Review Article Anabolic steroids in sport: biochemical, clinical and analytical perspectives Andrew T Kicman and DB Gower

Abstract Address Drug Control Centre King’s College London London SE1 9NN, UK Correspondence Dr AT Kicman E-mail: [email protected] This article has been prepared at the invitation of the Analytical Investigations Standing Committee of the Association of Clinical Biochemists.

International Olympic Committee accredited laboratories play a key role in upholding the principle of fair play and innate ability, as desired by the majority of sports competitors and spectators. Not only does doping damage the image of sport, but it can also be harmful to the individual. The great majority of samples test negative but, when an adverse Žnding is declared, the analytical data must be of a sufŽciently high standard to withstand legal challenges by third parties. The most widely misused performance-enhancing drugs are the anabolic–androgenic steroids, commonly referred to as ‘anabolic steroids’. This review attempts to address the complex issues concerning anabolic steroids in sport by considering the clinical, biochemical and analytical perspectives. Ann Clin Biochem 2003; 40: 321–356

Fifty years of anabolic steroids and sport: an overview The discovery of a process for synthesizing the steroid testosterone (T) from cholesterol in 19351 greatly accelerated scienti¢c investigations into the e¡ects of this endogenous androgen. T rapidly became noted for its anabolic properties in increasing muscle size and strength and in its androgenic properties of increasing virilization and aggression. In a comprehensive review of the history of the use of anabolic steroids in sport, Todd 2 noted that, as early as 1945, the science writer Paul de Kreuf had speculated that `it would be interesting to watch the productive power of [a] . . . professional group [of athletes] that would try a systematic supercharge with testosterone . . .’. In the early 1950s, the ¢rst suspicion that T was actually being administered to improve sporting performance came with the allegation that Soviet weightlifters were administering T to gain strength.2 Also, around that time, pharmaceutical companies were developing synthetic analogues of T with the aim of treating patients in a `catabolic state’, the objective being to enhance the anabolic e¡ects of T and to dissociate the unwanted androgenic e¡ects (particularly to minimize the viriliz ing e¡ects). Nandrolone (19-nortestosterone) was the ¢rst synthetic analogue &2003 The Association of Clinical Biochemists

of T to show enough anabolic-androgenic dissociation in animal experiments 3 to justify its introduction as a therapeutic agent.4 Subsequently, numerous analogues of T [and also its 5a-reduced metabolite, 5adihydrotestosterone (DHT)] were developed (see Fig. 1), such as methandienone (methandrostenolone), oxymesterone and stanozolol (for dates of patent awards see Ref. 5). With the growing availability of these licensed compounds, competitors began to experiment with them in an attempt to gain improvements in sporting performance without imparting the full androgenic e¡ects associated with T. During the 1960s, a number of deaths occurred in sports competitors from the use of stimulants and, as a result, in 1967 the International Olympic Committee (IOC) reestablished a medical commission (an IOC Medical Commission was Ž rst established in Athens in 1961; see Ref. 6) which banned the practice of doping in sport, the banned classes including stimulants and narcotics. Anabolic-androgenic steroids (AASs), commonly termed `anabolic steroids’, were not included as a banned class despite the suspicion of their extensive use by competitors at international level. For example, it was estimated that a third of the USA track and ¢eld team used AASs 2; also, the government of the German Democratic Republic had begun a clandestine programme of promoting AAS administration to their athletes (see `Undesirable 321

322

Kicman and Gower

e¡ects of AASs’, below). Some would argue that the IOC Medical Commission was in£uenced by the medical opinion at that time, in retrospect found to be incorrect, which maintained that the use of these steroids conferred no advantage in athletic performance. More likely, AASs were not included in the IOC

Figure 1.

(continued on page 323)

Ann Clin Biochem 2003; 40: 321–356

list primarily because, at that time, there was no analytical procedure that could meet the combined requirements of being sensitive, rapid and comprehensive. The biological £uid chosen for analysis was urine because (i) it was considered that the collection of blood was too invasive to the individual and, (ii)

Anabolic steroids in sport

after administration, many drugs and/or their metabolites are more concentrated in urine than in blood, allowing easier detection. In 1969, the ¢rst application of radioimmunoas say (RIA) for the measurement of steroids in biological £uids was published.7 Although the technique of RIA met the criteria necessary in being sensitive and rapid, the problem of how to develop a comprehensive screen still remained. At that time, there were 14 licensed orally active AASs and it was clearly impractical to have a speci¢c assay for each compound, let alone their metabolites. However, these steroids have a common 17a-alkyl substituent (12 with a 17 a-methyl group and two with a 17 a-ethyl group); these substituents sterically hinder ¢rst-pass hepatic metabolism, thus giving greater oral availability. The approach used, developed by Brooks et al.,8 was to raise immuno globulins that could target these two alkyl functions. Any presumptive positive sample could then be analysed by gas chromatography- mass spectrometry (GC/MS) in the full-scan mode (see `Con¢rmatory analysis’, below) for con¢rmatory identi¢cation.9 A

323

trial test targeting the orally active alkylated steroids was introduced at the Commonwealth Games in New Zealand in February 1974. Nine out of 55 samples failed the immunoassay screen and seven samples were con¢rmed to be positive by GC/MS. The IOC Medical Commission reacted quickly to these results as, by April of the same year, AASs were introduced as a banned class of compounds in the Anti-Doping Code (in the 1990s, this banned class was changed from AASs to `anabolic agents’ to incorporate out-ofcompetition testing for clenbuterol and other b2agonists, which are considered to have anabolic activity). In the latter half of the 1970s, RIA screens were developed to detect the presence of nandrolone in urine, this steroid being formulated for intramuscular injection.10,11 Subsequently, RIAs were developed for nandrolone metabolites12 and the orally active 1methyl steroid, methenolone. However, in the early 1980s, rapid improvements in mass spectrometers allowed IOC-accredited laboratories to develop a speci¢c and comprehensive screen which could detect

Figure 1. Structures of anabolic–androgenic steroids with corresponding diagnostic metabolites and examples of registered trade names. For a comprehensive list of trade names, together with the name and country of corresponding manufacturers, see Ref. 14. Superscripted numbers against trade names refer to 17b-hydroxyl-esteriŽed preparations: 1undecylenoate; 2acetate; 3propionate; 4 heptanoate; 5decanoate; 6hexahydrobenzylcarbonate. Ann Clin Biochem 2003; 40: 321–356

324

Kicman and Gower

as little as 1 mg/L of an AAS and/or its metabolite in urine13 (see `Screening for xenobiotic AASs in urine’, below). The advantages that a GC/MS screen conferred resulted in the accredited laboratories adopting this technique to replace RIA for screening purposes, an approach that is still used. The current availability of AASs may appear to be limited due to many being withdrawn as licensed products in numerous countries; however, they continue to be supplied as pharmaceutical preparations in others, and hence are comparatively easy to obtain on the international market (for details of existing manufacturers of each AAS listed in Fig. 1, see Ref. 14). Since the introduction of drug control in sport, the number of sports samples analysed worldwide has grown to over 100 000 annually, with approximately 1-2% of samples being reported positive by IOC-accredited laboratories, of which the majority were due to the presence of AASs. Strategies for evasion of detection of AAS adminis tration continue to this day, `natural’ hormones (see Fig. 2) being used despite their unfavourable anabolicandrogenic dissociation; it is more di¤cult to prove administration of such hormones as opposed to xenobiotic steroids. Indeed, T appears to be the most

popular choice amongst all the AASs administered, judging by IOC statistics on adverse ¢ndings. Doping with other `natural’ steroids includes DHT; urine samples collected from11Chinese swimmers, who had competed at the 1994 Asian Games held in Hiroshima, were declared positive for DHT by the IOC-accredited laboratory in Tokyo. Subsequently, the Olympic Council of Asia considered from the evidence that doping with DHT had occurred. In the same year (1994), Morales et al.15 published the e¡ects of replacement therapy of dehydroepiandrosterone (DHEA) in men and women of advancing age. At the beginning of 1995 these results were used by the media to liken DHEA to the `fountain of youth’. Prior to that, DHEA was not considered as a steroid of abuse probably due to its limited availability, even though the physician, Di Pasquale, mentioned as early as 1990 that DHEA `is used by some athletes as an anabolic agent’.16 With media attention, DHEA became widely available in the USA and marketed as a dietary supplement, although it must be emphasiz ed that steroids have no nutritional value. DHEA was reputedly used by some athletes before and during the 1996 Olympic Games in Atlanta. Although DHEA would have been considered a prohibited substance

Figure 2. Structures of ‘natural’ androgens (i.e. those produced as part of the steroidogenic pathway in man) that are administered for anabolic purposes. Testosterone and 5a-dihydrotestosterone are available as pharmaceutical preparations and the others as dietary supplements (NB: endogenous androstenediol is unsaturated between C-5 and C-6 whereas the exogenous form supplied as a dietary supplement is purported to be unsaturated between C-4 and C-5). Ann Clin Biochem 2003; 40: 321–356

Anabolic steroids in sport

under the Olympic Movement Anti-Doping Code, in December 1996 the IOC Medical Commission addressed concern as to whether it came under the banned class of anabolic agents by explicitly adding it as a named example (the Anti-Doping Code can be accessed via the Internet17). Currently, attention has focused on the commercial availability of other `dietary supplements’ or `prohormones’, these being androstenedione, androstenediol and their corresponding 19-nor analogues, which are being marketed as substances that can enhance muscle building. A further complication associated with the marketing of prohormones is that their purity is not guaranteed; for example, androstenedione supplements have been contaminated with 19-norandrostenedione in su¤cient quantities to cause a positive urine test result for 19-norandrosterone,18 the standard marker for detecting nandrolone use. At the present time, IOC-accredited laboratories play a key role in upholding the principles of fair play and innate ability, as desired by the majority of sports competitors and spectators. Not only does doping damage the image of sport, but it can also be harmful to the individual (see `Undesirable e¡ects of AASs’, below). The great majority of samples test negative but, when an adverse ¢nding is declared, the analytical data must be of a su¤ciently high standard to withstand legal challenges by third parties. This review attempts to address these complex issues.

Effects on body mass and athletic performance Use of AASs in the treatment of wasting disorders Ironically, whereas AASs can be abused in sport, their clinical usefulness in reversing the catabolic state of patients, such as those with severe burns or wasting diseases, has not been fully realized. Hence, many AASs have been withdrawn as licensed products in the UK and numerous countries world-wide. The exceptions are oxymetholone, which has bene¢cial e¡ects on damaged myocardium,19 stanozolol, used in the treatment of aplastic anaemia, T preparations (and also mesterelone but seldom used, if at all) for hormone therapy in male hypogonadism, and nandrolone decanoate for osteoporosis (but not an advocated treatment). Notwithstanding, recent reviews have highlighted the extreme clinical usefulness of AASs in muscle-wasting states in HIV-related diseases.20^22 For example, in a 30-week prospective pilot study,23 patients with HIV-related wasting were divided into an untreated group, a group receiving oxymetholone and another group receiving oxymetholone with ketotifen (ketotifen blocks the production of tumour necrosis factor, which plays an

325

important role in the pathogenesis of muscle wasting). The results showed that oxymetholone was safe and promoted weight gain but the addition of ketotifen did not further support weight gain. These results contrasted sharply with weight loss in the untreated group. From another series of studies, Bhasin et al.20 concluded that T administration to HIV-infected men with low circulating T concentrations resulted in weight gain, increased lean body mass and muscle strength.

Effects on muscle size and strength in man The action of AASs in stimulating growth of skeletal muscles in subjects with low circulating T, such as women and children, is undisputed. However, an early (1976) and comprehensive review of previous results concluded that there was little evidence for supraphysiological doses of T or synthetic AASs having any appreciable e¡ect on muscle size or strength in healthy men,24 although many of the studies reviewed had a lack of adequate control and standardization. More recent reviews suggest that the administration of AASs can consistently result in signi¢cant increases in strength if male athletes satisfy certain criteria, including the timing of doses and dietary factors.25^27 In attempts to clarify this problem, Bhasin et al.28 standardiz ed protein and energy intake in groups of men with weight-lifting experience. This study utilized the sensitive and advanced technique of magnetic resonance imaging for measuring small changes in muscle mass. Forty-three volunteers were assigned to one of four groups: placebo with no exercise; T (600 mg T enanthate per week for 10 weeks) with no exercise; placebo plus exercise; and T plus exercise. The men in the exercise group received controlled, supervised training 3 days per week during the treatment period. Muscle size, strength and fat-free mass increased in the placebo group with exercise and in the T group without exercise. In the group undergoing the combined regime, T treatment plus exercise, there were greater increases in muscle size and strength compared to either intervention alone; i.e. the e¡ects of combining supraphysiologica l doses of T with exercise were additive. Subsequent work showed that increases in fat-free mass, muscle size, strength and power are highly dose-dependent and correlated with serum T concentrations. 29 With respect to oral administration of `prohormones’, it has not been demonstrated at this time of writing that these supplements can enhance sporting performance or increase strength, power or muscle size. Following very large doses of DHEA (1600 mg/ day for 4 weeks), one study showed that there was neither an increase in lean body mass, nor an e¡ect on parameters of energy and protein metabolism in young men.30 Another study showed a loss of body fat Ann Clin Biochem 2003; 40: 321–356

326

Kicman and Gower

but no statistically signi¢cant increase in body mass, although the authors argued that these ¢ndings inferred an increase in muscle mass.31 With chronic ingestion of 300 mg/day of androstenedione in men, a dose that exceeds that recommended by most manufacturers of supplements, no augmentation of muscle size or strength with resistance training was observed compared with those receiving placebo.32 However, serum oestrogen concentrations can increase with oral androstenedione administration (e.g. Refs 32,33) and hence large doses to men may have adverse e¡ects associated with raised oestrogen (e.g. increased risk in the development of gynaecomastia). Supplements of DHEA or androstenedione may be of little or no bene¢t to healthy young men who wish to improve their strength and sporting performance if, as would be expected, any anabolic e¡ect is primarily mitigated through peripheral conversion of DHEA or androstenedione to T. Ingestion of DHEA can result in an increase in circulating DHEA and androstenedione but it is not resolved as to whether there is an increase in plasma T (e.g. Brown et al.34). This is not surprising because, in the adult male, the overall peripheral contribution of these precursor steroids to circulating T is small, approximately 95% originating directly from testicular secretion. Any contribution from exogenous DHEA or androstenedione will be largely moderated by the large amount of Tcontributed by the testis. Following very large doses of DHEA (1600 mg/ day for 28 days) to ¢ve young eugonadal men, free and total serum T did not increase.31 With regard to androstenedione, chronic administration of 300 mg/ day to male volunteers aged 19- 40 years resulted in considerable inter-individual variability to changes in serum T. King et al.32 showed no signi¢cant increase in serum T over an 8-week administration period (n=10), whereas Leder et al.33 reported a signi¢cant increase after 7 days of administration (n=14), but only a few of those subjects (n=4) had serum T concentrations that exceeded the reference range. For androstenediol, Uralets and Gillette 35,36 administered the ¢4 and the ¢5 forms to male volunteers and reported urinary concentrations of T in the former study. Following administration of the ¢4 form (100 mg, p.o.) to three male volunteers, large increases in urinary T concentrations were observed up to 20 h post-administration. However, plasma T was not assayed and it is probable that the increase in urinary T concentration in the adult male may be predominantly due to increased hepatic metabolism of androstenediol to T glucuronide rather than any large increase in circulating T. In young adult women, an increase in performance may be possible following ingestion of these supplements, as circulating T would be expected to increase. The plasma concentration of T (0Í7-2Í6 nmol/ L) is Ann Clin Biochem 2003; 40: 321–356

very approximately one-tenth of that found in men, and the proportion arising from peripheral conversion is much greater. Even though only 12-14% of androstenedione is converted peripherally to T,37,38 this amount accounts for about half the circulating T in women. As the peripheral contribution to blood T is far greater in young adult women than men, ingestion of modest amounts of DHEA, androstenediol or androstenedione is likely to raise circulating T. For example, even a single 100-mg oral dose of androstenedione to two women increased their plasma T concentration four- and sevenfold, respectively,39 whereas the same acute dose to men (n=10) did not a¡ect either free or total serum T.32 A recent study, in which a single dose of androstenedione (100 mg) was administered to ten young women, showed that the plasma T concentrations increased from *1nmol/ L to a maximum mean of 25 ¢1nmol/ L at 75 min and remained signi¢cantly di¡erent from control values between 30 min and 8 h post-administration.40 The mean exposure (as determined by area under the curve) to testosterone was greater than an order of magnitude compared to the control period, and the plasma concentrations observed were similar to those encountered in abuse of T for anabolic purposes. A similar pro¢le would be expected with chronic administration, but the risk of virilization precludes such a study. Given the concern raised by health care professionals about the medical consequences of misuse of anabolic steroids, this work supports the current consideration by UK and USA drug regulatory authorities in classifying androstenedione with the same status as T to regulate its availability.

