| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Internal Medicine, Division of Endocrinology and Metabolism, University of Michigan (E.V.D., C.A.J.), and Department of Veterans Affairs Medical Center (C.A.J.), Ann Arbor, Michigan 48109
Address all correspondence and requests for reprints to: Craig A. Jaffe, M.D., 3920 Taubman Center, Box 0354, Ann Arbor, Michigan 48109. E-mail: cjaffe{at}umich.edu.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Dexamethasone is primarily metabolized by hepatic CYP3A4 (9), an enzyme complex that is responsible for the metabolism of many xenobiotics (10, 11). Multiple xenobiotics are known to modify CYP3A4 activity (10, 11). In the 1960s it was observed that the antiepileptic diphenylhydantoin resulted in false positive DST (3, 8). Subsequently, several medications, including antiepileptics phenobarbital, primidone, ethosuximide, carbamazepine, and rifampin, have been shown to have a similar effect on DST by inducing the activity of CYP3A4 (4, 12, 13, 14, 15).
The FDA approved the antidiabetic agent troglitazone for use in patients with type 2 diabetes mellitus in 1997. It was eventually withdrawn from the market in the United States in March 2000 because of liver toxicity. Many of the patients who were receiving troglitazone were obese and also had hypertension, a constellation of findings that raises the possibility of Cushings syndrome. In addition, troglitazone was studied in clinical trials for the treatment of polycystic ovary syndrome, another condition that frequently needs to be differentiated from Cushings syndrome (16, 17).
In 1998 we evaluated a patient with the diagnosis of polycystic ovary syndrome who was participating in a clinical trial on the use of troglitazone for the treatment of this condition. She had clinical signs suggestive of Cushings syndrome. Twice she underwent a 1-mg overnight dexamethasone suppression test, and both times she failed to suppress her morning cortisol level. However, the 24-h urinary free cortisol level was within the normal range. We suspected that the DST was falsely abnormal because of induction of CYP3A4 by troglitazone. At that time, there was indirect evidence of CYP3A4 induction by troglitazone (Rezulin package insert) (17, 18, 19, 20, 21). As troglitazone was used in a patient population very likely to be screened for Cushings syndrome, we decided to study the effects of troglitazone on CYP3A4 systematically. We tested the effect of troglitazone on the activity of CYP3A4 and the results of the DST in eight healthy subjects. At the same time we have reexamined the value of alternative suppression tests for the screening for Cushings syndrome.
| Subjects and Methods |
|---|
|
|
|---|
The study was approved by the Institutional Review Board and the General Clinical Research Center of University of Michigan. We studied eight healthy subjects, five women and three men (mean age, 25 ± 2 yr; mean body mass index, 23 ± 2 kg/m2). All subjects had normal screening laboratory tests, including liver function tests. The study was conducted from October 1998 until February 1999.
Study design
The subjects were studied twice, first at baseline and then in an identical protocol after treatment with troglitazone (600 mg daily) for 28 d. Troglitazone was continued while they completed the protocol, and compliance was assessed by counting the number of returned tablets at the end of the study. After an overnight fast, baseline serum cortisol, corticosterone, and 11-deoxycortisol were measured at 0800 h on d 1, and a [14C]erythromycin breath test (ERMBT) was performed at 0800 h on d 1 for the assessment of CYP3A4 activity (22, 23). An overnight hydrocortisone suppression test was performed by administering a single dose of hydrocortisone (50 mg) at 1100 h and measuring serum corticosterone, 11-deoxycortisol, and cortisol at 0800 h on d 2 as previously described (14). After a 1-d washout period, a DST was performed. Dexamethasone (1 mg) was taken at 1100 h on d 3, and serum cortisol and dexamethasone were measured on d 4.
Assays
Serum cortisol was measured by enzyme immunoassay (Diagnostic Systems Laboratories, Inc., Webster, TX). Serum corticosterone was measured by RIA after multiple solvent extractions, and serum 11-deoxycortisol and dexamethasone were measured by RIA after chromatography by Esoterix (Calabasas Hills, CA).
Data analysis
The normal response to DST was defined as an 0800 h serum cortisol suppression below 1.8 µg/dl (1). The normal corticosterone response to hydrocortisone was defined as a more that 50% suppression comparing to baseline, as previously described (14). The specificity of each test was calculated as the ratio of the number of subjects with normal response to the total number of subjects. Paired t tests were used to compare dexamethasone levels and the ERMBT results before and during troglitazone administration. Results are expressed as the mean ± SE.
| Results |
|---|
|
|
|---|
Baseline 0800 h serum cortisol was similar before and during troglitazone treatment (P = 0.66). The same was true for baseline serum corticosterone (P = 0.40), and 11-deoxycortisol (P = 1.0).
