Journal of Analytical Toxicology,Vol. 22, October 1998

The Effect of Oral Dehydroepiandrosterone(DHEA) on the Urine Testosterone/Epitestosterone(T/E) Ratio in Human Male Volunteers* Thomas Z. Bosy1, Karla A. Moore 1,t, andAlphonsePoklis2 IArmed Forces Instituteof Pathology, Division of Forensic Toxicology, 1413 ResearchBoulevard, Building 102, Rockville, Maryland 20850-3125 and 2Departmentof Pathology,Medical College of Virginia, MCV Station Box 980165, Richmond, Virginia 23298-0165

Abstrad Dehydroepiandrosterone (DHEA) is an endogenous androgenic steroid produced by the ovaries and adrenal glands. Research suggests that DHEA can be converted to testosterone in peripheral tissues. Classified as a nutritional supplement, this compound may be purchased without a prescription. The military and international sports organizations prohibit the use of exogenous androgenic/anabolic steroids. Steroid-screening results are considered "positive" when the urinary ratio of testosterone to epitestosterone (T/E), an inactive synthetic byproduct, exceeds 6:1. Human volunteers ingested the recommended daily dose of 50 mg each morning for 30 days to determine if DHEA causes an adverse effect on this ratio. Urinary samples were colleded before ingestion and 2-3 h after ingestion. Urine samples were extracted using solid-phase columns and analyzed using a previously developed gas chromatography-mass spectrometry method. T/E results were compared to an average baseline generated from three urine samples obtained before the study. Mean baseline T/E ratios averaged 0.67 for the seven subjects (range 0.1-1.2). The mean T/E ratio after DHEA ingestion ranged from 0.03 to 2.11. Individual postdose T/E ratios ranged from 0.01 to 3.7. The results from these individuals indicate that the administration of DHEA at this dose, for this period of time, has a minimal effect on urine T/E ratios and would not be expected to result in a positive screen for testosterone abuse. One subject agreed to take a single dose of 250 rag. This acute, high dose caused his T/E ratio to increase by 40% relative to the predose value.

Introdudion Controlling steroid abuse requires an effective means to determine the administration of banned substances. For synthetic anabolic androgenic steroids, the identification of the parent steroid and metabolites in urine is evidence that abuse has taken place (1,2). For substances that are produced naturally, like testosterone, the mere presence of the substance in the urine obviously cannot constitute proof of an offense. The notion of * The opinions or assertions contained herein are lhe privale views of the aulhors and are not to be construed as official or rejecting the views of the Department of the Air Force, the Department of the Army, the Department of the Navy, or the Department ol the Defense. Author to whom correspondence should be addressed.

setting a concentration cutoff for urinary testosterone was abandoned because the range of values in normal subjects varies widely. A more realistic approach was developed in the early 1980s (3,4): measuring the ratio of testosterone to epitestosterone, the inactive 17-(z epimer of testosterone produced as a by-product of testosterone synthesis. The T/E ratio is approximately 1 in the majority of normal males. Exogenous testosterone administration results in an increase in the testosterone/ epitestosterone (T/E) ratio because exogenous testosterone is not incorporated into the manufacture of epitestosterone. Additionally, exogenous administration of some of the other anabolic steroids will affect this ratio through a negative feedback mechanism, which decreases testosterone production without affecting epitestosterone production (5). In recognition of this situation, the International Olympic Committee (IOC) and many national and international sport authorities have worded their rules to state that testosterone administration is banned and that the T/E ratio may not exceed 6 (6,7). This is also the criterion currently accepted by the Department of Defense in determining steroid abuse cases. This study was conducted in response to questions received from military criminal investigation offices concerning dehydroepiandrosterone (DHEA) consumption and its relevance to the investigation of steroid-abuse cases. Use of selected anabolic, androgenic steroids is illegal for service members. DHEA is classifted as a nutritional supplement and as such is readily available to the general population. Recently, athletes have begun taking DHEA, theoretically hoping to derive some competitive benefit from its conversion to testosterone despite previous scientific evidence to the contrary. Moreover, individuals in a position to have their urine tested for steroids have begun claiming that elevated T/E ratios are the result of innocent DHEA supplementation. Current preliminary evidence indicates that DHEA may indeed result in T/E ratios approaching 6. The military and drugtesting communities have since stopped reporting the T/E ratio until these issues are resolved. This study will contribute to the body of knowledge on this relatively new supplement and assist in the litigation of cases in which this drug has been implicated as responsible for abnormal test results in anabolic steroid-abuse cases.