Anabolic, androgenic or anticatabolic action? Androgens are responsible for the sexual dimorphism in skeletal muscle; hence, chronic administration of AASs to women will increase their muscle mass and strength. In contrast, despite the large number of studies in the 1960s and 1970s, it has only recently been clearly demonstrated (as noted above) that the administration of androgens can increase muscle mass in intact men.28 A possible biochemical explanation of this clearly apparent di¡erence in response between sexes may be drawn from studies with rodents. Dahlberg et al.41 showed that the fully developed intact male rat had a lower concentration of androgen receptors compared to the fully developed female rat. Castration increased the ligand a¤nity of these receptors. If this situation is analogous to the human, then it is possible that the large increase in T secretion observed during puberty in young men causes down-regulation of receptors in skeletal muscles. After puberty, the receptors in these muscles would be saturated by T; therefore administration of supratherapeutic doses of T or synthetic AASs would

Anabolic steroids in sport

evoke little, if any, additional e¡ect via the androgen receptors in skeletal muscle. This prompts the question:What other possible mechanisms may mediate an anabolic e¡ect in healthy intact men? Some athletes claim that the administration of AASs enables faster recovery from fatigue during training; AASs may counter the catabolic e¡ect in athletes with high concentrations of endogenous cortisol due to the stress of severe training schedules.42 Nonetheless, the role of AASs as glucocorticoid antagonists is unclear. As summarized by Hickson et al.,43 both in vitro and in vivo studies have failed to show a consistent positive interaction of these steroids with the glucocorticoid receptors in muscle. Sharpe et al.44 postulated that AASs do not occupy glucocorticoid receptor sites but down-regulate the glucocorticoid receptor content. Indirect evidence of an antiglucocorticoid e¡ect in humans comes from a case study in which a patient with androgen insensitivity syndrome received pharmacological doses of T enanthate (5 mg/kg/day for 10 days). As expected of a patient with androgen insensitivity, administration did not suppress gonadotropin secretion, nor were there changes in plasma sex hormone binding globulin concentration or sebum excretion. Despite this, administration resulted in a positive nitrogen balance of 3 g per 24 h, similar to that expected in healthy men;45 this supports the hypothesis that AASs can evoke an antiglucocorticoid e¡ect independent of the androgen receptor.

Other effects Although many of the androgenic e¡ects of AASs are undesirable, the psychological changes, particularly increased aggression which can result in antisocial behaviour in some, may be harnessed by others to help their training and increase their competitiveness.46 Finally, AASs may enhance performance by stimulating the production of erythrocytes. As discussed by Mooradian et al.,47 androgens stimulate the production of erythropoietin and also act directly on the erythropoietic stem cells in the bone marrow.Whether there is an increased erythropoietic e¡ect in healthy competitors is debatable, as the weight of evidence suggests that there is no improvement in aerobic performance of athletes treated with AASs. 25

Undesirable effects of AASs General The undesirable e¡ects induced by AAS administration have been extensively reviewed.22,25,48^50 The severity of the undesirable e¡ects depends on the gender, the type of steroid or combination of steroids, the size of dose and the duration of administration. With the exception of possible psychological changes, no adverse e¡ects from acute overdosage with AASs in

327

healthy adults have been reported. With chronic administration, adverse e¡ects can be manifested with pharmacological doses, and abusers run the risk of a permanent damage to health. The irreversible e¡ects of virilization in females and the stunting of linear growth in adolescents is of particular concern. The risks of undesirable e¡ects are augmented in competitors and body-builders who can take many times the therapeutic dose of one or several (`stacking’) AASs over sustained periods of time (`on cycles’of several weeks to months).49 The risk of developing serious e¡ects from chronic administration of AASs may appear to be relatively low when compared to the use of socially acceptable drugs such as tobacco. However, it took approximately 25 years to prove conclusively that smoking increases the risk of contracting lung cancer and, on that basis, it may take much longer to assess the risk of adverse e¡ects of AAS administration because it is predominantly an `underground’ activity. The doses of AASs administered can be far greater than the physiological or therapeutic doses in clinical studies and to administer such doses in a controlled study over a prolonged period of time would be unethical. Nonetheless, one of the largest pharmacological `experiments’ ran for three decades in the former German Democratic Republic (GDR) with a clandestine governmentally and scienti¢cally supported doping programme to improve sporting performance. With the collapse of the GDR in 1990, previously secret documents showed that AASs were administered to several thousand GDR athletes every year since the mid-1960s in an increasing number of discipline s. These previously secret doctoral theses, scienti¢c reports and reports made by some sports physicians for the Stasi, record damaging e¡ects in detail. Of note, is that nowhere did the GDR doctors describe a damaging e¡ect that is not described in the `Western’ literature.51 Increasing information is being gained from former athletes who are coming forward to describe the damage to their health. With such information and the possibility of follow-up studies to investigate the incidence of symptoms among former GDR athletes, compared to individuals not exposed to AASs, may help to elucidate just how harmful the use of these steroids is.

Effects on the liver Liver dysfunction is commonly cited as a result of chronic administration of AASs, especially with the C-17a-alkylated steroids, which are orally active 48 (see `Structure - activity relationships’, below), and the ability of AASs to induce liver disease is undisputed. Symptoms may appear to be related to large doses, as morphological evaluation of the liver in hereditary angioedema patients receiving long-term low to moderate doses of stanozolol or danazol did not induce Ann Clin Biochem 2003; 40: 321–356

328

Kicman and Gower

signi¢cant hepatic damage as assessed by laboratory tests or biopsy.52 Caution must be exercised in the interpretation of results from liver function tests as exercise can also increase alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels, as well as lactate dehydrogenase and creatine kinase (CK). Although the aminotransferase levels increase initially with steroid treatment, they can return to normal often while the steroids are still being used.48 Moreover, Dickerman et al.53 have questioned whether AAS hepatotoxicity has been overstated. Two groups of body-builders were studied, one taking self-directed regimens of AASs, the other not taking steroids. In both groups AST, ALT and CK were elevated, whereas glutamyltranspeptidase (GGT) levels remained normal, although the latter enzyme has generally been accepted in the past as a more speci¢c marker of hepatic dysfunction than the aminotransferases. Levels of CK were raised in groups taking exercise, whether on AASs or not. Patients with viral hepatitis (a control group) were the only individuals in whom there was a correlation between aminotransferases and GGT. The authors concluded that levels of CK and GGT as well as AST and ALT should all be taken into account when attempting to evaluate possible hepatotoxicity in the use of AAS therapy or abuse. An athlete with abnormal liver function due to steroid administration should be made aware of the dangers and advised to discontinue treatment immediately. If an abnormal liver function prevails, this is indicative of a severe liver disease or noncessation of administration. At this stage, analysis of a urine sample for AASs will show whether the athlete has truly discontinued administration. Although the incidences are rare, hepatic cholestasis (bile canal obstruction), peliosis hepatitis (blood-¢lled sacs in the liver) and liver tumours are associated with long-term AAS treatment. The increase in risk of developing these life-threatening conditions appears to be linked with administration of almost exclusively the orally active 17 a-alkylated AASs for periods greater than 6 months for cholestasis or peliosis hepatitis and 24 months for liver tumours.

Psychological effects As noted above, apart from a direct anabolic e¡ect, natural androgens and their synthetic analogues can potentially enhance performance by increasing aggression, an attribute that may be harnessed for improvements in training and competition. However, for some individuals, the psychological changes can be insidious, reportedly giving rise to violent behaviour (`roid rage’) and psychosis.54,55 For example, steroid misuse has been associated with murder, nearhomicide and violence towards women.54,56 AAS Ann Clin Biochem 2003; 40: 321–356

users may show signs of dependency;57 withdrawal can lead to clinical depression.

Effects on the cardiovascular system It is di¤cult to evaluate these e¡ects as most of those described are based on case reports. However, there are indications of a number of serious adverse e¡ects associated with AAS administration, which can result in premature mortality. A number of studies (e.g. Hurley et al.,58 Cohen et al.59) have shown that AAS administration results in a change in the lipoprotein pro¢le where low-density lipoprotein (LDL)-cholesterol is raised and highdensity lipoprotein (HDL)-cholesterol and apolipo protein-1concentrations are lowered. Such pro¢les are generally found in those having a greater risk of atheroma and hence coronary heart disease, albeit that there is some evidence that this unhealthy pro¢le may be reversed about a month after cessation of drug use (e.g. Refs 22,60). Whether this greater risk translates to users who are gaining in body weight while taking regular exercise remains to be proven but it is a cause for concern. Sader et al.61 studied 20 male body-builders (mean age 35 years), ten of whom were actively using AASs and ten who denied ever using them. The use of AASs was associated with signi¢cant decreases in both plasma cholesterol and HDL and there were signi¢cant increases in left ventricular mass and self-reported physical strength. Arterial reactivity (a measure of elasticity) of the right brachial or radial artery, at rest and with dynamic tests including the response to sublingual nitroglycerine, was not signi¢cantly di¡erent in either group of body-builders, nor the carotid intima-media thicknesses. Sullivan et al.,62 have described a young male bodybuilder who was consuming large quantities of AASs, and who presented to the local emergency department with symptomatic rapid atrial ¢brillation. An echocardiogram revealed signi¢cant septal hypokinesis, as well as posterior and septal wall thicknesses that were at the upper limit of normal for highly trained athletes. Ten weeks after stopping AAS ingestion, atrial ¢brillation did not occur. A study by Parssinen et al.63 showed that powerlifters taking AASs had a 4 ¢6-fold increased incidence of premature mortality compared with population controls (12Í9% compared with 3Í1%). Of the powerlifters who died prematurely (mean age of death 43 years), three had myocardial infarction, three committed suicide, one had hepatic failure and one had non-Hodgkin’s lymphoma. Although there is no known link between this last condition and AAS use, there is a suspected link with the use of growth hormone.64 These workers suggest the probable cause of these results is the use of AASs and other

Anabolic steroids in sport

concomitantly ingested drugs, although power training by itself does not increase health risks. Niemenin et al.65 have also reported serious cardiovascular side-e¡ects in four male patients who had ingested massive amounts of AASs during many years of strength training. All had cardiac hypertrophy (`athletes heart’), even though hypertrophy can occur with exercise alone. One patient had ventricular ¢brillation during exercise, one had arterial thrombus in the lower leg, and two patients had signs and symptoms of heart failure (one of these two having massive thrombi in right and left ventricles). The e¡ects of AASs on blood pressure in humans seems to be quite variable, some groups having reported signi¢cant increases in systolic blood pressure, others reporting no e¡ect.49 In a detailed study, Kuipers et al.60 reported on blood pressure, lipid pro¢les and liver function in male body-builders who self-administered AASs for 8 weeks. No changes were recorded in systolic blood pressure in this group compared with two others who were given AASs or placebo in double-blind studies. However, there was a signi¢cant increase (P50Í05) in diastolic blood pressure in the self-administered group. Whether the slight increase in diastolic pressure is dose-related is a subject for debate.

Effects on the prostate The World Health Organization guidelines for the use of androgens in men 66 a¤rms that currently there is no evidence that the administration of androgens in hypogonadal or normal men leads to the development of benign prostatic hyperplasia (BPH) or the progression of pre-clinical to clinical prostatic carcinoma. Although more long-term investigations are required, such may be the case for men receiving T for replacement therapy or for hormonal male contraception. However, the doses administered are relatively small compared to the amounts of AASs administered by athletes and body-builders who may have an increased risk of developing prostatic volume enlargement 67,68 and BPH. In AAS users, disproportionate growth of the inner part of the prostate may occur,69 the area known to be the histological origin of BPH, but any relationship to the early lesions of BPH remains unclear.

Virilization In women, AAS misuse causes virilization (e.g. Ref. 51). Androgenic e¡ects include an irreversible deepening of the voice, clitoral hypertrophy which can be permanent and other gynaecological disorders. Amenorrhoea is common and the libido may be greatly increased during steroid administration (the `on cycle’).Vellus develops into terminal hair giving rise to the male pattern of hair growth on the face and body.

329

In contrast, on the scalp, terminal hair is converted to vellus, making the female susceptible to the male pattern of baldness.

Other undesirable effects Administration of AASs results in an increase in the secretion of sebum and, in those individuals whose sebaceous glands are very responsive to androgens, `steroid acne’ is a frequent complication. Cystic acne can be extensive, particularly on the back, resulting in deep scarring. In some men, gynaecomastia can develop, which is probably due to an alteration in the hormonal balance between androgens and oestrogens from AASs that can undergo aromatization. Franke and Berendonk note that 12 male weightlifters in the DDR had such breast enlargement, often with enlarged nipples, that the tissue had to be removed by surgery.51 Chronic administration of AASs leads to disturbances of the hypothalamic - pituitary- gonadal axis, and the suppression of gonadotrophin secretion can result in infertility, testicular atrophy, disturbances of the menstrual cycle and secondary amenorrhoea. In men, impotence often can occur after cessation of a steroid cycle. Administration can lead to an acceleration of baldness in men who are genetically predisposed that way. Whether this is due to the exposure of the androgen receptors to large amounts of AASs and/ or 5 a-reduced metabolites with greater binding a¤nity to that of DHT remains to be elucidated.

