| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIAL |
United States Anti-Doping Agency, Colorado Springs, Colorado 80906
Address all correspondence and requests for reprints to: Larry D. Bowers, Ph.D., Senior Managing Director, Technical and Information Resources, United States Anti-Doping Agency, Colorado Springs, Colorado 80906. E-mail: lbowers{at}usada.org.
The first documented use of testosterone as a performance-enhancing substance in sport began in the 1950s. The success of Russian weightlifters during this era led Dr. John B. Zeigler to surreptitiously obtain information from their trainers that they were using testosterone. Later, Zeigler would use Dianabol to enhance the performance of weightlifters at the York Barbell Club beginning in 1958. Once Pandora’s Box was open, it was impossible to close. The growth and widespread use of anabolic steroids in many competitive sports eventually led the U.S. Congress to schedule testosterone and other anabolic agents under the Anabolic Steroids Control Act in 1990 (Public Law 101-647). For many years, the medical community argued that there was no proof that testosterone enhanced athletic performance. Finally, in 1996, Bhasin and co-workers showed that supraphysiological doses of testosterone indeed produced performance benefits (1), something the athletes had known for four decades.
The urinary testosterone to epitestosterone (T/E) ratio was the first test to be used for evaluation of abuse of a natural (or endogenous) substance such as testosterone to enhance performance in sport. The production of reliable quadrupole gas chromatograph/mass spectrometers in the mid-1970s allowed more widespread study of urinary steroid profiles. Donike and co-workers (2) observed an unusual distribution of testosterone concentrations in samples collected at the Moscow Olympics in 1980. In 1982, Donike proposed the use of the urinary T/E ratio based on the observation that epitestosterone was excreted in the urine in a relatively constant amount and as such constituted a way to compensate for the variability in testosterone concentration that occurred as a function of the diluteness of the urine (2). Because the primary metabolite of testosterone is testosterone glucuronide, the procedure developed by Donike used β-glucuronidase in the preparation of the samples for analysis. In 1983, the International Olympic Committee accepted a T/E ratio of greater than 6:1 as evidence of testosterone doping, based on the 99th percentile of the log normal distribution of ratio values established by Donike and his co-workers.
It was soon clear that the T/E ratio based on population reference ranges had limitations. For example, it had been noted in 1970 that genetic make-up could have a significant effect on urinary testosterone excretion, whereas epitestosterone excretion was unaffected (3). As population T/E ratio data accumulated in the anti-doping laboratories, it was clear that there were two distributions of urinary T/E ratios in both men and women, one with a maximum frequency at a T/E ratio of about 0.1 and a much larger population with a maximum frequency at a T/E ratio of about 1. Within the anti-doping community, the smaller population was referred to as low mode, and it was known that administration of testosterone to low-mode individuals did not result in a T/E ratio above the population threshold (4). Jakobsson and her colleagues (5) provided the mechanism for low-mode excretion in 2006 when they demonstrated that the UGT2B17 gene deletion was the cause of significant differences in testosterone glucuronide excretion between Korean and Swedish men. In their current paper in this issue, Jakobsson-Schulze et al. (6) report that when subjects deficient in the UGT2B17 gene (del/del) receive exogenous testosterone, their T/E ratio does not rise above the current population-based threshold of 4:1. One potential route of excretion of testosterone in UGT2B17 del/del genotypes is through other phase II metabolites such as testosterone sulfate. Borts and Bowers (7) directly measured the concentrations of testosterone and epitestosterone sulfate and glucuronide using HPLC/tandem mass spectrometry. Low-mode excretors did not produce larger than normal amounts of testosterone or epitestosterone sulfate. There was also no change in the relative production of sulfate conjugates. Because phase II metabolism does not explain the difference in testosterone handling in these individuals, it would be interesting to know what phase I metabolism products are increased in individuals with the UGT2B17 del/del polymorphism. This information may help develop a more sensitive test for testosterone abuse in these individuals. Nevertheless, applying a uniform threshold for T/E ratio to a global population of athletes was a difficult proposition.
The potential for individual-based reference ranges was discussed by Harris and Boyd (8) in the 1970s and had achieved some popularity in clinical laboratory medicine. At the 1993 Cologne Workshop, both Donike’s group and Baenziger and Bowers presented studies evaluating the variation of T/E ratio over time in individual athletes. Donike et al. (9) showed, based on results from 47 male athletes, that the coefficient of variation for the urinary T/E ratio was less than 30%. Baenziger and Bowers (10) showed that three patterns emerged from the study of 97 male athletes with at least one T/E ratio greater than 3: a pattern of elevated T/E ratios with consistent mean and coefficient of variation less than 30%, a highly variable ratio over time with coefficient of variation greater than 30%, and a consistent pattern of T/E ratios with a spike inconsistent with other values. Thus, in addition to the low-mode excretors, there was a small population of individuals who produced a consistent naturally elevated urinary T/E ratio. Sottas et al. have recently confirmed this intra-individual variability (11). The variability of the T/E ratio in women was up to 60%, primarily due to the lower urinary concentrations of testosterone and epitestosterone. Thus, comparing an athlete’s T/E ratio to previous or subsequent samples was much more effective in identifying testosterone doping than using a single T/E ratio in a single urine sample. In 1993, the International Olympic Committee Medical Commission advocated the longitudinal evaluation of the T/E ratio in cases when the individual’s T/E ratio was between 6 and 10. The World Anti-Doping Agency (WADA) decreased the threshold for T/E ratio to 4 in 2005. Thus, although individual-based review of suspicious T/E ratio changes is routinely performed by responsible anti-doping organizations, this unfortunately did not address the problem of low-mode excretors.