Reproduction(photocopying)of editorialcontentof thisjournalisprohibitedwithoutpublisher'spermission.

4.55

Journal of Analytical Toxicology, Vol. 22, October 1998

Experimental Materials DHEA capsules (50 rag) were obtained from The Vitamin Shoppe| (North Bergen, NJ). Testosterone, epitestosterone, 16-~r and DHEA standards were purchased from Sigma Chemical Co. (St. Louis, MO). N-MethyI-N-trimethylsilyltrifluoroacetamide (MSTFA) and ~-glucuronidase (from E. coli) were also purchased from Sigma Chemical Co. CLEAN SCREEN solid-phase extraction columns (ZSDAU020) were purchased from United Chemical Technologies, Inc. (Bristol, PA). All other chemicals were purchased from Sigma Chemical Co. Solvents were HPLC grade, purchased from Fischer Scientific Products/Fischer Scientific (Fair Lawn, New Jersey).

Subjects, dosing, and urine collection Seven healthy, male volunteers participated in the study. Demographic information is provided in Table I. The study was conducted under the guidelines and approval of the Institutional Review Board, Medical College of Virginia (Richmond, VA),and each volunteer gave informed consent. Participants submitted three urine samples taken on 3 separate days for baseline determination of their T/E ratios. The study protocol then instructed the participants to collect a urine sample each morning before administration of a 50-rag capsule of DHEAand a urine sample approximately 3-h post-dose (with no voiding after administration of the DHEAcapsule until the 3-h collection) for 30 consecutive days. Results of studies performed previously in this laboratory show peak urine levels of DHEAoccur approximately 2-h post-dose. This suggests that peak plasma levels occur some time before that, and the 3-h measure of urine T and E was chosen in an attempt to maximize the urine T generated from the dose of DHEA.To determine the effect of an acute high dose of DHEA, one subject agreed to take a single 250-rag dose. This single high dose was given to determine the changes such a dose would have on the T/E ratio measured over a 24-h period.

16-r was added (20 ng/mL) and the sample mixed and pH adjusted to 5.5-6.5 using concentrated sodium phosphate mono- or di-basic. The samples were then centrifuged at 3000 rpm for 5 rain. The sample supematant was poured over a C-8 solid-phase extraction column (Worldwide Monitoring Clean Screen ZCDAU020,United Chemical Technologies) conditioned with 3 mL of methanol, 3 mL of deionizedwater, and 3 mL of 100mMpH 6.0 phosphate buffer. The samples were added to the columns, allowed to flow via gravity, and washed with 3 mL deionized water. Vacuum was applied at > 10 mm of mercury for 10 rain to dry the samples, which were then eluted into clean tubes with 3 mL of methanol. Eluates were then dried under a stream of nitrogen after which 2 mL of 200mM phosphate buffer (pH 7.0) and 250 units of [~-glucuronidasewere added, vortex mixed, and allowed to incubate at 50~ for 1 h. Samples were then cooled, and the pH was adjusted to 10-11 using a 1:1 mixture of NaHCO3/Na2CO3.Five milliliters of n-butylchloride was then added to each sample. The tubes were capped and shaken vigorously for 10 rain and then centrifuged at 3000 rpm for 5 rain. The organic layer was transferred to clean tubes and dried under a stream of nitrogen. Afterthe samples were dry, the tubes were put into a desiccator and dried further under vacuum for 30 rain. The dried samples were then derivatized with 50 ~L of MSTFA/NH4I/dithioerythritol(1000:2:5,v/w/w)and incubated at 70~ for 20 rain. The samples were centrifuged at 3000 rpm for 1 rain and transferred directly to gas chromatography (GC) injection vials. One microliter was injected for gas chromatographic-mass spectrometric (GC-MS) analysis.

Instrumentation GC-MS analysis was performed using a 5890 GC (HewlettPackard, Palo Alto, CA) interfaced to an HP 5972 MS.The GC was equipped with a DB-1MS, fused-silica cross-linked methyl silicone capillary column (15 m • 0.25-ram i.d., 0.25-1Jmfilm-thickness, J&W Scientific, Folsorn, CA) with helium (1 mL/min) as the carrier gas. Splitless injections were performed with the GC oven temperature programmed at 150~ (held 1 rain) to 225~ at 2~ The final ramp was to 300~ at 50~ and held Extraction and derivatization for 3 min. The injector and MS temperature was 280~ The MS Urine samples were used for solid-phase extraction using a was operated in the selected ion monitoring (SIM) mode, and modified protocol from Donike et al. (8,9). Five milliliters of ions 432, 417, (T, E, DHEA), and 520 (internal standard) were urine was pipetted into 16 • 100 mm borosilicate glass test tubes. collected. A six-point standard curve was run with each assay. Quantitation was determined comparing the integrated peak areas of a sample with that Table I. Demographics of Subjects and Results of the Effect of DHEA on of the standard curve. The integrated peak area Baseline TE Ratios* of the 432 ion was used for all quantitations. Standard Concentrations of both T and E were varied to Weight Baseline Mean T/E T/E Range deviation generate the standard curve. Subject Age (Ibs.) T/E (Mean) ( D H E A ) (DHEA) 1 2 3 4 5 6 7