In response to the 1-mg overnight DST, all subjects had 0800 h cortisol below 1.8 µg/dl (mean, 0.66 ± 0.08 µg/dl) before troglitazone treatment. During troglitazone treatment, 0800 h cortisol level after dexamethasone was above 1.8 µg/dl in all subjects (mean, 9.0 ± 1.8 µg/dl). Therefore, when the criterion of 0800 h cortisol suppression to below 1.8 µg/dl was used, the specificity of the DST in ruling out Cushings syndrome was 100% before treatment, but 0% during troglitazone treatment.
|
|
Before treatment, after 50 mg hydrocortisone, 0800 h serum corticosterone levels decreased by more than 50% in all subjects (mean, -87 ± 2%). The corticosterone response to HST was similar during troglitazone treatment (-85 ± 5%; P = 0.91). Therefore, the specificity of the HST, using more than 50% suppression of 0800 h corticosterone as a criterion of normalcy, was 100% both before and during troglitazone treatment.
|
After hydrocortisone administration, before troglitazone treatment, 0800 h serum cortisol levels decreased below 5 µg/dl in five of seven subjects and by more than 50% of baseline in six of seven subjects, giving a specificity of 86%. During troglitazone treatment, 0800 h cortisol level decreased below 5 µg/dl in five of eight subjects and to below 50% of baseline in six of eight subjects, with a specificity of only 75%.
Erythromycin breath test (Fig. 4
)
The activity of CYP3A4, as measured by the percentage of [14C]erythromycin exhaled at 20 min after radiolabeled erythromycin injection, increased during the troglitazone treatment by 27 ± 9% (P = 0.022).
|
| Discussion |
|---|
|
|
|---|
CYP3A4 is the most important drug-metabolizing P450 and is affected by a wide variety of commonly used medications. Induction of this enzyme leads to accelerated metabolism of a large number of xenobiotics, including cyclosporine, oral contraceptives, antibiotics, HMG-CoA reductase inhibitors, and corticosteroids, particularly dexamethasone. Induction of CYP3A4 results in false positive DST, because of accelerated metabolism of dexamethasone. In the case of troglitazone, the effects on the results of the DST were of particular importance because there is frequently a need to rule out Cushings syndrome in patients most likely to be taking troglitazone, such as obese patients with diabetes mellitus or obese women with polycystic ovary syndrome.
At the time this study was conducted there were only indirect data suggesting that troglitazone induced the activity of CYP3A4. The manufacturer reported decreased levels of terfenadine, ethinyl estradiol, and norethindrone when administered together with troglitazone (17). Decreased serum cyclosporine level in a patient receiving troglitazone had also been reported (19, 20). In our study, we used the ERMBT to directly assess the effects of troglitazone on CYP3A4 activity (22).
This test has been extensively validated and shown to reliably quantify the activity of CYP3A4 (23, 24, 25, 26, 27, 28). Although there is a large intersubject variability in the results of the ERMBT, the results for a given subject are remarkably consistent (22). Treatment with glucocorticoids or rifampicin increased the ERMBT values by 55% and 120%, respectively (22). In our subjects troglitazone increased CYP3A4 activity by 27%, verifying that the reported drug interactions of troglitazone were due to induction of CYP3A4 activity. Although this degree of induction was less than those reported for several other xenobiotics (20), it resulted in a clinically significant effect on dexamethasone. The serum dexamethasone levels 9 h after dexamethasone administration decreased by 66%, and false positive DST occurred in all subjects. Our findings are in agreement with a recent study that showed that the area under the curve of dexamethasone for 24 h after dexamethasone administration was decreased by 49% in subjects receiving troglitazone (29).
Our data are consistent with the effects of several other xenobiotics. Diphenylhydantoin, a well known inducer of CYP3A4, decreased the half-life of dexamethasone by 51% (2). Moreover, diphenylhydantoin resulted in abnormal 2-d, 2-mg DST or 1-mg, overnight DST (3, 14). Similarly, administration of rifampin resulted in low plasma dexamethasone concentrations and false positive 1-mg overnight DST (6).
Since our study was designed, there have been additional reports of clinically important effects of troglitazone on the metabolism of simvastatin, atorvostatin, and dexamethasone (29, 30, 31, 32). More direct evidence of troglitazone-induced CYP3A4 activity comes from in vitro studies using primary cell cultures of human hepatocytes (33). Troglitazone increased immunoreactive CYP3A4 protein, CYP3A4 mRNA, and testosterone 6ß-hydroxylase activity in a dose-dependent fashion (34). The in vitro ability of troglitazone to induce CYP3A4 was intermediate between those of rifampin and dexamethasone (34).