Structure–activity relationships Classically, synthetic AASs are considered as chemical analogues of T or DHT, being designed to enhance the protein anabolic e¡ect relative to the unwanted androgenic e¡ects for clinical purposes (e.g. wasting diseases). Nonetheless, complete dissociation of the anabolic e¡ects from the androgenic e¡ects is not possible, as much evidence suggests there is only one type of active androgen receptor 47 (as an adjunct, the development of non-steroidal molecules called selective androgen receptor modulators with distinct tissue selectivity potential may o Ú er unique therapeutic potential to androgen therapy by eliminating or diminishing the undesirable eÚ ects 70,71). A mechanism for partial dissociation can be based on the marked di¡erence in 5a-reductase activity in the body, which is very high in androgenic target tissue but negligible, if at all, in skeletal muscle.72 One way of increasing anabolicandrogenic dissociation is to administer an AAS that binds with high a¤nity to the androgen receptor but upon reduction to a 5a-metabolite binds with less (e.g. nandrolone).73 Hence, in androgenic tissue, this potent steroid will be readily converted to a less active Ann Clin Biochem 2003; 40: 321–356

330

Kicman and Gower

metabolite, whereas the parent compound will predominate in skeletal muscle tissue. Historically, from the determination of myotrophicandrogenic indices and results from nitrogen balance studies (e.g. Refs 3, 74), some general structure activity relationships were deduced. Both models compared the e¡ect of AAS administration to that of a reference steroid administered (T or 17 a-methyltestosterone for oral routes or the equivalent propionate esters for parenteral routes). Apart from removal of the C-19 angular methyl group from the A/B-ring junction, most increases in anabolic potency and anabolicandrogenic dissociation result from modi¢cations to the A-ring of Tor DHT (see Fig. 3). Oral activity can be conferred by attachment of a methyl or ethyl group to C-17 (the 17 a-alkylated steroids) which prevents deactivation by ¢rst-pass metabolism by sterically hindering oxidation of the 17 b-hydroxyl group. Oral activity can also be conferred by attachment of a methyl group at C-1 as in methenolone or mesterolone but these two AASs are relatively weak in activity. Parenteral preparations do not require a 17a-alkyl group but the 17 b-hydroxyl group is esteri¢ed with an acid moiety 75 to prevent rapid absorption from the oily vehicle, which is usually arachis oil plus a small amount of benzyl alcohol. Oral formulations of DHEA, androstenedione, 19norandrostenedione, androstenediol and 19-norandrostenediol are widely marketed as `dietary supplements’ or `prohormones’. These steroids lack a 17 aalkyl substituent so they undergo extensive ¢rst-pass metabolism. To date, no studies have been undertaken to investigate the activity of 19-norandrostenedione and 19-norandrostenediol but, intuitively, one may consider them somewhat analogous to androstenedione and androstenediol, respectively. DHEA and

androstenedione do not bind to the androgen receptor but they are substrates for conversion to T and hence are sometimes referred to as `weak androgens’. The form of androstenediol that is marketed as a supplement is purported to have the locant of the double bond between C-4 and C-5 (the ¢4 form; e.g. Androdiol 1 ), whereas androstenediol produced endogenously is unsaturated between C-5 and C-6 (the ¢5 form). The androgenic activity of the endogenous steroid is poorly understood and some of its activity may be as a result of metabolic conversion to T (and DHT). However, recent evidence has shown that androstenediol itself can activate androgen receptor target genes in the presence of androgen receptors.76 Further, the ¢5 form is known to have oestrogenic properties, the e¡ects being mediated by the oestrogenic receptor. Whether the ¢4 form has androgenic and oestrogenic activity is not known but, if it possesses the oestrogenic properties associated with that of the ¢5 form, these properties would be considered most undesirable amongst male users of anabolic agents. The ¢4 form is often advertised as a prohormone that is more e¤ciently converted to T compared to androstenedione.

Metabolism of AASs General aspects As xenobiotic AASs are often metabolized extensively, with little parent steroid being excreted into the urine, it is important to identify these metabolites for drug monitoring purposes. An overview of the phase I and phase II metabolic pathways of AASs has been published earlier.77 For many of these steroids, there can be more than one diagnostic metabolite. Figure 1

Figure 3. Structural modiŽcations to the A- and B-rings of testosterone which increase anabolic activity; substitution at C-17 confers oral or depot activity (i.m. ˆ intramuscular). Ann Clin Biochem 2003; 40: 321–356

Anabolic steroids in sport

lists the major AASs, with an example of a principal diagnostic metabolite in each case. In a very comprehensive paper, SchÌnzer and Donike 78 describe the metabolism of AASs in man, the chemical synthesis of the major metabolites, and their GC retention times (determined on OV-1 and SE -54 columns) with the electron impact mass spectra of trimethylsilyl derivatives. Some veterinary steroids [mibolerone (17b-hydroxy-7 a,17 a-dimethylestr-4-en3-one) and trenbolone] are also abused by some sports competitors, and the urinary metabolites of these in man have been characterized.79,80 In the analytical context, it should be noted that not all AASs give rise to unique metabolites; i.e. common glucuronidated metabolites can be produced due to shared pathways of metabolism. 17a-Methyl-5 aandrostane-3 a,17 b-diol is a major metabolite of mestanolone, methyltestosterone and oxymetholone, probably because the latter two steroids are converted in the body to the mestanolone intermediate. 17 aMethyl-5 b-androstane-3 a,17b-diol is a major metabolite of 17 a-methyltestosterone, methandienone and methandriol, the latter two steroids being endogenously converted to the 17 a-methyltestosterone intermediate. Another general consideration concerns epimerization for steroids that possess the 17b-hydroxy-17amethyl con¢guration, the 17a-hydroxy-17 b-methyl epimer being detected in human urine following drug administration (e.g. Refs 81-88). The mechanism proposed by Edlund et al.89 is based on the formation of an unstable tertiary sulphate intermediate. Elimination of the sulphate group can also cause the loss of the angular methyl group attached at C-13, resulting in the formation of18-nor-17,17-dimethyl-13(14)-ene steroids. The metabolic pathways of nandrolone, methandienone and stanozolol are now described in more detail to illustrate the extent of metabolism that can occur.

331

Metabolism of nandrolone Nandrolone (19-nortestosterone, Fig. 1) was synthesized in 1950 by Birch,90 and by 1958 a paper described identi¢cation of two (of the three) major urinary metabolites following intramuscular injection of an aqueous suspension of the steroid.91 Only about 0Í4% of injected nandrolone (usually administered as the decanoate ester) is excreted unchanged except for conjugation with glucuronic acid. A much greater proportion (about 30%) is excreted as the tetrahydro17-oxosteroid metabolites. In this respect, the metabolism of nandrolone is similar to that of T, both steroids undergoing 4-ene-3-oxo reduction and oxidation of the 17b-hydroxy group (see Fig. 4). The major metabolites of nandrolone are 3 a-hydroxy5a-estran-17-one (19-norandrosterone), 3 a-hydroxy5b-estran-17-one (19-noretiocholanolone) and 3 bhydroxy-5 a-estran-17-one (19-norepiandrosterone), but with no 3b-hydroxy-5 b-estran-17-one (19-norepietiocholanolone) being observed.92 For con¢rmatory analysis of positive samples, following hydrolysis of steroid glucuronides, the abundance of 19-norandrosterone was usually found to be greater than that of 19-noretiocholanolone. Relative to these two metabolites, 19-norepiandrosterone glucuronide is present in smaller amounts, although there is some evidence to suggest that larger quantities may be present as the sulphate conjugate.12 This is consistent with the excretion of other 3 b-hydroxylated steroids, such as epiandrosterone, these sulphate conjugates being poorly cleaved with screening procedures that incorporate enzymatic hydrolysis (see `Screening for xenobiotic AASs in urine’, below).

Metabolism of methandienone A summary of the metabolism of methandienone is shown in Fig. 5. The 1,4-diene structure renders the A-ring somewhat resistant to metabolic reduction and

Figure 4. Phase I metabolism of nandrolone to I (19-norandrosterone), II (19-noretiocholanolone) and III (19-norepiandrosterone). Ann Clin Biochem 2003; 40: 321–356

332

Kicman and Gower

Figure 5.

Phase I metabolic reactions for methandienone.

the 17 a-alkyl group of methandienone sterically hinders oxidation of the 17-oxo group. As a consequence, hydroxylation at C-6 is favoured,93 6 b-hydroxymethandienone being one of the earliest AAS metabolites to be found in urine collected from athletes.9 This major metabolite is resistant to phase II metabolism, being almost exclusively excreted as the free steroid. Other major unconjugated metabolites subsequently identi¢ed were the 17a-hydroxy epimer of methandienone (17-epimethandienone) and the 6bhydroxylated metabolite of epimethandienone.84,86 In the conjugate fraction of urine, three major bishydroxylated metabolites (5 b-H orientation) have been identi¢ed;88 these steroids have a 3a-hydroxyl group which facilitates conjugation to glucuronic acid in the human. A fourth metabolite has a 1-ene-3-oxo con¢guration but has also been found in the conjugate fraction,88 presumably because of formation of a 3hydroxyl group by keto- enol tautomerism, which can then be conjugated with uridine diphosphoglucuronic acid.

Metabolism of stanozolol Historically, the detection of doping with stanozolol has been di¤cult for a variety of reasons, not least because it is extensively metabolized, the urinary metabolites had not been fully characterized and the N,O-bis-trimethylsilyl (bis-TMS) derivatives had poor GC qualities on packed columns.9 As opposed to most anabolic steroids which have neutral properties (note: Ann Clin Biochem 2003; 40: 321–356

the phase II metabolites of AASs are analysed as aglycones), the pyrazole moiety confers the parent steroid and metabolites with polar properties. Hence, extraction e¤ciency is governed by the choice of pH of the aqueous phase 83 and chromatographic performance by the number of unwanted polar sites exposed on a GC column employing a non-polar liquid phase (e.g. methylsilicone). With the advent of fused-silica capillary columns in the 1980s, great improvements were achieved in chromatographic resolution, these columns having in the region of 100 000 theoretical plates. The bis-TMS derivatives of stanozolol metabolites can thus be resolved from interfering urinary components and are more stable on these columns. Nonetheless, each column needs to be checked after installation as the number of active sites that can interact with the pyrazole ring appear to vary and these sites can cause adsorption problems. The metabolites of stanozolol have now been characterized83,94 (see Fig. 6). Hydroxylation occurs at C-30, C-4 and C-16 and epimerization at C-17. Hydroxylation at the 16 a and 16 b positions is a major biotransformation route, formation of the a- or b-epimer being subject to inter-individual variation. Phase II metabolism appears to be by glucuronide conjugation with hydroxyl groups at the C-3 0, C-4 and C-16 positions.94 As the N-glucuronide occurs to a major extent in the rat,95 the pyrazole moiety of stanazolol may also be metabolized by N-conjugation in humans.

Collection, handling and storage of samples Urine Urine is the chosen biological £uid for analysis. The sampling procedure for doping control is described in detail within Appendix C of the Olympic Movement Anti-Doping Code.17 Appropriate doping control o¤cers (called independent sampling o¤cers in the UK) witness a urine sample being delivered into a collection vessel, not least to ensure that no manipulation of the sample occurs. Much time and e¡ort has been devoted to ensuring that the sample kits and chain-ofcustody documentation can withstand legal challenges. The athlete is requested to select two coded glass bottles, each assigned a unique code. The urine is then divided between the two bottles, ideally into about 70 mL for screening purposes (usually designated the `A-sample’) and 30 mL for con¢rmatory analysis (the `B -sample’). In some cases, the volume of urine available for analysis is considerably smaller. The bottles are sealed using tamper-proof lids and then sent to the laboratory within a sealed shipping container. The sampling o¤cer is usually requested to measure and

Anabolic steroids in sport

333

Figure 6. Metabolism of stanozolol: I, 3 0 -hydroxystanozolol; II, epistanozolol; III, 3 0 ;-hydroxyepistanozolol; IV, 4b-hydroxystanozolol; V, 16a-hydroxystanozolol; VI, 16b-hydroxystanozolol. record the speci¢c gravity and pH of urine remaining in collection vessels, prior to disposal of that urine. On arrival at the laboratory, the A-sample seal is broken and the urine is analysed. If the A-sample fails a drug test, the B-sample seal is broken at a later date and the analysis repeated, competitors having the option to witness this process with representatives of their choice, such as an independent scienti¢c expert and a legally quali¢ed individual. As specimens are collected at many locations within a country, delays during transit can be as much as several days. These samples are usually at ambient temperature and, as no preservative is added, general degradation of the sample may occur (e.g. increase in pH due to ammonia formation). Markers of general degradation include a pH above that physiologically possible (pH 48Í3) and/or a large presence of 5 aandrostanedione and free steroids that were originally glucuronidated (e.g. androsterone, etiocholanolone).96 However, the absence of signs of general degradation is not conclusive proof that sample integrity has been maintained. Storage of samples in IOC-accredited laboratories is at about +48C or 7208C; in the UK, samples are stored at 7208C. In clinical chemistry, maintaining sample integrity prior to analysis is extremely important and, as a corollary, the apparent lack of precaution in preservation of sports samples in transit is completely alien to that philosophy. However, if preservative is added to sports samples, it is possible that an adverse ¢nding may be challenged on the basis

that such samples failed the test because of adulteration with a foreign material. Although AASs are not thermally labile there has been some concern about the possibility of urinary microbial production of T causing an adverse ¢nding 97^99 when the approved test for detecting doping with T is used. Steroid-transforming microorganisms are common in nature and urine can contain bacteria and yeast due to normal commensals and natural exfoliation. Also, in adult women, symptomatic and asymptomatic urinary tract infections can be present, and urine can be directly contaminated during collection by normal gastrointestinal £ora present on the perineum. However, de la Torre et al.100 did not ¢nd T production following inoculation of urine with selected organisms, but they consider that microbial contamination may hamper interpretation of results. Recently, Kicman et al.101 showed that an increase in urinary T can occur following inoculation of urine with a strain of Candida albicans, but that the increase is minor and hence of little evidential value for any individual sport case. It is di¤cult to comment further as to the possibility of how much T can be formed by di¡erent strains of C. albicans. Interpretation will be helped by isotope ratio MS (see `Detection of ``natural’’ androgen administration’, below) of urinary steroids to determine whether T is of exogenous origin as well as comparing the T liberated by glucuronidase hydrolysis and/or that in the free steroid fraction to total T. Notwithstanding, Ann Clin Biochem 2003; 40: 321–356

334

Kicman and Gower

overcoming the quasi-legal di¤culty of adding a suitable chemical preservative to samples to reduce the possibility of microbial action appears to be a straightforward way of addressing this speci¢c issue.

Additional biological matrices for analysis As well as urine, biological matrices for analysis of AASs could include blood, saliva, sweat and hair. Of these, blood and hair analysis appear to be the most promising as useful adjuncts to urinalysis for AASs, but much work needs to be done to establish these techniques as suitable methods for evidential analysis as part of drug control in sport.

Blood Blood sampling has been performed and is increasingly being established as an additional biological matrix for analysis in doping control. In 1989, the Medical Commission of the Fe¨ de¨ ration Internationale de Ski permitted blood sample collection to screen for heterologous blood transfusion at the World CrossCountry Ski Championship s of that year (no heterologous blood was detected amongst the 66 samples analysed). The IOC Medical Committee ¢rst approved blood tests in 1993 and the following year tests were carried out at the Winter Games in Lillehammer, Norway. With the advent of recombinant DNA technology, human growth hormone and erythropoietin became widely available and concern about misuse prompted the development of tests for these hormones which incorporate blood collection (e.g. as performed at the Sydney 2000 Olympic Games). Currently, the collection of blood in human sports is performed to detect blood doping (autologous and non-autologous), erythropoietin administration and, sometime in the future, for the detection of human growth hormone administration. Whether blood collection will facilitate detection of administration of AASs remains to be seen, but it could be particularly helpful in assessing whether natural androgen administration has occurred.102

Scalp hair Scalp hair analysis for AASs has gained in popularity during the past decade. Not only is scalp hair easily obtained from subjects but the samples are stable with no need for temperature and pH control or addition of preservative during transportation to the laboratory. At least 60 pharmaceuticals and drugs of abuse, together with some AASs such as nandrolone and stanozolol, can be measured.103 In addition, there are reports of the detection of T and its esters, nandrolone and its esters and methandienone in scalp hair after parenteral administration. Especially useful is the fact that a number of samples can be taken (without personal `invasion’ or embarrassme nt) to allow con¢rAnn Clin Biochem 2003; 40: 321–356

mation of an earlier result or for analysis of another possible analyte. It is recommended that approximately 200 mg of scalp hair should be cut o¡ as close as possible to the skin of the posterior vertex region of the scalp, and wrapped in aluminium foil for storage at room temperature.104 However, despite the potential usefulness of hair analysis in drug abuse in human sports, it is generally accepted that a negative hair result cannot exclude administration of a detected drug or its precursor and should not over-rule a positive urinary result105 (however, in a Court of Law, the hair analysis result is bound to cause some ambiguity to the urinary results). There are good reasons for this recommendation, as the incorporation of drugs into hair is variable and depends on a number of factors, including the binding a¤nity with melanin and the hydrophilicity and membrane permeability of the drug concerned.106 The latter parameter is based on the pH gradient between blood and the acidic hair matrix. Thus, basic drugs would be expected to have a high membrane permeability; this is borne out by Rollins et al.,107 who showed that a single dose of codeine could be detected in hair for 8 weeks. Similarly, Henderson et al.108 found that the threshold dose of cocaine injected intravenously was 25-35 mg if it was to be detected in hair. On the other hand, anabolic steroids, despite their lipophilicity, have low membrane permeabilities and, after a single intramuscular dose of 50 mg of nandrolone decanoate, neither the unesteri¢ed steroid nor its decanoate could be detected in the hair of a 37-year-old man.109 In contrast, his urine showed positive for nandrolone metabolites (19-norandrosterone and 19-noretiocholanolone) for at least 8 months post-injection. Similar results110 were obtained with men who had received single-dose administration of T enanthate, propionate or undecanoate, or nandrolone and its decanoate. Such ¢ndings contrast with those by Gaillard et al.,111 who analysed hair and urine collected from racing cyclists for amphetamines, corticosteroids and AASs using GC/MS (by positive chemical ionization), highperformance liquid chromatography/ MS/MS and GC/ MS/MS, respectively. The authors found that, for amphetamines and AASs, the percentage of positive ¢ndings in urine was less than for scalp hair; for AASs, none was detected in urine (out of 30 analyses) whereas two were detected in hair (out of 25 analyses). Although these results should be interpreted with caution, such ¢ndings appear to support hair as a useful matrix for analysis of AASs. Generally, physiological levels in hair for T and DHEA are in the range 1-10 pg/mg hair, while the limit of detection for nandrolone is approximately 1pg/mg hair. Kintz et al.,109 make the important point that, given current methodologies, hair of athletes would be far more likely to test positive for AAS abuse if they have