38 39 36 37 40 49 32

180 222 192 152 210 175 201

1,03 0,34 0.10 0,56 0.90 1,2 0.56

' Two-tailed t-VALUE ~= 0.025; df= 6)=3.143; Itl = 0.05.

456

1.10 0.43 0,03 0,98 1,68 2.11 0,77

0.53-2,61 0.16-1,02 0,0-0,06 0,53-1.43 0,90-3.00 1,0-3.7 0,39-1.2

0.40 0.18 0.01 0.25 0.52 0.51 0.18

Statistical analyses The null hypothesis states the T/E ratio would be unaffected by the administration of DHEA. Because it was possible that the T/E ratios could increase or decrease, pre- and post-dosing ratio means were compared using a two-tailed, paired t-test. Significant differences were reported for p < 0.025. The two-tailed t-value ((x = 0.025, df = 6) for significance was 3.143.

Journal of Analytical Toxicology, Vol. 22, October 1998

Results To verify the accuracy of package labeling for the DHEAused in this study, a random selection of capsules was dissolved in deionized water, and the drug was extracted. The results, as compared with a reference standard, indicated that the package labeling was factual regarding purity and amount. Quantitations were achieved by comparing integrated peak areas of the samples with the integrated peak areas of the standard curve. Baseline mean T/E ratios ranged from a low of 0.1 to a high of 1.2. Urine samples used to calculate these values were collected on three separate days and exhibited minimal variability (data not shown). Representative pre- and post-dose chmmatograms are seen in Figures 1 and 2, respectively. DHEA pre-dose amounts are significantly lower than the post-dose concentrations. These sampies are from approximately halfway through the treatment regimen and depict the relative increase in abundance as a result of one 50-mg capsule. This increase was consistent across test subjects and varied little during the course of the treatment. T/E ratios after 30 daily doses of 50 mg DHEA are shown in Figure 3. Most subjects showed an immediate increase in T/E ratios after the first day of treatment. Over the 30-day period, the T/E ratios exhibited varying degrees of fluctuation. For example, subject 2 had a baseline mean of 0.34 and a treatment mean of 0.43; in contrast, the baseline mean of subject 6 was 1.20, whereas his treatment mean almost doubled to 2.11. These two subjects marked the least and greatest variations, respectively, observed between the baseline and treatment mean. Subject 6 also showed the greatest difference between his baseline mean of 1.2 and peak treatment ratio of 3.7. All subjects, except subject 3, exhibited a peak T/E ratio at least double their baseline mean at some point during the study. However,t-test results indicated that the difference between the baseline means and the treatment means were not significant (twotailed t-value [o~ = 0.025, df = 6] = 3.143; Itl = 0.05). No peak T/E ratios approached the 6:1 threshold that suggests testosterone abuse. Additionally, all urine testosterone concentrations examined in this study were below 120 ng/mL (data not shown). No adjustment was made to any sample correcting for urinary specific gravity or creatinine concentrations. The results of a single 250-mg dose of DHEA

0oooo7o0ooi

3

60000

2

i 50000

4--U

30000

1

4

20000

100OO 28.00

29.00

30.00

31.00

32.00

33.00

34,00

35.00

36.00 3 % 0 0 38.00

39.00

Time (min) Figure 1. Typical chromatogram of a pre-DHEA dose (50 rag) urine sample. Peak identification: I, DHEA; 2, epitestosterone; 3, testosterone; 4, 16-co-OH-testosterone(IS).

200000i

150000 100000

5OOO0 ~ o ~ 28.00

29.00

30.00 31.00

32.00

33.00

34,00

35.00

36.00

37.00

38.00

39,00

Time (min) Figure 2. Typical chromatogram of a post-DHEA dose (50 mg) urine sample. Peak identification: I, DHEA; 2, epitestosterone; 3, testosterone; 4, 16-0c-OH-testosterone(IS).

t

2

3 4

5

6

7 8

9 10 11 12 13 t4 IE 11 17 111 l g ;Z0 21 22 .23 24 25 215 27 28 29 30 31

Day. Figure 3. Daily urine T/'Eratios. Urine samples were collected approximately 3 h after each 50-rag dose. TiE ratios were calculated by dividing the integrated area of the testosterone peak by the integrated area of the epitestosterone peak, "Day 1" values = baseline mean T/E ratio.

