at a reduced level of significance (adjusted according to
Bonferroni procedure) to keep the type I error less than or equal to
0.05. Data are given as the mean ± SEM
unless stated otherwise.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The neuroendocrinological assessment at baseline revealed exaggerated
cortisol secretion in the combined Dex-CRH test in both treatment
groups (Fig. 3, left, Day 1)
compared with healthy controls. Both groups thus displayed
dysregulation of the hypothalamo-pituitary-adrenal system, with a
comparable degree of dysregulation. The mean morning cortisol plasma
concentration was normal in all subjects.
|
Neuroendocrinological follow-up in the group treated with placebo and
fluocortolone revealed that the time course of the plasma cortisol
concentration changed little on day 30 (1 day after the end of steroid
medication; Fig. 3, center, Day 30) as well as on day 75.
Corticosteroids alone thus did not substantially alter HPA system
dysregulation. In contrast, when fluocortolone was combined with
moclobemide, cortisol secretion in the Dex-CRH test on day 30 was lower
than in the reference population and was indistinguishable from that in
controls on day 75 (Fig. 3, right, Day 75). Combined
moclobemide and fluocortolone thus led to normalization of HPA
axis activity. The neuroendocrine indicators are listed in Table 2. It was further found that the AUC-Cort
was significantly different from that in the reference population in
both patient groups at baseline. At follow-up, AUC-Cort in the
moclobemide group was not different from that in controls at both
assessments. In the placebo group, a trend (P = 0.067)
toward HPA system hyperactivity was retained at the final visit.
|
The therapeutic effect of corticosteroid treatment on neurological impairment was similar in both groups, with mean reductions of the EDSS on day 30 of 1.2 ± 0.3 points (moclobemide group) and 1.2 ± 0.3 (placebo group) and comparable further improvement until day 75. The HAMD score was more markedly reduced in the moclobemide group; however, the mean change was small, as expected from the generally low scores at baseline. The one patient with a HAMD score of 19 received moclobemide and improved to a HAMD score of 10.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Time course of HPA axis dysregulation in multiple sclerosis
Dysregulation of the HPA system in MS patients has been demonstrated in a number of studies (10, 11, 12, 13, 14, 18). Among the tests employed to date, the combined Dex-CRH test was especially sensitive at detecting HPA axis hyperactivity (18), and we therefore chose it as the main measure for neuroendocrinological assessment in this trial. The mild degree of cortisol hypersecretion was comparable to the range we (12, 18) and others (14) had observed in relapsing-remitting MS in earlier investigations. We believe that it reflects a disturbance of negative feedback at the level of the hypothalamus or pituitary, where the administered dexamethasone suppresses HPA system activity, mediated via the glucocorticoid, and partly the mineralocorticoid receptor. An abnormally low number or altered function of these receptors could explain this phenomenon.
The origin of the HPA axis dysregulation in MS is currently unclear, and studying its time course may contribute to understanding this phenomenon. Among functional disturbances, activation of hormone secretion, for example by cytokines or inflammatory mediators may be involved (5, 6), which would be consistent with a normalization upon treatment with antiinflammatory agents. Alternatively, it may merely reflect anatomical damage to the central nervous system, as it was found to correlate with the degree of neurological impairment (18). In this latter case, HPA axis dysregulation would be expected to hardly respond to treatment. Our present results argue in favor of a functional rather than a structural disturbance, as the response of the HPA system can be effectively manipulated depending on the choice of treatment.
The steroid regimen we used did not produce substantial suppression of the HPA axis, as assessed by the Dex-CRH test 1 day after the end of treatment. This appears contradictory to the usually observed pattern of HPA system down-regulation upon exogenous steroid application. The slow tapering from a 100-mg to a 10-mg daily dose, and the 1-day interval before the Dex-CRH test probably allowed for sufficient recovery of HPA system reactivity. There are three previous studies on the activity of the HPA axis in the context of steroid treatment for MS relapse (39, 40, 41). All of these studies focused on concerns about endocrinological safety and were therefore designed to primarily detect possible subnormal HPA system activity. Miro et al. (39) performed ACTH stimulation and metyrapone tests immediately after cessation of therapy. They showed that oral corticosteroids (prednisone, tapered from 1 mg/kg) did not result in gross suppression of cortisol secretion and concluded that they were endocrinologically safe. No baseline values are reported. Wenning et al. (40) and Levic et al. (41) investigated patients treated with iv methylprednisolone (500 mg daily for 5 days and 1000 mg daily for 7 days, respectively), and their results are therefore not directly comparable to ours. These studies detected subnormal basal cortisol levels up to 5 days after cessation of treatment (40), but only transiently reduced ACTH and cortisol responses to iv CRH (40) or insulin-induced hypoglycemia (41) for 1–3 days. These studies do not comment on the phenomenon of overactivity of the HPA system, and the follow-up period was at most 3 weeks after steroid treatment (41). It is therefore difficult to compare these trials with our present work. Most importantly, however, oral treatment did not suppress HPA system responsiveness (39), as was the case in our series.