Anabolic steroids in sport

been chronically administered. As research into hair analysis has progressed, at least two important factors have been identi¢ed which should be given serious consideration. Inuence of hair colour on analytical results. Melanin is the major (85-93%) constituent of hair104 and the principal component for binding of drugs. Both phaeomelanin and eumelanin are present in hair, with black or brown hair containing greater quantities of the latter form of melanin. It is not surprising, therefore, that dark hair binds larger amounts of drugs than does red and blonde hair.112 In this context, several authors103,104,109 have highlighted the very real possibility of racial bias when African ^ American drug abusers might be identi¢ed more frequently than Caucasians. Stability of drugs in human hair. Clearly, this factor is of crucial importance in the interpretation of drug analyses. Several recent studies have shown that dyeing, and especially bleaching, of hair can reduce the content of numerous drugs, including opiates, cocaine, cannabis, nicotine,T and its undecanoate and nandrolone.111,113 These results have important consequences; ¢rst, Kidwell114 has suggested the possibility of incorporating dyes to reduce the binding of drugs in dark hair, and thus to minimize the inequity problem noted above. Secondly, the routine colouring or bleaching of their hair by athletes (whether for aesthetic or drug-evasion purposes), can lead to missing identi¢cation of, for example, T and other AASs.109 A particular problem occurs when athletes shave their heads; the only recourse the analyst then has is to ask for hair samples to be taken from other parts of the body ^ hopefully without too much inconvenience. However, much more information is needed on the concentrations of AASs or their metabolites in axillary or pubic hair.

335

Although quadrupole GC/MS instruments are currently being used by all IOC laboratories for screening purposes, the IOC now requires that, for low concentrations of selected AASs, analytical methods must be capable of reaching `the 2 ng/mL detection limit’. The IOC has requested that high-resolution mass spectrometry (HRMS) and tandem MS (MS/MS) are used by the accredited laboratories for screening to increase the limit of detection of selected AASs (see Olympic Movement Anti-Doping Code17). GC/HRMS was employed for the ¢rst time for screening at an Olympic Games in 1996 (Atlanta).

Sample preparation A £ow chart of sample preparation employing mixedphase extractions and enzymic hydrolysis is given in Fig. 7. In the past, a separate method for the isolation of certain AASs and their metabolites that are largely excreted unconjugated (free steroids) from urine was employed, but these steroids can be now adequately screened for without this additional procedure being necessary. However, extraction of free steroids, in

Screening for xenobiotic AASs in urine Introduction Since the early 1980s, IOC-accredited laboratories switched from using RIA to screening by benchtop quadrupole GC/MS because of the large improvements that these instruments o¡er in speci¢city, sensitivity and data handling, together with a reduction in cost. Increased chromatographic resolution was obtained with the use of superior capillary columns, and increased sensitivity was achieved by using electron impact MS in the selected-ion monitoring (SIM) mode. In addition, with automated sample injection and short chromatographic run times (typically 20 30 min because of oven temperature programming), large sample throughput made low-resolution GC/MS the preferred analytical tool.

Figure 7. Schematic (generic) of sample preparation for the screening of anabolic-androgenic steroids and their metabolites from urine. Ann Clin Biochem 2003; 40: 321–356

336

Kicman and Gower

which liquid- liquid extraction of urine is followed by derivatization, is still used for con¢rmation of certain steroids (e.g. for analysis of £uoxymesterone). Internal standards are added to the urine. Our laboratory routinely employs a pure trideuterated mixture (16,16,17a-2H) of T, epitestosterone, DHT and 5a-androstane-3 a,17 b-diol. This eliminates error due to sample extraction and variability in e¤ciency of steroid derivatization (see below). As many AASs or their metabolites are excreted primarily as glucuronides in the human, equivalent deuterated internal standards should ideally also be conjugated, but these, together with other certi¢ed AAS reference materials, have only relatively recently become commercially available. The National Analytical Reference Laboratory (NARL) within the Australian Government Analytical Laboratories (AGAL) is accredited for the preparation and characterization of steroid reference materials in accordance with ISO Guide 34 for Reference Material production. The materials are supplied with a comprehensive analysis report. Their characterization and certi¢cation has been reviewed and approved by an independent scienti¢c panel. To date, 60 anabolic steroid reference materials have been prepared and characterized. Steroid glucuronides are not volatile so, prior to GC/MS analysis, the conjugate must be cleaved. Enzymic, rather than acid, hydrolysis is chosen because, although the latter o¡ers speed and simplicity, it can also generate non-speci¢c interference and steroid dehydration products. In sports drug testing, much emphasis is placed on measuring trace amounts of steroid analytes and any unpredictable interference on a GC/MS chromatogram can hinder interpretation. IOC laboratories prefer to perform a preliminary clean-up to optimiz e the conditions for enzymic hydrolysis; a clean-up removes enzyme inhibitors such as lactones and ascorbic acid which are often present in urine. Initially, solid- liquid extraction is used, employing either C8 or C18 cartridges, which have generally superseded Amberlite XAD-2 resin or equivalent; methanol is used to elute the free and conjugated steroids. After evaporation, the dried extract is resuspended in a suitable bu¡er at the

optimal pH for the enzyme preparation, usually an extract from Helix pomatia or Escherichia coli. The digestive juice of H. pomatia is a rich source of steroid glucuronidase and aryl sulphatase, the latter being of no practical value as aryl steroids are minor metabolites of AASs. H. pomatia is used by some laboratories, rather than a pure glucuronidase preparation, because it is more economical to do so and because some believe that its limited alkyl sulphatase activity is of value. However, it is important to be aware that artefacts can be produced using preparations of H. pomatia, including the potential formation of T in urine.115^117 Following deconjugation, liquid- liquid extraction is performed with a non-polar organic solvent, usually diethyl ether or t-butylmethyl ether, the neutral steroids being favourably partitioned into this. The organic layer is evaporated to dryness, and the extract further dried in a vacuum desiccator containing phosphorous pentoxide and potassium hydroxide to ensure removal of all water present (which can prevent complete derivatization, see below); the steroids are then derivatized.

Derivatization For screening purposes, steroid hydroxyl and oxo functions are converted to TMS ether and enol derivatives, respectively, to render them more volatile and thermally stable and to improve their chromatographic characteristics. The derivatization mixture used consists of N-methyl-N-trimethylsilyltri£uoroacetamide (MSTFA) with a catalyst, iodotrimethylsilane (ITMS), such a reaction yielding isomerically pure TMS enol derivatives.118 The reaction is shown in Fig. 8. As ammonium iodide (NH 4I) is more stable under storage than ITMS, NH4 I can be reacted with MSTFA to generate fresh ITMS in situ, but an antioxidant (e.g. ethanethiol) must then be included in the mixture to suppress the formation of iodine.119 Mass spectra of the TMS enol and TMS ether derivatives of some AASs may show one signi¢cant ion. This can enhance sensitivity for screening purposes (e.g. 6 b-hydroxymethandienone), but it is of limited diagnostic value for con¢rmation. In such cases, for con¢rmation, TMS ether derivatives are formed

Figure 8. Silylation resulting in formation of a trimethylsilyl (TMS) enol TMS ether derivative. The example shown is a diagnostic metabolite of mesterolone. Ann Clin Biochem 2003; 40: 321–356

Anabolic steroids in sport

337

instead; i.e. only the alcoholic hydroxyls undergo derivatization, using a reaction mixture of MSTFAimidazole (100:2, v/v).

Screening by GC/MS Screening is performed by GC/MS, supplemented with GC/HRMS and/or GC/MS/MS. With respect to chromatography, the TMS derivatives of AASs separate well on fused-silica capillary columns (20-30 m), using cross-linked methylsilicone as the stationary phase and helium as the mobile phase. Rapid separation of components is based on the use of oven temperature programming and detection with a lowresolution quadrupole mass ¢lter mass spectrometer. The mass spectrometer is operated in the SIM mode, up to four ions being monitored for the screening of each analyte;120^122 examples are shown in Fig. 9. The limit of detection, and in some cases quanti¢cation, for most AASs can be as small as 1 mg/L (*100 pg on-column) using SIM GC/MS, but this will vary depending on the nature of the urinary matrix and the degree of fragmentation of respective analytes. Although GC/MS continues to be used in IOCaccredited laboratories, screening for AASs is supplemented by the application of more sophisticated MS, i.e. HRMS (currently using magnetic sector instruments) or MS/MS.123 The use of GC/magnetic sector MS for screening of analytes is not a new concept; it has already been used for routine analysis of dioxins in the environment. In 1993, Horning and Donike124 showed the potential of HRMS for the screening of AASs, at moderate resolution in the SIM mode. Resolution in terms of MS may be de¢ned as the ability to distinguish two ions with a small mass di¡erence; i.e. M/ ¢M, where M is the mass of the ion being detected and ¢M is the di¡erence between that mass and an adjacent mass in a mass spectrum. Two peaks of similar intensity are considered to be resolved if the valley between the two peaks is equal to 10% when using magnetic sector instruments (and 50% when using quadrupoles). For example, a GC coupled to a magnetic sector instrument operating at a resolution of 5000 can monitor an ion of mass m/z 500 ¢0 atomic mass units (amu) and thus distinguish it from a coeluting chromatographic interferent with an m/z of 500 ¢1. An example showing the mass resolution required to separate a nandrolone metabolite from a vitamin E metabolite is given by Mueller et al.125 The term screening by HRMS, as used within the sports community, is somewhat of a misnomer as these instruments are operated at moderate resolution (3000 -5000) rather than high resolution (i.e. 410 000). Nonetheless, this `moderate’ resolution is far superior to the unit mass resolution ( +0¢5 amu) obtained with quadrupole mass ¢lters, thus o¡ering

Figure 9. Selected ion detection of the trimethylsilyl derivatives of oxymesterone, oxymethelone, bolasterone and bolasterone metabolite (M1) excreted in urine. In each ‘mini’ ion chromatogram, the abscissa is time (min) and the ordinate is abundance (arbitrary units). These examples are taken from Fig. 5 of the article by Donike et al.,122 reproduced with permission from the International Athletic Foundation. Copyright & 1988. Ann Clin Biochem 2003; 40: 321–356

338

Kicman and Gower

considerable improvement in the limit of detection compared to quadrupole MS equipment, with identical sample preparation and GC conditions. Moderate resolution, rather than high resolution, is chosen for screening purposes because a compromise has to be made between su¤ciently eliminating other interfering ion signals compared to the ion mass of choice whilst maintaining an adequate signal for detection purposes. In a direct comparison, Kokkonen et al.126 concluded that GC/HRMS was 2-10 times more sensitive for detecting metabolites of methandienone compared to quadrupole GC/MS, the gain in detection sensitivity depending on the analyte (curiously, despite supposedly identical column conditions, di¡erent retention times on the SIM chromatograms between the two techniques are displayed in their paper). The superiority of GC/HRMS as a routine screening technique, especially for metabolites of stanozolol and methandienone, became apparent when there was a large increase in the number of positive samples identi¢ed by the IOC-accredited laboratory in Cologne. In 1995, out of 6700 samples tested by this laboratory, 116 positives were reported of which 41 were detected by quadrupole GC/MS, whereas an additional 75 were identi¢ed solely by their GC/HRMS screen.127,128 As a consequence of these ¢ndings, the IOC place much greater emphasis on the detection of certain anabolic agents at low concentrations, i.e. the metabolites of the AASs, nandrolone, methandienone, 17 a-methyltestosterone and stanozolol, together with the b2-agonist clenbuterol. IOC-accredited laboratories target these speci¢c analytes so that they can be detected at a urinary concentration as low as 2 ng/mL; a corollary of a presumptive positive is that these laboratories must be capable of con¢rming the presence of these compounds at this low concentration (see `Con¢rmatory analysis’, below). For many IOC-accredited laboratories, the high cost of screening by HRMS, and the requirement of skilled operators to achieve the best results, currently preclude them from using such instruments. To ful¢l the request made by the IOC, such laboratories have chosen to use tandem MS (MS/MS), as theoretically a similar enhancement in limit of detection might be expected by monitoring a selected product ion from collision-induced dissociation of a precursor ion. Tandem MS can be performed either in space using a sequential set of quadrupole rods or in time using a quadrupole ion trap (QIT), the latter being a threedimensional analogue of the linear quadrupole mass ¢lter. The QIT has no advantage, in terms of improved sensitivity, over quadrupole mass ¢lters for comprehensive screening of AASs by SIM, despite the use of selected-ion storage,129 and this appears to be the case Ann Clin Biochem 2003; 40: 321–356

with QIT MS/MS (of course, improved sensitivity can be gained by targeting a limited number of metabolites of AASs that yield high-intensity precursor ions130). In contrast, GC/HRMS can o¡er such an advantage within a 10 -20% mass shift (e.g. m/z 500 - 400), budget and expertise permitting.

ConŽ rmatory analysis When a presumptive positive is found by screening, additional tests are performed with the derivatized extract, usually by subjecting it to analysis by `full’ scan GC/MS. If the data are consistent with a positive ¢nding, a second aliquot is taken from the A-sample and the analytical procedure is repeated, in keeping with the Laboratory Procedures in the Olympic Movement Anti-Doping Code17 (Appendix D, part 2.1.2). If appropriate, the sample extraction procedure can be modi¢ed to obtain a cleaner and more concentrated extract, as can the derivatization procedure for characterization purposes, although the latter is seldom necessary for con¢rmation of steroid identi¢cation. Criteria for con¢rmation are a concordance in retention time and scan data compared with those of reference urine, a minimum of three diagnostic ions being mandatory. The reference urine is prepared by spiking a reference standard into a negative (blank) urine that has been shown not to contain any substances of interest. When a reference standard is not available for addition, the data must be compared with those of appropriate elimination urine collected in a controlled study, with supporting documentation to that e¡ect. There can be a loss of sensitivity when a quadrupole GC/MS is operated in the scan mode, although this can be redressed in part by con¢rmation using a QIT. Figure 10 shows background-subtracted mass spectra of the bis-TMS derivatives of a metabolite of methenolone, m/z 446 being the molecular ion and m/z 208, 195 and 179 being the fragment ions. In this investigation,131 a criterion for a sample to be considered positive was that all four characteristic ions had to be a minimum of three times above background noise for m/z range of 150 -550. Another criterion was that positive con¢rmation of methenolone required concordance in retention times compared to the reference standard. Matching the retention time to a reference standard is extremely important because some analytes can give identical spectra; for example, Goudreault and Masse132 found that the 17 a-epimer of methenolone, a metabolite that is present in minute amounts, gives an identical mass spectrum to that of methenolone but a di¡erent retention time. The same is the case for most of the corresponding 17-epimers of 17a-methyl steroids,133 whether underivatized or TMS-derivatized.