457

Journal ot Analytical Toxicology, Vol. 22, October 1998

are shown in Figure 4. This dose resulted in a 40% increase in the subject's T/E ratio relative to the pre-dose value. The peak T/E ratio after this 250 mg single dose was 1.2, which, again, is much lower than the 6.0 traditionally used as the cutoff indicating testosterone abuse. This increase in T/E ratio occurred 2-h post-dose and persisted until the fifth hour. The ratio returned to pre-dose levels approximately 7-h post-dose. As in the initial study, the urinary testosterone concentration did not surpass 120 ng/mL at any time point tested.

cortex, ovary, and testis) and such peripheral tissues as skin, adipose tissue, breast, lung, endometrium, prostate, liver, epididymis, and brain. Two essential enzymes for the conversion of DHEA to active androgens/estrogens are 3]3-hydroxysteroid dehydrogenase and 171B-hydroxysteroid dehydrogenase (3). Neither of these enzymes is present in skeletal muscle (12). Without the expression of both these enzymes, it is difficult to conclude that this tissue has the capability of generating testosterone from either DHEA or DHEA-S. This being the case, the use of high doses of DHEA, while having presumed effects in those tissues that possess the enzymatic activity, would not result in the production of testosterone in muscle. Discussion As stated, no evidence exists that DHEA promotes muscle growth. Therefore, the Drug Enforcement Administration had DHEA is an endogenous, steroid that is also produced in a determined DHEAis not an anabolic steroid and is not subject to sulfated form (DHEA-S) (10). Production of both of these comregulation under the Controlled Substances Act. According to pounds occurs primarily in the adrenal glands. Normal concenthe Federal Food, Drug and Cosmetic Act (FFDCA, Sect. 201), trations in adults exceed that of any other steroid except a "dietary supplement" contains one or more of the following: cholesterol. DHEA-S levels in adult men are 100-500 times "...a vitamin, a mineral, an herb/botanical, an amino acid, a higher than testosterone and 1000-10,000 times higher than dietary substance used to increase intake and/or a concentrate, estradiol concentrations in women. DHEA and DHEA-S intermetabolite, constituent or extract". DHEAfalls under this proviconvert, with the equilibrium favoring conversion from DHEA-S sion as long as therapeutic claims are not made, and in 1994, the to DHEA. Dietary Supplement and Education Act opened the door to overAlthough neither of these compounds has intrinsic androthe-counter distribution of DHEA. genic, estrogenic or other classical hormonal activity, both, espeThe fame, and often fortune, that results from participation in cially DHEA-S, are converted into androgens and estrogens in national and international sporting events is a powerful psychoperipheral tissues (11). Approximately 50% of total androgens in logical and economic motivator. The pressure to succeed has men are derived from these compounds. In women, 75% of forced athletes to look for ways to enhance their performance. estrogen formation is from DHEA/DHEA-S before menopause, The first documented case of athletes using performanceand it rises to almost 100% after menopause. enhancing compounds was by the Russian weightlifting team in It has been well demonstrated that the biosynthesis of androthe 1950s. The compound they used turned out to be testosgens from DHEA is limited to the appropriate target tissues terone (4). Although testosterone does have androgenic (maswithout leakage of significant amounts of active androgens into culinizing) activity, it is the anabolic (muscle building/reduction the circulation (11). This local biosynthesis and action of androin muscle loss) effects that are exploited by athletes. gens, termed "intracrinology", eliminates the inappropriate Beginning in 1972, the IOC instituted full-scale urine drug exposure of other tissues to androgens, thus minimizing the testing at the Munich Olympic Games. The IOC has banned the risks of undesirable side effects. Conversion of DHEA occurs priuse of synthetic anabolic steroids since 1974 and banned the use marily in the "classic" steroidogenic tissues (placenta, adrenal of exogenous testosterone in 1984. However, because testosterone is an endogenous product, another means of detection was required to determine "abuse". In 1984, the ]OC and other sports federations agreed to use the method of Donike (6) to measure T/E ratios. They agreed that a ratio greater than 6:1 would be an indicator of abuse. Since the appearance of DHEA as an over-the.2 ' counter product, investigators have reported that I'~ o9 steroid abusers with positive urine drug tests are claiming that their use of DHEA, not exogenous testosterone, is accounting for their elevated T/E ratios, much as the "nasal inhaler defense" was used by methamphetamine abusers in the 1980s and early 1990s. It is well established that DHEAis not converted to testosterone in muscle, and any increased Hours post-dose production of testosterone in target tissues does not appear to "leak" into the circulation (10,11). Figure 4. Hourly urine T/E ratios following a single 250-mg dose of DHEA. Ratios were calculated The steroids produced in peripheral tissues are as in Figure 3. used locally, and only inactive metabolites are 13

12

05

.