Modification of HPA system activity and possible clinical effects
Essentially normal reactivity of the HPA axis was achieved by a combination of steroids and moclobemide. As steroids alone had no comparable effect, the influence on the HPA system is largely attributable to moclobemide. The most probable mechanism of this influence is an increase in glucocorticoid receptor (GR) content in the hypothalamus, thus restoring the usual level of HPA suppression by exogenous dexamethasone. Peiffer et al. demonstrated that antidepressants do increase the amount of GR messenger ribonucleic acid in the hypothalamus in healthy rats fed amitriptyline (32). Biochemical and pharmacological studies confirmed the increase in GR protein, and behavioral testing indicated an attenuation of HPA axis reactivity to established stressors with both moclobemide (34) and amitriptyline (33). A higher number of GRs could explain how the feedback inhibition by dexamethasone is improved, reflected by diminished cortisol secretion compared with baseline. The more pronounced effect on lowering cortisol secretion compared with ACTH secretion indicates that the chronic stimulation of the adrenal glands is diminished, but may also be the result of additional peripheral effects. Another possible explanation might, of course, be that the physical and psychological stress associated with neurological impairment led to HPA activation, and that it resolved secondary to clinical improvement. However, this seems unlikely because the treatment groups had a very similar clinical course, but behaved differently with respect to neuroendocrine regulation. Still, the sample size is small, and the influence of unrecognized confounding effects cannot be excluded.
In the present patient group, the effects of moclobemide cotreatment on HPA system dysregulation were moderate. Statistically, a significant difference compared with controls was present at baseline, and HPA overactivity persisted until the final visit in patients receiving fluocortolone alone. With moclobemide cotreatment, no difference from controls was detected at follow-up assessments. Direct comparison of the two treatment groups did not, however, result in differences reaching statistical significance. We think that this failure to show significant group differences in the direct comparison is partly due to the small sample size. Further, this study comprises patients with relapsing-remitting MS only, the group with the mildest degree of HPA axis activation (18).
Although our study sample was too small to detect clinically relevant differences in the effects of the two treatments, this was not the primary goal of the study. The difference is expected to be subtle and would require larger numbers of patients to be detected. This is also evident from a recent study comparing different regimens of steroid treatment in MS (42). In addition, a possible clinical benefit of normalizing HPA axis responsiveness would probably be evident with long-term therapy only. The mean depression scores (HAMD) may seem unexpectedly low in a chronically ill population. Strict exclusion of patients who met criteria for major depression and attribution of somatic symptoms to MS rather than depression have probably contributed to these low scores.
Chronic hypersecretion of cortisol, as suggested by the known HPA axis hyperactivity in MS, leads to target tissue adaptation, making the tissues less sensitive to the effects of steroids, among others in the immune system. Moreover, recent evidence (43) suggests that in aged rats, hypercortisolism is a crucial factor limiting the proliferation of neural stem cells in the hippocampus, and that a reduction of corticosteroid levels restores normal formation of neurons even in adult mammals. Although this has not been prospectively studied in MS, the dysregulation of the HPA system may therefore contribute to the pathogenesis of the disease, and its normalization may be beneficial.
Our present data demonstrate that the HPA system dysregulation can be modified by pharmacological intervention. Further investigation is required to determine whether the observed endocrinological changes influence the clinical outcome in the long term.
Received July 10, 2000.
Revised October 10, 2000.
Revised December 15, 2000.
Accepted December 28, 2000.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C Heesen, D C Mohr, I Huitinga, F T. Bergh, J Gaab, C Otte, and S M Gold Stress regulation in multiple sclerosis-current issues and concepts Multiple Sclerosis, March 1, 2007; 13(2): 143 - 148. [Abstract] [PDF]
|
|
|
|
 |
|
 L. M. L. van Winsen, D. F. R. Muris, C. H. Polman, C. D. Dijkstra, T. K. van den Berg, and B. M. J. Uitdehaag Sensitivity to Glucocorticoids Is Decreased in Relapsing Remitting Multiple Sclerosis J. Clin. Endocrinol. Metab., February 1, 2005; 90(2): 734 - 740. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Gottschalk, T. Kumpfel, P. Flachenecker, M. Uhr, C. Trenkwalder, F. Holsboer, and F. Weber Fatigue and Regulation of the Hypothalamo-Pituitary-Adrenal Axis in Multiple Sclerosis Arch Neurol, February 1, 2005; 62(2): 277 - 280. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. HARBUZ, A. J. CHOVER-GONZALEZ, and D. S. JESSOP Hypothalamo-Pituitary-Adrenal Axis and Chronic Immune Activation Ann. N.Y. Acad. Sci., May 1, 2003; 992(1): 99 - 106. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
). 
, The cortisol response in 27 healthy controls
studied previously.