In October 2006, at the Fifth Annual Symposium on Anti-Doping Science organized by the U.S. Anti-Doping Agency (USADA), about 70 scientists from around the world gathered to discuss how to better use the results of urine steroid testing. With nearly 157,000 urine anti-doping tests performed on Olympic sport athletes worldwide in 2006 (12), there is an extensive base on which to evaluate an individual’s test results. Techniques such as the personalized reference interval and reference change limit models (8) were discussed. Sottas and co-workers (11, 13) have taken the first steps toward validating the use of Bayesian statistics to detect variation from individual norms. Using any of these approaches, the individual with a consistent T/E ratio of 0.1 could be evaluated as having committed a doping offense with a T/E ratio much less than the current 4:1 population threshold based on their individual history of test results. The WADA is moving to make available individual-based evaluations (also known within the anti-doping community as passports) using the Bayesian approach. Although some refinements may be required, individual-based reference range evaluations will enable the anti-doping organizations to identify abuse of testosterone much more efficiently.
Another approach to detecting the use of testosterone abuse is the use of gas chromatography/combustion/isotope ratio mass spectrometry (GC/C/IRMS), which can measure very small differences in the 13C/12C ratio of testosterone or one or more of its metabolites. These differences arise as a result of whether the synthetic precursor for the testosterone came from a plant that fixes nitrogen via a three-carbon or four-carbon pathway. Steroids produced by the body have a 13C/12C ratio similar to the composition of the diet (e.g. corn-fed cattle). Because the carbon source for synthetic testosterone is usually soy or yams, the 13C/12C ratio is different. The ratio is measured relative to a certified reference standard, and the ratio of 13C/12C is reported as a 
-value with units of per mil (
). The -value of the steroid of interest is usually compared with the -value of another steroid, referred to as the endogenous reference compound (ERC), which is either a precursor of testosterone or a member of another steroid pathway and which is not affected by abusing testosterone (14, 15, 16). For example, 5
- and/or 5β-androstanediol (metabolites of testosterone) can be compared with 5β-pregnanediol (ERC) (16). Other compounds used have been androsterone, etiocholanolone, and testosterone. WADA deems that a difference in -values (
) between the steroid of interest and the ERC greater than 3 is a doping violation.
The GC/C/IRMS test has many advantages over the T/E ratio when the test is positive. Unfortunately, a negative test is uninformative because increased production of endogenous testosterone due to administration of LH or human chorionic gonadotropin or the use of a testosterone preparation with a similar -value (e.g. orchic extract) would give a negative result even in the presence of grossly increased amounts of testosterone. There are also limitations related to the amount of the glucuronide of testosterone or its metabolites in the urine, so Schulze’s observations on the altered excretion of testosterone glucuronide from an individual with UGT2B17 del/del or other UGT del/del genotypes may have relevance even with the GC/C/IRMS approach.
The approaches to detection of abuse of testosterone in sports have advanced significantly since Donike’s initial population studies on the T/E ratio. Although low-mode excretors with the del/del UGT2B17 genotype continue to pose some difficulty for doping control, both the individual-based reference ranges and GC/C/IRMS test provide opportunities to deal with these issues. The latter test can be used in both targeted and randomly selected samples even when the T/E ratio is less than 4:1. As an example, several athletes with urinary T/E ratio lower than the population-based threshold have received sanctions when their urine was shown to contain exogenous testosterone by the GC/C/IRMS test.
The purpose of an anti-doping program is deterrence of use. A successful anti-doping program, such as that envisioned in the WADA Code, involves education (e.g. www.usantidoping.org/what/education/; www.usadakids.org; www.thatsdope.org), testing and results management, and research. Collection of samples at the right times (given the characteristics of the test) is a critical element of testing. Test distribution planning must achieve a balance between in-competition and no-notice out-of-competition collections. On the education front, USADA, the American College of Sports Medicine, and other organizations have begun an initiative (Professionals Against Doping in Sports) to engage physicians in educating athletes under their care about steroid abuse and to stress the unethical nature of assisting athletes with doping. I would welcome contact from any medical professional who wishes to assist in this initiative. USADA, WADA, and other anti-doping organizations have funded significant research initiatives in the anti-doping area. For example, USADA has funded a project to improve the sample throughput rates, measurement uncertainty, detection limits, and cost of GC/C/IRMS analysis (17). As long as testing is an integral part of the anti-doping program, understanding both the biology and analytical capabilities of the test armamentarium are important to its success. Schulze and co-workers’ contribution to this understanding is another step forward in the fight against doping in sport.
Footnotes
Abbreviations: ERC, Endogenous reference compound; GC/C/IRMS, gas chromatography/combustion/isotope ratio mass spectrometry; T/E, testosterone to epitestosterone; USADA, U.S. Anti-Doping Agency; WADA, World Anti-Doping Agency.
Received May 5, 2008.
Accepted May 20, 2008.
References
This article has been cited by other articles:
 |
|
 |
|
  S. W. Smith Chiral Toxicology: It's the Same Thing...Only Different Toxicol. Sci., July 1, 2009; 110(1): 4 - 30. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Lippi, U. G. Longo, and N. Maffulli Genetics and sports Br. Med. Bull., February 9, 2009; (2009) ldp007v1. [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 |