Anabolic steroids in sport

339

Figure 10. Background-subtracted full-scan mass spectra of the trimethylsilyl (TMS) enol TMS-ether derivatives of methenolone (top) and its major metabolite (bottom), 3a-hydroxy-1-methylen-5a-androstan-17-one in a urinary extract. The insets show the respective chemical structures (underivatized). Figure from Kicman et al.,131 reproduced with permission of the American Association for Clinical Chemistry, Inc. (AACC). Copyright & 1994. With analysis of lower concentrations of AASs, the data may not be of su¤cient quality. In such cases, the use of SIM for con¢rmation is permissible, but adequate criteria for identi¢cation need to be met. To

help achieve international consistency in reporting, a working group of directors of IOC met `to harmonize, in a defensible way, the analytical part of reporting for low concentrations of anabolic steroids, when using Ann Clin Biochem 2003; 40: 321–356

340

Kicman and Gower

the recently implemented techniques of HRMS and MS/ MS’. As a result, a document was circulated to all accredited laboratories entitled `A nalytical Criteria for Reporting Low Concentrations of Anabolic Steroids (August 1998). This document is now in the public domain, appended to the publication of the Nandrolone Review.134 The criterion for chromatography is that the retention time or relative retention time of the analyte should not di¡er by more than 1% from that of the same substance in the positive control urine analysed in the same batch unless the shift can be explained (e.g. by column overloading). Perhaps surprisingly, the criteria for MS are the same for low and high mass resolution, where evaluation of a single spectrum must include a minimum of three diagnostic (quali¢er) ions. If three ions are not available then a second derivative should be prepared or a second ionization or fragmentation technique should be used. For MS/MS, the minimum of three ions may include the precursor ion. For low-resolution MS and HRMS, the relative abundance of any of the ions must not di¡er by more than 5% (absolute) or 20% (relative), whichever is greater, from that of the positive control urine, whereas for MS/MS the amount is 10% and 25%, respectively. In the future, HPLC/MS/ MS may be useful for con¢rmation (and possibly screening) of AASs.135^140 The direct measurement of steroid glucuronides (and sulpho-conjugates) in urine is attractive but analysis is not simple. Considerable research and development are required to evaluate the potential application of HPLC/MS/MS and whether it can be used as a comprehensive tool for analysis of AASs.

Estimating time of administration Following declaration of an adverse ¢nding, the analyst is sometimes required by a Drug Advisory Committee to give an opinion of the time, dose and route of administration, for evidential purposes. It is useful also to gauge how e¡ective a drug test may be by estimating the degree of retrospective detection following administration. Unfortunately, the pharmacokinetic parameters of many xenobiotic AASs and their metabolites have not been determined, and even urinary excretion rates against time have not been extensively studied. If these data were readily available, and samples had been collected from an athlete at regular intervals following a positive ¢nding, interpretation is still confounded by the variability in dilution associated with single-pass urine specimens. To compensate for variability, creatinine correction is not advocated because athletes use creatine as a `dietary supplement’. Furthermore, a number of studies on animals, children and adults have shown that creatine metabolism and creatine (or creatinine) Ann Clin Biochem 2003; 40: 321–356

excretion is a¡ected by the administration of AASs.141 Adjustment by urine -speci¢c gravity is a crude and unestablished technique and is not employed (except when adjusting the `relative signal’ of a positive control urine for the nandrolone metabolite, 19-norandrosterone, when this analyte is detected in a sample but the speci¢c gravity of the sample is 41Í020; for more information, see Ref. 134). Despite all of the above, many are still interested in an opinion as to what length of time will pass before an individual may test negative following cessation of administration. As a very rough guide, a user of orally active 17 a-alkylated steroids may test negative as little as a week following withdrawal, using the conventional approach of GC coupled to a quadrupole mass spectrometer (see `Screening by GC/MS’, above). In contrast, adverse ¢ndings for nandrolone metabolites can occur many months following a single intramuscular administration of nandrolone decanoate, the formulation being speci¢cally designed for slow release of the esteri¢ed drug (see `Nandrolone’, below).

Detection of doping with ‘natural’ steroids General aspects In the context of this section, the term `natural’ refers to steroids that are structurally identical to those produced endogenously (e.g. T, DHT, DHEA and androstenedione; see Fig. 2). Other workers prefer the term `endogenous’ steroids, although it is a misnomer because the steroids administered are by de¢nition from an exogenous source; hence, the term `pseudoendogenous’ has been proposed recently. In the UK, public attention has recently focused on the `nandrolone problem’ with the spate of adverse ¢ndings in 1999 and an inquiry was undertaken (the results of which were published in the Nandrolone Review134). In the past, the presence of 19-norandrosterone in urine was correctly interpreted as evidence of administration of nandrolone. Although the possibility was previously considered, only with the recent application of GC/HRMS, and also current reports in the literature, using puri¢cation procedures prior to analysis by quadrupole GC/MS, has indicated that trace amounts of 19-norandrosterone are naturally excreted by some men and non-pregnant women (see `Nandrolone’, below). Applying laboratory standard operating procedures, these trace amounts are far too small for full scan data to be of su¤cient evidential quality to prove the presence of this metabolite. Nonetheless, as the use of SIM is permissible for con¢rmatory analysis and will be required for identi¢cation of trace amounts, it is important to consider the possible sources of 19-norandrosterone. In adult women, and possibly in men, a source could be

Anabolic steroids in sport

from endogenous nandrolone production; hence nandrolone is discussed in this section.

Nandrolone In vitro studies in the 1960s on the human ovary142 and prostate143 suggested that 19-nor steroids may play an intermediate role in the aromatization process.144,145 In the 1980s, GC/MS data showed that nandrolone was present in human follicular £uid146 and in plasma throughout gestation.147 In the UK, in the 1980s, an immunoassay for the screening of nandrolone metabolites in sports competitors’ urine was applied, and it was known that raised assay values could result because of pregnancy.12 The `decision limit’ chosen for the immunoassay screen was high (100 ng/mL) compared with the amounts detectable using present-day analysis by GC/MS (*2 ng/mL). The immunoassay was devised as a general screen and the high cut-o¡ chosen then was su¤cient for the purpose. At that time, 19-norandrosterone was viewed as exclusively a xenobiotic steroid by those in IOCaccredited laboratories, so these results highlighted that the approach of increasing the sensitivity of doping control methods might increase the possibility of detecting endogenously produced nandrolone metabolites. In the 1990s, preliminary reports by those working in IOC-accredited laboratories concluded that 19-norandrosterone was present in urine collected during pregnancy. This conclusion was based on observations of an increased SIM GC/MS signal (low resolution) and/or diagnostic product spectra following GC/MS/MS;148^150 a current investigation has demonstrated that it is not uncommon for the concentration of 19-norandrosterone to exceed 5 ng/mL during pregnancy.151 With the application of GC/HRMS for doping control it became apparent that low concentrations of 19norandrosterone in urine may occur in untreated men and non-pregnant women. Subsequently, three studies152^154 showed that the urinary concentration in untreated men is 50Í6 ng/mL (two of the studies using quadrupole GC/MS following puri¢cation procedures). Full scan spectra were not presented, presumably because of the di¤culty in analysing such low concentrations. Nonetheless, the study by Le Bizec et al.153 met the criteria for reporting low concentrations of AASs by providing a minimum of three diagnostic ions (m/z 420, 405, 315; these ions correspond to the bis-TMS derivatives of 19-norandrosterone). For these three studies the test population was small (n=47 in total), was not a representative racial mix and was composed mainly of non-athletes. Further useful but unpublished data (to date) came from the 1996 Winter Olympics in Nagano, the preliminary results showing that, of the 370 male competitors tested,

341

only ¢ve showed urinary concentrations of 19norandrosterone 40Í1 ng/mL and no concentration exceeded 0Í4 ng/mL; of the 251 female competitors tested none excreted more than 5 ng/mL (communication from Dr M Ueki, Director of the IOC Accredited Laboratory in Japan, to the Expert Committee on Nandrolone, the committee being commissioned by UK Sport; see Ref. 134). Notably, it cannot be excluded that some competitors were using nandrolone, which would positively bias the data. These data support the minimum reporting concentration for IOC-accredited laboratories of 2 ng/mL for 19-norandrosterone in urine from male subjects and 5 ng/mL in urine from the non-pregnant women (compared to control urine). However, the database potentially available from analysis of sports samples needs to be broadened and it is highly desirable that such data are brought into the public domain for peer review. Furthermore, in untreated healthy men there is no direct evidence to date that nandrolone is produced endogenously but, notwithstanding, the urinary excretion rate of 19-norandrosterone increases under stimulation with human chorionic gonadotrophin.155 The increase was small, resulting in urinary nandrolone metabolite concentrations of 51ng/mL as measured by GC/MS following hydrolysis with b-glucuronidase and extensive puri¢cation. Le Bizec et al.153 and Saugy et al.156 have demonstrated that the urinary concentration of 19-norandrosterone may be increased after exercise; the study designs, however, were inappropriate to support the hypothesis that a concentration of 19-norandrosterone above the laboratory reporting threshold can arise as a result of endogenous production. Subsequent studies (all on male subjects) have shown that urine collected after exercise can have higher concentrations of 19-norandrosterone,157,158 but this may be simply the result of dehydration rather than induction of nandrolone secretion.159 Recently, Le Bizec et al.160 have provided preliminary results that 19-norandrosterone formed from nandrolone administration is exclusively conjugated to glucuronic acid, whereas a proportion (*30%) of that produced from an endogenous source is sulpho-conjugated. Although these results are potentially of interest, the scienti¢c underpinning of this study is severely lacking; i.e. no data concerning possible variation in e¤ciency of hydrolysis of sulpho-conjugates and extraction are provided, the nandrolone was administered to a small number of volunteers, the dose was tiny (5 mg) and by an oral route only. Analysis employing carbon isotope MS, which can be used to distinguish between exogenous and endogenous sources of steroids in urine (see `Detection of ``natural’’ androgen administration’, below), has been applied to 19-norandrosterone.161,162 However, Ann Clin Biochem 2003; 40: 321–356

342

Kicman and Gower

even with speci¢c immunoa¤nity chromatography before analysis to optimize sample preparation, improvements in instrument sensitivity by at least a factor of 10 is required for successful analysis of samples containing 19-norandrosterone near the chosen reporting concentration for IOC-accredited laboratories. Other sources of nandrolone may be in trace amounts contained within poor-quality processed meat (from the injection site of animals illegally treated with nandrolone), and the athlete is advised to avoid such products; also, the o¡al from the boar163 and horse should likewise be avoided as these species produce signi¢cant amounts of nandrolone endogenously. Apart from nandrolone, administered steroids that can be converted to 19-norandrosterone include norethisterone and so-called `dietary supplements’ that contain 19-norandrostenedione and/or 19norandrostenediol. Norethisterone is a 19-norprogestogen that is used as a contraceptive in women and for the treatment of menstrual dysfunction; a major metabolite is tetrahydronorethisterone and 19-norandrosterone is a minor one. The presentation of cases in which the urinary concentration of 19-norandrosterone exceeds 5 ng/mL, in the presence of tetrahydronorethisterone, are rare and are investigated on an individual basis to examine whether the ¢nding is compatible with contraceptive medication. Dietary or nutritional supplements that contain the `prohormones’ 19-norandrostenedione and 19norandrostenediol are designed as oral formulations, a typical `recommended’ dose being around 100 mg daily. Currently, there are no papers described in scienti¢c journals showing that ingested prohormones are metabolized to nandrolone, although an article in the proceedings of a workshop supports this possibility.164 In a quasi-legal context, some experts have commented that there was ambiguity as to whether 19-norandrostenediol and 19-norandrostenedione were included or not in the IOC Prohibited Classes of Substances (Class C, Anabolic agents) under `A nabolic-androgenic steroids and related substances’. Any possible ambiguity was removed from 31 January 1999, when these compounds were named speci¢cally in Appendix A of the Olympic Movement Anti-Doping Code.17 The pharmacokinetic properties of nandrolone have been determined after intramuscular administration of the drug as the phenylpropionate and decanoate ester.165^167 Formulations of esteri¢ed nandrolone are designed for depot activity, being slowly released from the arachis oil vehicle into the general circulation, where cleavage is performed by blood esterases. The mean absorption and elimination half-lives of nandrolone following intramuscular injection of the Ann Clin Biochem 2003; 40: 321–356

decanoate ester are quoted in days.165^167 The release rate of this esteri¢ed drug and the time of maximum concentration reached after administration varies depending on the injection site and the injection volume; for example, for a 100-mg dose in a 1-mL injection volume, a mean (standard deviation) absorption half-life of 12Í0 (0Í9) days has been measured following administration into a deltoid site compared to 7Í7 (0Í6) days into a gluteal site.167 In contrast, after oral administration of the `prohormone’ steroids, the major proportion of the dose will be converted by ¢rst-pass metabolism into inactive tetrahydro-17-oxo metabolites, conjugated with glucuronic acid, and then rapidly eliminated. Following a 50-mg dose of 19-norandrostenedione, urinary concentrations of 19-norandrosterone exceeded 100 000 ng/mL in the ¢rst urine voids,35 whereas intramuscular administration of 50 mg of nandrolone decanoate resulted in concentrations of 250 -750 ng/mL in the ¢rst week.92 Excessively high urinary concentrations of 19-norandrosterone are indicative of oral administration of a `prohormone’. However, with lower concentrations (e.g. 5100 ng/mL), it is impossible to distinguish the source, although 19-norandrosterone from metabolism of a precursor steroid is likely to have an apparent elimination halflife (in terms of urinary excretion rate) of only a few hours.

Testosterone (T) Application of hormone ratios A test based on determining whether a urine concentration of Texceeds the upper limit of a reference range would be insensitive because of the wide variability in excretion associated with a single-pass urine collection. To overcome the problem, Brooks et al.10 introduced the concept of the hormone ratio in 1979, the use of a ratio being considered to be independent of urinary £ow rates. The ratio of T to luteinizing hormone (T/LH) was originally proposed, but this necessitated two separate assay procedures being performed and, in retrospect, immunoprocedures are generally accepted as not having the discriminatory power of MS for evidential analysis.168 In1982, the test adopted by the IOC for detection of T administration was based on the GC/MS determination of the ratio of T to its 17a-epimer, epitestosterone (EpiT), following glucuronide hydrolysis169 (T/EpiT ratio; often referred to as the T/E ratio). The T/EpiT decision limit was derived empirically from an observed distribution of measurements in specimens collected from a large number of individuals. In healthy men and women, the median T/EpiT ratio approximates unity, but supraphysiologica l doses of T cause an increase in the ratio as a result of increased excretion of T (see Fig. 11),

Anabolic steroids in sport

343

Figure 11. Ion chromatogram (m/z 432) of the trimethylsilyl (TMS) enol TMS ether derivatives of testosterone (T) and epitestosterone (EpiT) in a urinary extract with a T/EpiT ratio 46. Good chromatography is important for the resolution of the adrenal steroid metabolite, 11b-hydroxyandrosterone (11-OHA), from T. the laboratory reporting threshold chosen being T/EpiT=6.