Our study also allowed us to investigate an alternative to the DST. This is of practical importance because many commonly used xenobiotics affect the activity of CYP3A4. In patients suspected of having a false positive DST result, 24-h urine collection for UFC would be an acceptable alternative. The sensitivity and specificity of this test are excellent (35). However, the use of 24-h UFC can be misleading in patients with episodic Cushings syndrome (36) as well as in patients with renal insufficiency (37). Late night salivary cortisol measurement is becoming a widely accepted and practical alternative and might eventually become the test of choice in screening for Cushings syndrome (33). However, the availability of well validated, alternative suppression tests would be useful, especially in cases of mild Cushings syndrome where the combination of different diagnostic tests is required.
One alternative is the HST, which was first described by Meikle et al. (14) in 1974. The HST takes advantage of the fact that the metabolism of hydrocortisone is minimally affected by the induction of CYP3A4 (2). The protocol tests the ability of hydrocortisone instead of dexamethasone to suppress pituitary ACTH secretion. An intermediate product of the steroid biosynthesis pathway that is ACTH dependant and can be measured without cross-reaction with hydrocortisone, such as corticosterone, is used as an indicator of ACTH suppression. Suppression of 0800 h plasma corticosterone to less than 50% of the baseline after receiving 50 mg hydrocortisone 9 h previously was considered normal.
In the first report of HST, the specificity of the test in patients treated with antiepileptic medications was excellent (14). Only 1 of 15 patients who was taking 3 different agents failed to suppress plasma corticosterone levels below 300 ng/dl. The HST was also performed in 3 patients with known Cushings syndrome, and all three failed to suppress serum corticosterone.
In our study, we have further explored the HST by measuring 11-deoxycortisol and serum cortisol levels. It was expected that serum 11-deoxycortisol, which is the immediate precursor of cortisol, would decrease after the administration of hydrocortisone. This was confirmed by our data. However, the use of 11-deoxycortisol did not offer any advantage over serum corticosterone. Interestingly, because of the short plasma half-life of hydrocortisone, we were able to detect a suppression of endogenous cortisol after hydrocortisone administration. However, the specificity of the HST was low when serum cortisol levels were used as the indicator of ACTH suppression.
Unlike troglitazone, the newer thiazolidinediones that are currently available for clinical use do not appear to significantly affect the activity of CYP3A4. In one clinical study rosiglitazone resulted in a small, not clinically significant, decrease in nifedipine concentrations, suggesting that rosiglitazone is a very weak inducer of CYP3A4 (38). In another study, rosiglitazone did not affect the metabolism of ethinyl estradiol and norethindrone (39). Pioglitazone had no effect on the area under the curve of dexamethasone (29). In the same study, troglitazone, but not pioglitazone, decreased the dexamethasone-induced hyperglycemia and hyperinsulinemia. Similarly, troglitazone decreased the plasma concentration of simvastatin, but pioglitazone had no effect (31). Therefore, the currently available thiazolidinediones would be unlikely to affect the results of DST.
In conclusion, we have shown that troglitazone specifically induces the activity of CYP3A4, resulting in accelerated metabolism of dexamethasone and a false positive 1-mg overnight DST. This effect is similar to that of antiepileptic medications and rifampin and gave us the opportunity to evaluate screening tests for Cushings syndrome. We have shown that the HST is a useful alternative to the DST in the setting of medications that affect CYP3A4 activity. However, the sensitivity and specificity of this test in patients with high clinical suspicion of Cushings syndrome remain to be determined.
| Footnotes |
|---|
Abbreviations: DST, Dexamethasone suppression test; ERMBT, [14C]erythromycin breath test; HST, hydrocortisone suppression test.
Received November 18, 2002.
Accepted April 1, 2003.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Collino, N. S.A. Patel, and C. Thiemermann Review: PPARs as new therapeutic targets for the treatment of cerebral ischemia/reperfusion injury Therapeutic Advances in Cardiovascular Disease, June 1, 2008; 2(3): 179 - 197. [Abstract] [PDF] |
||||
![]() |
M. A. Peraza, A. D. Burdick, H. E. Marin, F. J. Gonzalez, and J. M. Peters The Toxicology of Ligands for Peroxisome Proliferator-Activated Receptors (PPAR) Toxicol. Sci., April 1, 2006; 90(2): 269 - 295. [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 |