0

458

5

.

10

.

.

,

15

20

.

25

Journalof Analytical Toxicology,Vol. 22, October 1998

released into the circulation. It is possible, however, that testosterone produced locally, used on site, metabolized, and excreted as the common glucuronide conjugate, could contribute to urine testosterone levels, thereby affecting the individual's urinary TIE ratio. Although serum levels of free testosterone are not elevated as a result of DHEA administration (10), urine levels could increase. This may result as a consequence of a high firstpass effect contributing to urinary testosterone metabolites and not to systemically active testosterone in the plasma. If this hypothesis is correct, then it may be possible for peripheral tissues to produce enough testosterone and excrete the corresponding metabolite to significantly alter the TIE ratio in those individuals using/abusing DHEA. The results of this study suggest that it is possible for DHEA to increase urine testosterone levels but that this increase is not sufficiently pronounced to be mistaken for testosterone abuse. Additionally, a single, acute high-dose DHEA administration had no significant effect on the TIE ratio. A stu@ of the effect of chronic, high-dose DHEA consumption is currently underway.

References I.

R.V. Brooks, G. Jeremiah, W.A. Webb, and M. Wheeler. Detection of anabolic steroid administration to athletes. J. Steroid Biochem. 11:913-917. (1979). 2. C.K. Hatton and D.H. Catlin. Detection of androgenic anabolic steroids in urine. Ther. Drug Monit. 7(3): 655-668 (1987). 3. I. Belkien, T. Schurmeyer, R. Hano, P.O. Gunnarsson, and E. Nieschlag. Pharmacokinetics of 19-nortestosterone esters in normal men. J. Steroid Biochem. 22:623-629 (1985).

4. D.H. Catlin and D.A. Cowan. Detecting testosterone administration. Clin. Chem. 38(9): 1685-1686 (1992). 5. A.T. Kicman, R.V. Brooks, S.C. Collyer, D.A. Cowan, M.N. Nanjee, G.J. Southan, and M.J. Wheeler. Criteria to indicate testosterone administration. Brit. J. Sports Med. 24:253-264 (I 990). 6. M.L. Donike, K.R. Barwald, K. Klosterman, W. Schanzer and J. Zimmermann. The detection of exogenous testosterone. In Sport: Leistung und Gesundheit, H. Heck, W. Hollmann, H. Leisen, and R. Rost, Eds. Deutsche Artzte-Verlag, Koln, Germany, 1982, 293-300. 7. R.V. Brooks, A.T. Kicman, N.S.E. Nanjee, and G.J. Sothan. Detection of the administration of testosterone. Abstract 1305, Proceedings of the 12th International Congressof Clinical Chemists, Rio de Janeiro, Brazil, 1984. 8. M. Donike, J. Zimmermann, K.-R. Barwald, W. Schanzer, V. Christ, K. Klostermann, and G. Opfermann. Routinebestimmung yon anabolike im ham. DeutscheZ. Sportmed. 35:14-24 (1984). 9. M Donike and J. Zimmermann. Zur darstellung von trimethylsilyl-, triethylsilyl-, und tertbutyldimethylsilyl-enolathern yon ketosteroiden fur gas-chromatographische und massenspektrome-trische untersuchungen. J. Chromatogr. 202:483-486 (1980). 10. F. Labrie, A. Belanger, L. Cusan, and B. Candas. Physiological changes in dehydroepiandrosterone are not reflected by serum levels of active androgens and estrogens but of their metabolites: intracrinology. J. Clin. Endocrinol. Metab. 82:2403-2409 (1997). 11. F. Labrie. Intracrinology. ~viol. Cell Endocrinol. 78: Cll 3-Cl1B (1991). 12. F. Labrie, A. Belanger, J. Simard, V. Luu-The, and C. Labrie. DHEA and peripheral androgen and estrogen formation: intracrinology. Ann. NY Acad. Sci. 774:16-28 (1995).

Manuscript received January 22, 1998; revision received June 9, 1998.

459