Endogenous sources of EpiT The T/EpiT ratio may be augmented as a consequence of dose -dependent inhibition of testicular steroidogenesis. The testis secretes EpiT (as well as a proposed biosynthetic precursor, androst-5-ene-3 b,17 a-diol170), overall contributing to approximately 95% of the pool of urinary EpiT glucuronide in eugonadal men.171 When supraphysiological doses of T are taken, suppression of LH secretion decreases the urinary excretion of EpiT glucuronide;172^174 for example, intramuscular administration of 200 mg of T enanthate weekly for 16 weeks decreases urinary EpiT to 510% of pre-treatment values.174 In women, the regulation of EpiT production is not known but, as EpiT has been identi¢ed in follicular £uid,175 some augmentation of the ratio may occur through suppression of ovarian steroidogenesis. Adrenal stimulation can cause an increase in EpiT production but this appears to be countered in men by a concomitant decrease in testicular steroidogenesis, probably as a result of induced hypercortisolaemia. Hence, stressinduced changes in adrenal and testicular steroidogenesis result in no change or only a small decrease in the urinary T/EpiT ratio in eugonadal men.171

Increases in T/EpiT ratios The IOC Medical Code stipulates that, in the case of T/ EpiT 46, it is mandatory that the relevant medical authority conducts an investigation before the sample is declared positive. Usually it is concluded that surreptitious T administration has occurred, but occasionally the athlete may have a physiologically increased ratio,176^181 being a `natural biological outlier’. In addition, the possibility of a pathological condition (e.g. a T-secreting tumour) accounting for an augmented ratio in a sports competitor must not be neglected, although there is no such case report described in the scienti¢c literature. Investigations include a review of T/EpiT results from previous tests, subsequent tests and also any results of endocrine investigations. In the event that previous T/EpiT results are not available, the athlete should be tested unannounced at least once a month for 3 months.

Intra-individual proŽ ling of T/EpiT ratios With an adverse ¢nding, investigating the T/EpiT results from previous and subsequent tests; i.e. assessing the T/EpiT intra-individual (within-subject) variability is useful in determining whether an o¡ence has occurred. The application of T/EpiT intraindividual pro¢ling was ¢rst discussed by Donike et al.182 In a subsequent article,183 the statistical test proposed by Harris184,185 was applied for biochemical Ann Clin Biochem 2003; 40: 321–356

344

Kicman and Gower

analyses to assess the suitability of using such data. This statistical test evaluates whether it is appropriate to use an intra-individual (subject-based) reference range, as opposed to an inter-individual (populationbased) reference range, for assessing changes in biochemical status. However, to date, there are very limited data on intra-individual variation of T/EpiT ratios presented in the peer-reviewed literature. In their article on detection of T and xenobiotics, Catlin et al.186 have reviewed the data on intra-individual variability. They present their criteria for determining whether T doping has occurred in men, based on T/ EpiT ratio data from drug-free males who showed an intra-individual coe¤cient of variation (CV) 560% (variation from the collection of three or more samples of urine taken at monthly or longer intervals). In contrast, they report an example of a case of a T user, sampled four times, whose initial T/ EpiT was 8Í2, a mean of 3Í0 and a CVof 114% (see Fig. 12). This pattern was considered to be typical of an individual who is caught and then discontinues T administration. In these authors’ experience, most T users who provide three or more urine samples have a CV 460%. However, those with CV 560% and T/EpiTof 6-10 are tentatively classi¢ed as `naturally increased’. The global incidence of athletes who fail a T/EpiT test but fall into the `tentative’classi¢cation (CV 560%) is not known, but a very large proportion was reported in Sweden;181 27 of the 28 individuals had a CV 543%; no action was taken because it was not possible to undertake further investigations at that time. Hence, what may be required is to determine the likelihood that a particular T/EpiT ratio falls outside the range of values expected in an individual rather than assigning

a`blanket’ CV threshold of 60% (Dr L Bowers, personal communication). Of interest is whether there is a di¡erence in intra-individual variability depending on whether samples are collected over a matter of days compared to months. As a logical ¢nal step, it could be argued that there is little need for a reporting T/EpiT threshold of 6, that is based on inter-individual variation, if each individual’s variation in T/EpiT ratio is stored on an international database. Additional studies could re¢ne the value of T/EpiT intraindividual pro¢ling, as opposed to recommending immediate clinical investigation to an athlete whose sample has a T/EpiT 46.

Ancillary tests to detect T administration As noted above (see previous section), ancillary investigations can include an invasive test to di¡erentiate between natural T/E outliers and T users by administering the pharmaceutically licensed antifungal agent, ketoconazole.177,178 This drug in the appropriate dose inhibits the cytochrome P450 17a-hydroxylase/ 17,20-lyase, the enzyme that converts pregnenolone to dehydroepiandrosterone and progesterone to androstenedione, these products being intermediates in the steroidogenic pathway to T. In eugonadal men, including natural outliers, ketoconazole administration results in a large and acute decrease in the excretion rate of urinary T, which contrasts with a much smaller decrease in the rate of EpiT excretion, so the T/EpiT ratio decreases considerably. Conversely, in control subjects who have received T, ketoconazole does not cause a decrease in the T/EpiT ratio because the origin of T is from the exogenous source. The ketoconazole test o¡ers athletes a rapid means of

Figure 12. Testosterone/epitestosterone (T/E) proŽles of three athletes sampled four times over 170 days on relatively short notice. The T/EpiT ratio of the Žrst sample from athlete A (Žlled bars) was 8¢2, followed by subsequent T/EpiT ratios of 1¢2, 1¢3 and 1¢4 (mean 3¢0, CV ˆ 114%). The T/EpiT ratios from athlete B (hatched bars) showed minimal variability (mean ˆ 0¢97, CV ˆ 9¢8%); athlete C (open bars) showed greater variability (mean ˆ 2¢3, CV ˆ 40%) but gave ratios still within the laboratory norms for non-T-using controls. Figure from Catlin et al.,186 reproduced with permission from the American Association for Clinical Chemistry, Inc. (AACC). Copyright & 1997. EpiT, epitestosterone; CV, coefŽcient of variation. Ann Clin Biochem 2003; 40: 321–356

Anabolic steroids in sport

proving their innocence, and it has been applied with success in several European countries. However, the test has not been adopted internationally because some questions remain as to whether an individual’s rights may be compromised by being coerced into taking an antifungal drug to prove that they are not using T. Other tests to distinguish T/EpiT ratio outliers have been described. One approach is based on the determination of serum 17-hydroxyprogesterone/T,187,188 but this method is yet to be generally established. Another approach is based on taking into account sulphate conjugates of T and EpiT, in conjunction with measurement of the aglycones,189 but subsequent data gained by HPLC/MS/MS analysis did not support the usefulness of this strategy.139 An abnormally large T/LH ratio accompanying a high T/EpiT ratio is indicative of a T user (in addition, there is some evidence to support that the T/LH ratio is a more sensitive retrospective marker of chronic administration of T than the T/EpiT ratio190). Urinary LH measurement between laboratories is yet to be standardized although some laboratories have determined their own T/LH reference range, and exercise stress appears to attenuate rather than augment the ratio.191 Currently, there is much interest in the application of carbon isotope ratio MS (see `Detection of ``natural’’ androgen administration, below) and this procedure appears to be the most promising way forward.

Detection of administration of other ‘natural’ androgens 5a-Dihydrotestosterone (DHT)

It was predicted that some AAS users would switch to administering DHT in an attempt to beat the tests for T administration, as administration of this potent androgen would not perturb the T/ EpiT ratio.192 Subsequently, a number of samples collected at the 1994 Asian Games were found to have an abnormal urinary steroid pro¢le, consistent with 5 a-DHT administration (see`Fifty years of anabolic steroids and sport’, above). Tests developed to detect DHT adminis tration in men were based on the urinary concentration of DHT/EpiT, with secondary markers being that of the ratios of the DHT metabolite, 5a-androstane-3 a,17b-diol, to EpiT, 5 b-androstane-3 a,17b-diol and LH,193^195 The ratio of androsterone to etiocholanolone is also augmented following DHT administration,194 but this appears to be a less sensitive marker than the other ratios selected.195

DHEA, androstenedione and androstenediol (prohormones) Detection methods based on abnormal urinary concentration of DHEA, androstenedione and androstenediol or their metabolites may be successful

345

during periods of administration. However, following cessation of administration, the decrease to basal concentrations will be rapid, which means that such tests will lack sensitivity in terms of retrospective detection. The application of carbon isotope ratio MS may be useful for improving retrospective detection of administration (see next section). With administration, changes in the T/ EpiT ratio may occur, but in men this is probably due more to hepatic metabolism of these prohormones to T glucuronide than to an increase in circulating T. Results from preliminary studies, which have been performed exclusively with male volunteers, show that, although the T/EpiT ratio may be augmented with modest doses of DHEA or androstenedione, following administration there is a wide variability in response and the reporting threshold (T/Epi=6) is generally not exceeded. In women, it would be expected that the T/ EpiT ratio would be a better marker of administration due to the large proportion of T produced from peripheral conversion of the administered prohormones (see`E¡ects on muscle size and strength’, above). For screening of DHEA administration, a urinary concentration threshold of 300 mg/L of DHEA glucuronide has been proposed but, based on this threshold, a single replacement dose (50 mg) can be detected for only up to 8 h post-administration.196 This study, together with another in which DHEA was adminis tered repeatedly (50 mg for 30 days; n=7 men),197 reported a minimal e¡ect on the T/EpiT ratio. In contrast, Bowers198 found that one of four male subjects showed a dose -dependent increase in the T/ EpiT ratio with DHEA ingestion (50, 100 and 150 mg/day for 3 days), resulting in the T/EpiT threshold being exceeded. In another small study,36 a single administration of 200 mg of DHEA (n=3 men) resulted in T/EpiT 46 in one subject being exceeded, but he had a relatively high basal T/EpiT of 4. Levesque¨ and Ayotte have published their criteria for the detection of oral administration of androstenedione in the proceedings of a workshop.199 These authors comment that the T/EpiT ratio may be raised, but not systematically, the urinary concentrations of androsterone and etiocholanolone are increased to abnormal `levels’and the presence of the characteristic metabolites 6a-hydroxyandrostenedione, 6b-hydroxyandrosterone and 6 b-hydroxyetiocholanolone are also diagnostic, being present as glucuro- and sulphoconjugates.200 Their criteria may be useful for detecting presumptive positives but the criteria are unlikely to be su¤cient to withstand the rigours of a strong legal challenge. All these metabolites are also produced naturally; hence the lack of decision limits for the concentration of these analytes in urine (in cases where T/EpiT 56) makes it di¤cult for the analyst to con¢rm an adverse ¢nding. Van Eenoo et Ann Clin Biochem 2003; 40: 321–356

346

Kicman and Gower

al. 201 describe in a preliminary report the results of statistical analysis of urinary concentrations of androstenedione and other endogenous steroids, in samples (n=305) collected for doping analysis in Flanders. Using a non-parametric approach, the faroutside values for the concentration of androstenedione and ratio of androstenedione/EpiT were 23 ng/mL and 1¢2, respectively.Van Eenoo et al. regard the far-outside value as the decision limit in doping analysis (note: Laidler et al.202 used twice the faroutside value as a decision limit for human chorionic gonadotrophin concentration in urine collected from sportsmen).

Detection of ‘natural’ androgen administration by determination of the carbon isotope ratio

A number of recent articles 203^208 have reported the potential of gas chromatography/combustion/ isotope ratio mass spectrometry (GC/C/IRMS) as an analytical tool for detecting T administration. Further, this technique may be useful in the future for the analysis of nandrolone metabolites, although currently it lacks su¤cient sensitivity for this speci¢c purpose (see above). With GC/C/IRMS, analytes from urine are separated on the GC column and are combusted catalytically to carbon dioxide and water in a furnace at high temperature. A Na¢on 1 membrane or a cryogenic trap removes the water so that only carbon dioxide enters the mass spectrometer, between pulses of reference standard carbon dioxide obtained from a natural carbonate source and calibrated against an international standard. The three major ions monitored at m/z 44, 45 and 46 correspond to the isotopes 12C16O16O, 13 16 16 C O O and 12C18O16O, respectively. Measurements of ion currents at m/z 45 and m/z 44 are chosen to calculate the ratio of 13C to 12C. The isotopic abundance of the analyte is then compared to the reference gas using a d (delta) scale, expressed in parts per thousand, usually quoted as parts per mille ( %):

d13 C…%† ˆ

…13 C

12

C†sample ¡ …13 C 13

… C

12

12

C†reference

C†reference

£ 1000

The natural abundance of 13C is *1Í11%, the human diet consisting of plant and animal sources with varying 13C content relative to 12C. Endogenously produced steroids should thus have a 13C/12C that re£ects an average of that in the carbon sources ingested. T used in pharmaceutical formulations is now generally synthesized from stigmasterol, which is obtained from soya beans and has a lower 13C content (see Fig. 13). Even though the amount of 13C in pharmaceutical T and endogenously produced T approximates 1Í11% of the 12C present, GC/C/IRMS can accurately measure a distinct di¡erence in the Ann Clin Biochem 2003; 40: 321–356

Figure 13. Endogenously produced steroids have a 13C/12C content that reects an average of that in the carbon sources ingested; testosterone in pharmaceutical formulations is synthesized from soy, which has a smaller 13C content.

isotope ratio between them, as combusted carbon dioxide products. Following administration of T, the 13 C content of pooled T decreases whereas the isotope ratio of the C 21 biosynthetic precursors of T (i.e. pregnenolone and progesterone) should remain unchanged. Detection of doping with T administration can be based on comparison of the carbon isotope ratio of the combusted urinary metabolites of T (i.e. 5 a- and 5 b-androstanediol) to that of pregnenolone (i.e. a pregnanediol, 5 b-pregnane-3 a,20 a-diol) in the urine from an individual (see Fig. 14). An absolute cut-o¡ value for the isotope ratio of urinary T or its metabolites may also be helpful. For the detection of `natural steroid’administration, the majority of recent studies have shown that GC/C/ IRMS can match the capability of the established hormone ratios approach. Nonetheless, further investigations are required to evaluate whether this interesting approach has su¤cient sensitivity ; reference ranges need to be established and there should be standardization of measurements between laboratories. In addition, it needs to be substantiated that enzymes in the steroid biosynthetic pathway do not di¡erentiate between steroid substrates with di¡erent 13 C content. Despite this, GC/C/IRMS appears to be a powerful tool and may become even more so as more work is done to improve sensitivity and to develop and evaluate the potential of this technique.

Human chorionic gonadotrophin Human chorionic gonadotrophin (hCG) is a glycoprotein hormone misused by some male athletes to stimulate endogenous T production or to prevent

Anabolic steroids in sport

347

Figure 14. Values of d13C of derivatized steroids (acetates) from a urinary extract of a subject before and after injection of 250 mg of testosterone heptanoate (oenanthate). The open circles present pregnanediol, the closed squares 5b-androstanediol and the closed triangles 5a-androstanediol. Pregnanediol (5b-pregnane-3a,17b-diol) is a C21 metabolite of progesterone (a precursor of testosterone in the biosynthetic pathway); the androstanediols are metabolites of testosterone (from an endogenous or exogenous source). Reproduced from Shackleton et al.,207 with permission from Elsevier Science. Copyright & 1997.

testicular atrophy during prolonged administration of AASs. As the testes contribute to *95% of the pool of urinary T and EpiT glucuronide in eugonadal men,171 hCG stimulation results in a contemporaneous increase in the urinary excretion rate of both EpiT and T and hence there are only minor changes in the urinary T/EpiT ratio (e.g. Cowan et al. 209). HCG coadministered with T suppresses changes in the urinary T/EpiT ratio, 210 the increased epitestosterone production diminishing augmentation of the ratio that would otherwise occur solely with T administration. In those men who have not used hCG, its reported presence in urine is due to an hCG-secreting tumour that otherwise could have remained undetected for a dangerous period of time. The administration of hCG to female athletes would stimulate ovarian steroidogenesis; this is unlikely to enhance performance,211 thus samples collected from women are not tested for hCG. In November 1987, following the detection of unnaturally large concentrations of hCG in a number of sports samples analysed in the UK,212 the IOC Medical Commission banned the administration of hCG. The section on laboratory procedures of the current Olympic Movement Anti-Doping Code17 states: `For hCG: a validated immunoassay to detect and quantitate hCG. For con¢rmation, a second di¡erent immunoassay is required.’ IOC-accredited laboratories employ commercial immunometric assays for the detection of hCG. Laidler et al.,202 using the Serono MAIAclone immunoradiometric assay, proposed a

decision limit of 10 IU/L in ultra¢ltered urine samples, above which, after con¢rmatory procedures, a sample is considered positive. With con¢rmation, it is highly desirable that the concentration of hCG measured by the `second di¡erent immunoassay’ agrees with that of the ¢rst immunoassay used. However, variance in results is possible due to the di¡erence in immunoglobulin speci¢city between kit manufacturers, in particular the cross-reaction with nicked hCG and hCG b-core fragment present in the urine following administration. Stenman et al.213 comment that, if appropriately standardiz ed and validated methods are used, investigators should be able to detect self-administration of hCG in men as reliably as AASs are being detected by MS methods. Nonetheless, if methods based on softionization MS for end-point measurement were to achieve the sensitivity associated with immunoassays for protein hormones, then they are likely to be adopted for con¢rmatory purposes.123 The discriminatory power of MS has been demonstrated by analysis of tryptic digests of hCG, using matrix-assisted laser desorption ionization time-of-£ight (MALDI-TOF)/ MS 214 (see Fig. 15). Immunoa¤nity capture with improved sample immobilization protocols may enable laser desorption MS to be employed for the control of hCG. HPLC/MS using electrospray ionization has been used to identify tryptic fragments of hCG corresponding to 25 IU/L of intact hCG, following a solidphase immunoa¤nity trapping technique applied to 10 mL of urine.215 Both MALDI-TOF/MS and HPLC/ Ann Clin Biochem 2003; 40: 321–356

348

Kicman and Gower

Figure 15. MALDI-TOF mass spectrum of a tryptic digest of reduced and S-carboxymethylated human chorionic gonadotrophin in acyano-4-hydroxycinnamic acid matrix. Fragments are labelled as described in the text and glycopeptides are designated by *. Signals from 30 laser shots were summed to generate the spectrum, and the m/z axis was calibrated internally using fragment masses 1227¢3 and 1928¢4 Da. Figure from Laidler et al.,214 reproduced with permission from John Wiley and Sons Ltd. Copyright & 1995. MS have the potential for con¢rming the presence of hCG in urine but it remains to be established whether accurate quanti¢cation is possible.

Pharmacological manipulation to evade detection of AAS administration Co-administration of EpiT with T

Classi¢ed documents51 saved after the collapse of the GDR revealed that since 1983 a pharmaceutical company had produced preparations of EpiT propionate exclusively for the governmental doping programme. The administration of EpiT with T simultaneously or sequentially enables an athlete to manipulate the test for T administration if the test is based solely on determination of the T/EpiT ratio. Only *1% of T is excreted unchanged, apart from conjugation to glucuronic acid, compared with *30% of EpiT. Hence, an injection of these esteri¢ed steroids in a ratio of *30 parts T to1part EpiT will result in a raised plasma T but an unremarkable T/EpiT ratio, albeit that the T/LH ratio can be augmented.216 An approach that was used to detect EpiT adminis tration was to set a threshold for reporting a urinary EpiT concentration, but this strategy was unlikely to be Ann Clin Biochem 2003; 40: 321–356

of great practical value as urinary concentrations of steroids vary greatly depending on the degree of dilution of the urine. The IOC Medical Commission did introduce a limit on the urinary concentration of EpiT of 150 mg/L in the Anti-Doping Medical Code but it later changed the threshold to 200 mg/L.186 The use of IRMS has been proposed for the screening of T, 206 and such an approach could also be applied to detect co-administration of EpiT.217

Probenecid Probenecid is a competitive inhibitor of renal tubular transport mechanisms and, because it inhibits the reabsorption of uric acid, one of its main therapeutic uses is for the treatment of hyperuricaemia. Probenecid may also interfere with renal elimination of certain compounds, reducing their concentration in urine. The majority of AASs and their metabolites undergo phase II metabolism and, although comprehensive studies have not been undertaken, probenecid appears to reduce the excretion of the associated glucuronide conjugates. In 1988, the administration of probenecid was banned by the IOC Medical Commission because its use is considered as a `pharmacological manipulation’ that alters the integrity and validity of urine samples. Geyer et al. 218 showed

Anabolic steroids in sport

that the urinary concentration and excretion rate of 19-norandrosterone, the diagnostic metabolite of nandrolone, in a volunteer who was self-administering nandrolone decanoate was greatly reduced following administration of 3 g of probenecid. Screening for probenecid is usually incorporated with the method for analysis of AASs (i.e. by SIM GC/MS of the TMS ether derivative).

5 -Reductase inhibitors The activity of 5 a-reductase is implicated in BPH,

hirsutism and possibly male-pattern baldness, and speci¢c inhibitors of both isozymes of 5a-reductase are available for treatment purposes.219 The most wellknown inhibitor is ¢nasteride, which is a more e¡ective inhibitor of the type II than the type I isozyme 220 and has been most successfully used in the treatment of BPH. Finasteride may be improperly used to reduce the unwanted androgenic e¡ects of AAS abuse and more insidiously to evade detection by reducing the formation of diagnostic 5 a-reduced metabolites. Geyer et al.221 describe in a workshop proceedings a considerable reduction in the urinary excretion rate of 19-norandrosterone when ¢nasteride (5 mg) was administered after administration of nandrolone. As a consequence, the expert committee undertaking the Nandrolone Review for UK Sport recommended that the IOC Medical Committee should consider adding this and similar inhibitors to the list of banned masking agents.134 Urinary excretion of ¢nasteride is mainly as acidic metabolite(s), a major metabolite being the monocarboxylic acid as a result of one of the t-butylmethyl groups of ¢nasteride undergoing oxidation. 222 Developmental work will be required to enable analysis of this acidic metabolite to be incorporated into existing screening methods operated by IOC-accredited laboratories (if ¢nasteride is added to the list of banned masking agents).

Role of the expert in evidential analysis In April 1999 the new Civil Procedure Rules instigated by the Final Report by Lord Woolf on Access to Justice (1996) came into e¡ect. In that Final Report, there is a section on expert evidence, where Lord Woolf writes: `There is widespread agreement with the criticisms I made in the interim report of the way in which expert evidence is used at present, especially the point that experts sometimes take on the role of partisan advocates instead of neutral fact ¢nders or opinion givers.’ The Final Report also emphasizes that the expert’s function is to assist the court. It is in keeping with the spirit of the `Woolf Report’ that expert witnesses should not be partisan when

349

giving opinion or advice to a sports arbitration panel in connection with a positive ¢nding. IOC and ISO17025 accreditations help to ensure that data reported by IOC laboratories are usually of a su¤ciently high standard to withstand critical scrutiny (it is intended that the responsibility of the IOC in accreditating sports drug testing laboratories is to be passed over to that of the World Anti-Doping Agency in 2004). However, much time can be wasted on unbalanced and uncritical presentation of poor quality scienti¢c evidence that has been published.168 This brings into disrepute the competence of the witness and diminishes the value of science to the procedure. Good science in the ¢eld of drug control in sport should follow the normal principles of good science. Thus, it is important to present research and developmental work in good quality peer-reviewed journals; in turn, this should assist experts in giving a balanced opinion to arbitration panels.

Acknowledgements The authors wish to thank the following: Professors David Cowan (Drug Control Centre, King’s College London) and Vivian James (Chemical Pathology, St Mary’s Hospital, Imperial College London), Drs Larry Bowers (US Anti-Doping Agency) and Andrew Hutt (Department of Pharmacy, King’s College London) and Ms Michele Verroken (UK Sport) for helpful discussion; Dr Hendrik Neubert and Mr Chris Walker (Drug Control Centre, King’s College London) and Mrs DM Gower for assisting with the preparation of the manuscript.

References 1 Ruckzika L, Wettstein A, Kaegi H. Sexualhormone VIII Darstellung von Testosterone unter Anwendung gemischter Ester. Helv Chim Acta 1935; 18: 1478 2 Todd T. Anabolic steroids: the gremlins of sport. J Sport History 1987; 14: 87-107 3 Hershberger LG, Shipley EG, Meyer RK. Myotrophic activity of 19nortestosterone and other steroids determined by modiŽ ed levator ani muscle method. Proc Soc Exp Biol Med 1953; 83: 175-80 4 Kruskemper HL. Anabolic Steroids. New York: Academic Press, 1968: 236 5 Budavari S, ed. The Merck Index. Whitehouse Station, NJ: Merck, 1996 6 Beckett AH. Use and abuse of drugs in sport. J Biosoc Sci Suppl 1981; 7: 163-78 7 Abraham GE. Solid-phase radioimmunoassay of estradiol-17b. J Clin Endocrinol Metab 1969; 29: 866-70 8 Brooks RV, Firth RG, Sumner NA. Detection of anabolic steroids by radioimmunoassay. Br J Sports Med 1975; 9: 89-92 9 Ward RJ, Shackleton CH, Lawson AM. Gas chromatographic-mass spectrometric methods for the detection and identiŽ cation of anabolic steroid drugs. Br J Sports Med 1975; 9: 93-7 Ann Clin Biochem 2003; 40: 321–356

350

Kicman and Gower

10 Brooks RV, Jeremiah G, Webb WA, Wheeler M. Detection of anabolic steroid administration to athletes. J Steroid Biochem 1979; 11: 913-7 11 Hampl R, Starka L. Practical aspects of screening of anabolic steroids in doping control with particular accent to nortestosterone radioimmunoassay using mixed antisera. J Steroid Biochem 1979; 11: 933-6 12 Kicman AT, Brooks RV. Radioimmunoassay for nandrolone metabolites. J Pharm Biomed Anal 1988; 6: 473-83 13 Catlin DH, Kammerer RC, Hatton CK, Sekera MH, Merdink JL. Analytical chemistry at the Games of the XXIIIrd Olympiad in Los Angeles, 1984. Clin Chem 1987; 33: 319-27 14 Swiss Pharmaceutical Society. Index Nominum: International Drug Directory. Stuttgart: Medpharm ScientiŽ c Publishers, 1995 15 Morales AJ, Nolan JJ, Nelson JC, Yen SS. Effects of replacement dose of dehydroepiandrosterone in men and women of advancing age. J Clin Endocrinol Metab 1994; 78: 1360-7 16 Di Pasquale MG. Dehydroepiandrosterone (DHEA). Beyond Anabolic Steroids. Ontario: MGD Press, 1990: 63 17 Olympic Movement Anti-Doping Code. Access via http:// www.olympic.org (under ‘Ask a question’ type ‘Anti-Doping Code’); date last accessed 8 October 2002 18 Catlin DH, Leder BZ, Ahrens B, et al. Trace contamination of overthe-counter androstenedione and positive urine test results for a nandrolone metabolite. JAMA 2000; 284: 2618-21 19 Tomoda H. Effect of oxymetholone on left ventricular dimensions in heart failure secondary to idiopathic dilated cardiomyopathy or to mitral or aortic regurgitation. Am J Cardiol 1999; 83: 123-5, A9 20 Bhasin S, Woodhouse L, Storer TW. Proof of the effect of testosterone on skeletal muscle. J Endocrinol 2001; 170: 27-38 21 Bogin V, Shaw-Stiffel T. Androgenic anabolic steroids and other therapies for HIV-related wasting. J Clin Ligand Assay 1999; 22: 268-78 22 Shahidi NT. A review of the chemistry, biological action, and clinical applications of anabolic-androgenic steroids. Clin Ther 2001; 23: 1355-90 23 Hengge UR, Baumann M, Maleba R, Brockmeyer NH, Goos M. Oxymetholone promotes weight gain in patients with advanced human immunodeŽ ciency virus (HIV-1) infection. Br J Nutr 1996; 75: 129-38 24 Ryan AJ. Athletics. In: Kochakian CD, ed. Anabolic-Androgenic Steroids, Vol. 43. Berlin: Springer-Verlag, 1976: 515-34 (see also Addendum, p. 723) 25 Haupt HA, Rovere GD. Anabolic steroids: a review of the literature. Am J Sports Med 1984; 12: 469-84 26 Alen M, Hakkinen K. Androgenic steroid effects on serum hormones and on maximal force development in strength athletes. J Sports Med Phys Fitness 1987; 27: 38-46 27 Strauss RH, Yesalis CE. Anabolic steroids in the athlete. Annu Rev Med 1991; 42: 449-57 28 Bhasin S, Storer TW, Berman N, et al. The effects of supraphysiologic doses of testosterone on muscle size and strength in normal men. N Engl J Med 1996; 335: 1-7 29 Bhasin S, Woodhouse L, Casaburi R, et al. Testosterone doseresponse relationships in healthy young men. Am J Physiol Endocrinol Metab 2001; 281: E1172-81 30 Welle S, Jozefowicz R, Statt M. Failure of dehydroepiandrosterone to in uence energy and protein metabolism in humans. J Clin Endocrinol Metab 1990; 71: 1259-64 31 Nestler JE, Barlascini CO, Clore JN, Blackard WG. Dehydroepiandrosterone reduces serum low density lipoprotein levels and body Ann Clin Biochem 2003; 40: 321–356

32

33

34

35

36

37

38

39

40

41

42

43

44

45 46

47 48 49 50

fat but does not alter insulin sensitivity in normal men. J Clin Endocrinol Metab 1988; 66: 57-61 King DS, Sharp RL, Vukovich MD, et al. Effect of oral androstenedione on serum testosterone and adaptations to resistance training in young men: a randomized controlled trial. JAMA 1999; 281: 2020-8 Leder BZ, Longcope C, Catlin DH, Ahrens B, Schoenfeld DA, Finkelstein JS. Oral androstenedione administration and serum testosterone concentrations in young men. JAMA 2000; 283: 77982 Brown GA, Vukovich MD, Sharp RL, Reifenrath TA, Parsons KA, King DS. Effect of oral DHEA on serum testosterone and adaptations to resistance training in young men. J Appl Physiol 1999; 87: 2274-83 Uralets VP, Gillette PA. Over-the-counter anabolic steroids 4androsten-3,17-dione; 4-androsten-3b,17b-diol; and 19-nor-4-androsten-3,17-dione: excretion studies in men. J Anal Toxicol 1999; 23: 357-66 Uralets VP, Gillette PA. Over-the-counter ¢5 anabolic steroids 5androsen-3,17-dione; 5-androsten-3b,17b-diol; dehydroepiandrosterone; and 19-nor-5-androsten-3,17-dione: excretion studies in men. J Anal Toxicol 2000; 24: 188-93 Horton R, Tait JF. Androstenedione production and interconversion rates measured in peripheral blood and studies on the possible site of its conversion to testosterone. J Clin Invest 1966; 45: 301-13 Bardin CW, Lipsett MB. Testosterone and androstenedione blood production rates in normal women and women with idiopathic hirsutism or polycystic ovaries. J Clin Invest 1967; 46: 891-902 Mahesh VB, Greenblatt GR. The in vivo conversion of dehydroepiandrosterone and androstenedione to testosterone in the human. Acta Endocrinol 1962; 41: 400-6 Kicman AT, Bassindale T, Cowan DA, Dale S, Hutt AJ, Leeds AR. The effect of androstenedione ingestion on plasma testosterone in young women; a dietary supplement with potential health risks. Clin Chem 2003; 49: 167–9 Dahlberg E, Snochowski M, Gustafsson JA. Regulation of the androgen and glucocorticoid receptors in rat and mouse skeletal muscle cytosol. Endocrinology 1981; 108: 1431-40 Hervey GR. What are the effects of anabolic steroids? In: Davies B, Thomas G, eds. Science and Sporting Performance. Oxford: Clarendon Press, 1982: 120-36 Hickson RC, Czerwinski SM, Falduto MT, Young AP. Glucocorticoid antagonism by exercise and androgenic-anabolic steroids. Med Sci Sports Exerc 1990; 22: 331-40 Sharpe PM, Buttery PJ, Haynes NB. The effect of manipulating growth in sheep by diet or anabolic agents on plasma cortisol and muscle glucocorticoid receptors. Br J Nutr 1986; 56: 289-304 Wu FC. Endocrine aspects of anabolic steroids. Clin Chem 1997; 43: 1289-92 Brooks RV. Some aspects of the action of anabolic steroids on athletes. In: Landry F, Oraban WAR, eds. Physical Activity and Human Well-Being. Quebec: Symposia Specialists, 1978: 219-29 Mooradian AD, Morley JE, Korenman SG. Biological actions of androgens. Endocr Rev 1987; 8: 1-28 Di Pasquale MG. Anabolic Steroid Side-Effects: Facts, Fiction and Treatment. Ontario: MGD Press, 1990: 63 Graham S, Kennedy M. Recent developments in the toxicology of anabolic steroids. Drug Saf 1990; 5: 458-76 Landry GL, Primos WA Jr. Anabolic steroid abuse. Adv Pediatr 1990; 37: 185-205

Anabolic steroids in sport 51 Franke WW, Berendonk B. Hormonal doping and androgenization of athletes: a secret program of the German Democratic Republic government. Clin Chem 1997; 43: 1262-79 52 Cicardi M, Bergamaschini L, Tucci A, et al. Morphologic evaluation of the liver in hereditary angioedema patients on long-term treatment with androgen derivatives. J Allergy Clin Immunol 1983; 72: 294-8 53 Dickerman RD, Pertusi RM, Zachariah NY, Dufour DR, McConathy WJ. Anabolic steroid-induced hepatotoxicity: is it overstated? Clin J Sport Med 1999; 9: 34-9 54 Pope HG Jr, Katz DL. Homicide and near-homicide by anabolic steroid users. J Clin Psychiatry 1990; 51: 28-31 55 Pope HG Jr, Katz DL. Affective and psychotic symptoms associated with anabolic steroid use. Am J Psychiatry 1988; 145: 487-90 56 Choi PY, Pope HG Jr. Violence toward women and illicit androgenic-anabolic steroid use. Ann Clin Psychiatry 1994; 6: 21-5 57 Yesalis CE, Vicary JR, Buckley WE, Streit AL, Katz DL, Wright JE. Indications of psychological dependence among anabolic-androgenic steroid abusers. NIDA Res Monogr 1990; 102: 196-214 58 Hurley BF, Seals DR, Ehsani AA, et al. Effects of high-intensity strength training on cardiovascular function. Med Sci Sports Exerc 1984; 16: 483-8 59 Cohen JC, Faber WM, Benade AJS, Noakes TD. Altered serumlipoprotein proŽ les in male and female power lifters ingesting anabolic-steroids. Physician Sportsmed 1986; 14: 131-6 60 Kuipers H, Wijnen JA, Hartgens F, Willems SM. In uence of anabolic steroids on body composition, blood pressure, lipid proŽ le and liver functions in body-builders. Int J Sports Med 1991; 12: 413-8 61 Sader MA, GrifŽ ths KA, McCredie RJ, Handelsman DJ, Celermajer DS. Androgenic-anabolic steroids and arterial structure and function in male body-builders. J Am Coll Cardiol 2001; 37: 224-30 62 Sullivan ML, Martinez CM, Gallagher EJ. Atrial Ž brillation and anabolic steroids. J Emerg Med 1999; 17: 851-7 63 Parssinen M, Kujala U, Vartiainen E, Sarna S, Seppala T. Increased premature mortality of competitive powerlifters suspected to have used anabolic agents. Int J Sports Med 2000; 21: 225-7 64 Magnavita N, TeoŽ li L, Leone G. Hodgkin’s lymphoma in a cyclist treated with growth hormone. Am J Hematol 1996; 52: 65-6 65 Nieminen MS, Ramo MP, Viitasalo M, et al. Serious cardiovascular side-effects of large doses of anabolic steroids in weight lifters. Eur Heart J 1996; 17: 1576-83 66 World Health Organization. Risks and BeneŽ ts of Androgen Therapy. Guidelines for the Use of Androgens in Men. Geneva: WHO, 1992: 5-6 67 Wemyss-Holden SA, Hamdy FC, Hastie KJ. Steroid abuse in athletes, prostatic enlargement and bladder out ow obstruction: is there a relationship? Br J Urol 1994; 74: 476-8 68 Holmang S, Marin P, Lindstedt G, Hedelin H. Effect of long-term oral testosterone undecanoate treatment on prostate volume and serum prostate-speciŽ c antigen concentration in eugonadal middleaged men. Prostate 1993; 23: 99-106 69 Jin B, Turner L, Walters WA, Handelsman DJ. Androgen or estrogen effects on human prostate. J Clin Endocrinol Metab 1996; 81: 4290-5 70 Negro-Vilar A. Selective androgen receptor modulators (SARMs): a novel approach to androgen therapy for the new millennium. J Clin Endocrinol Metab 1999; 84: 3459-62

351

71 Roy AK, Tyagi RK, Song CS, et al. Androgen receptor: structural domains and functional dynamics after ligand-receptor interaction. Ann N Y Acad Sci 2001; 949: 44-57 72 Celotti F, Cesi PN. Anabolic-steroids: a review of their effects on the muscles, of their possible mechanisms of action and of their use in athletics. J Steroid Biochem Mol Biol 1992; 43: 469-77 73 Toth M, Zakar T. Relative binding afŽ nities of testosterone, 19nortestosterone and their 5-a-reduced derivatives to the androgen receptor and to other androgen-binding proteins: a suggested role of 5-a-reductive steroid-metabolism in the dissociation of myotropic and androgenic activities of 19-nortestosterone. J Steroid Biochem Mol Biol 1982; 17: 653-60 74 Potts GO, Arnold A, Beyler AL. Dissociation of the androgenic and other hormonal activities from the protein anabolic effects of steroids. In: Kochakian CD, ed. Handbook of Experimental Pharmacology, Vol. 43. Berlin: Springer-Verlag, 1976: 361-401 75 van der Vies J. Pharmacokinetics of anabolic steroids. Wien Med Wochenschr 1993; 143: 366-8 76 Miyamoto H, Yeh S, Lardy H, Messing E, Chang C. Delta5androstenediol is a natural hormone with androgenic activity in human prostate cancer cells. Proc Natl Acad Sci USA 1998; 95: 11083-8 77 Gower DB, Houghton E, Kicman AT. Anabolic steroids: metabolism, doping and detection in equestrian and human sports. In: Makin HLJ, Gower DB, Kirk DN, eds. Steroid Analysis. Glasgow: Blackie Academic & Professional, 1995: 468-526 78 Scha¨nzer W, Donike M. Metabolism of anabolic steroids in man: synthesis and use of reference substances for identiŽ cation of anabolic steroid metabolites. Anal Chim Acta 1993; 275: 23-48 79 Bowers LD. Metabolic pattern of mibolerone. In: Donike M, Geyer H, Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis. Proceedings of the 13th Cologne Workshop on Dope Analysis. Koln: Sport und Buch Strauss, 1996: 81-2 80 Uralets VP, Gillette PA, DuVall B, Latven RK. Trenbolone: screening and conŽ rmation. In: Donike M, Geyer H, Gotzmann A, MareckEngelke U, eds. Recent Advances in Doping Analysis. Proceedings of the 13th Cologne Workshop on Dope Analysis. Koln: Sport und Buch Strauss, 1996: 83-94 81 Masse R, Bi HG, Ayotte C, Du P, Gelinas H, Dugal R. Studies on anabolic steroids. V. Sequential reduction of methandienone and structurally related steroid A-ring substituents in humans: gas chromatographic-mass spectrometric study of the corresponding urinary metabolites. J Chromatogr 1991; 562: 323-40 82 Masse R, Bi HG, Ayotte C, Dugal R. Studies on anabolic steroids. II: Gas chromatographic/mass spectrometric characterization of oxandrolone urinary metabolites in man. Biomed Environ Mass Spectrom 1989; 18: 429-38 83 Masse R, Ayotte C, Bi HG, Dugal R. Studies on anabolic steroids. III. Detection and characterization of stanozolol urinary metabolites in humans by gas chromatography-mass spectrometry. J Chromatogr 1989; 497: 17-37 84 Durbeck HW, Buker I. Studies on anabolic steroids. The mass spectra of 17a-methyl-17b-hydroxy-1,4-androstadien-3-one (Dianabol) and its metabolites. Biomed Mass Spectrom 1980; 7: 437-45 85 Durbeck HW, Buker I, Scheulen B, Telin B. GC and capillary column GC/MS determination of synthetic anabolic steroids. II. 4Chloro-methandienone (oral turinabol) and its metabolites. J Chromatogr Sci 1983; 21: 405-10 86 Durbeck HW, Buker I, Scheulen B, Telin B. Gas chromatographic and capillary column gas chromatographic-mass spectrometric Ann Clin Biochem 2003; 40: 321–356

352

87

88

89

90 91 92

93 94

95

96

97 98 99 100

101

102

103 104 105 106

Kicman and Gower determination of synthetic anabolic steroids. I. Methandienone and its metabolites. J Chromatogr 1978; 167: 117-24 Bi H, Masse R. Studies on anabolic steroids 12. Epimerization and degradation of anabolic 17b-sulfate-17a-methyl steroids in human: qualitative and quantitative GC/MS analysis. J Steroid Biochem Mol Biol 1992; 42: 533-46 Schanzer W, Geyer H, Donike M. Metabolism of metandienone in man: identiŽ cation and synthesis of conjugated excreted urinary metabolites, determination of excretion rates and gas chromatographic-mass spectrometric identiŽ cation of bishydroxylated metabolites. J Steroid Biochem Mol Biol 1991; 38: 441-64 Edlund PO, Bowers L, Henion J, Covey TR. Rapid-determination of methandrostenolone in equine urine by isotope-dilution liquidchromatography tandem mass-spectrometry. J Chromatogr B 1989; 497: 49-57 Birch AJ. Hydroaromatic steroid hormones. Part 1. 19Nortestosterone. J Am Chem Soc 1950; 44: 367-8 Engel LL, Alexander J, Wheeler M. Urinary metabolites of administered 19-nortestosterone. J Biol Chem 1958; 231: 159-64 Masse R, Laliberte C, Tremblay L, Dugal R. Gas chromatographic/ mass spectrometric analysis of 19-nortestosterone urinary metabolites in man. Biomed Mass Spectrom 1985; 12: 115-21 Rongone E, Segaloff F. In vivo metabolism of d-1,17-amethyltestosterone. Steroids 1963; 1: 179-84 Schanzer W, Opfermann G, Donike M. Metabolism of stanozolol: identiŽ cation and synthesis of urinary metabolites. J Steroid Biochem 1990; 36: 153-74 Clay KL, Watkins WD, Murphy RC. Metabolism of pyrazole. Structure elucidation of urinary metabolites. Drug Metab Dispos 1977; 5: 149-56 Ayotte C, Charlebois A, Lapointe S, Barriault D, Sylvestre M. Validity of urine samples; microbial degradation. In: Schanzer W, Geyer H, Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis (4). Proceedings of the Manfred Donike Workshop, 14th Cologne Workshop on Dope Analysis, 17th to 22nd March 1996. Koln: Sport und Buch Strauss, 1997: 127-37 Bilton RF. Microbial production of testosterone [letter]. Lancet 1995; 345: 1186-7 Birchard K. Why doctors should worry about doping in sport. Lancet 1998; 352: 42 Honour JW. Testing for drug abuse. Lancet 1996; 348: 41-3 de la Torre R, de la Torre X, Alia C, Segura J, Baro T, TorresRodriguez JM. Changes in androgenic steroid proŽ le due to urine contamination by microorganisms: a prospective study in the context of doping control. Anal Biochem 2001; 289: 116-23 Kicman AT, Fallon JK, Cowan DA, Walker C, Easmon S, Mackintosh D. Candida albicans can produce testosterone: impact on the T/E sports drug test. Clin Chem 2002; 10: 1799-801 Peng SH, Segura J, Farre M, de la Torre X. Oral testosterone administration detected by testosterone glucuronidation measured in blood spots dried on Ž lter paper. Clin Chem 2000; 46: 515-22 Rivier L. Is there a place for hair analysis in doping controls? Forensic Sci Int 2000; 107: 309-23 Wennig R. Potential problems with the interpretation of hair analysis results. Forensic Sci Int 2000; 107: 5-12 Sachs H, Dressler U. Detection of THCCOOH in hair by MSD-NCI after HPLC clean-up. Forensic Sci Int 2000; 107: 239-47 Nakahara Y, Takahashi K, Kikura R. Hair analysis for drugs of abuse. X. Effect of physicochemical properties of drugs on the incorporation rates into hair. Biol Pharm Bull 1995; 18: 1223-7

Ann Clin Biochem 2003; 40: 321–356

107 Rollins DE, Wilkins DG, Krueger GG. Codeine disposition in human hair after single and multiple doses. Eur J Clin Pharmacol 1996; 50: 391-7 108 Henderson GL, Harkey MR, Zhou C, Jones RT, Jacob P 3rd. Incorporation of isotopically labeled cocaine and metabolites into human hair: 1. Dose-response relationships. J Anal Toxicol 1996; 20: 1-12 109 Kintz P, Cirimele V, Ludes B. Pharmacological criteria that can affect the detection of doping agents in hair. Forensic Sci Int 2000; 107: 325-34 110 Segura J, Pichini S, Peng SH, de la Torre X. Hair analysis and detectability of single dose administration of androgenic steroid esters. Forensic Sci Int 2000; 107: 347-59 111 Gaillard Y, Vayssette F, Pepin G. Compared interest between hair analysis and urinalysis in doping controls. Results for amphetamines, corticosteroids and anabolic steroids in racing cyclists. Forensic Sci Int 2000; 107: 361-79 112 Cone EJ. Mechanisms of drug incorporation into hair. Ther Drug Monit 1996; 18: 438-43 113 Jurado C, Kintz P, Menendez M, Repetto M. In uence of the cosmetic treatment of hair on drug testing. Int J Legal Med 1997; 110: 159-63 114 Kidwell DA. Is hair testing culturally biased and why is this a concern in the US? Second European Meeting on Hair Analysis, Martigny, Switzerland, 14-16 June 1999 115 Vanluchene E, Eechaute W, Vandekerckhove D. Conversion of free 3b-hydroxy-5-ene-steroids by incubation with Helix pomatia. J Steroid Biochem 1982; 16: 701-3 116 Messeri G, Cugnetto G, Moneti G, Serio M. Helix pomatia induced conversion of some 3b-hydroxysteroids. J Steroid Biochem 1984; 20: 793-6 117 Houghton E, Grainger L, Dumasia MC, Teale P. application of gaschromatography-mass spectrometry to steroid analysis in equine sports: problems with enzyme hydrolysis. Org Mass Spectrom 1992; 27: 1061-70 118 Donike M, Zimmermann J. Preparation of trimethylsilyl, triethylsilyl and tert-butyldimethylsilyl enol ethers from ketosteroids for investigations by gas-chromatography and mass-spectrometry. J Chromatogr 1980; 202: 483-6 119 Opfermann G, Schanzer W. Trimethylsilylation: aspects for derivatisation. In: Schanzer W, Geyer H, Gotzmann A, MareckEngelke U, eds. Recent Advances in Doping Analysis (4). Proceedings of the 14th Cologne Workshop on Dope Analysis 17th to 22nd March 1996. Koln: Sport und Buch Strauss, 1997: 247-52 120 Masse R, Ayotte C, Dugal R. Studies on anabolic steroids. I. Integrated methodological approach to the gas chromatographicmass spectrometric analysis of anabolic steroid metabolites in urine. J Chromatogr 1989; 489: 23-50 121 Ayotte C, Goudreault D, Charlebois A. Testing for natural and synthetic anabolic agents in human urine. J Chromatogr B 1996; 687: 3-25 122 Donike M, Geyer H, Gotzmann A, et al. Dope analysis. In: Bellotti P, Benzi G, Ljungqvist A, eds. International Athletic Foundation World Symposium on Doping in Sport: OfŽ cial Proceedings, Florence, 10-12 May, 1987: International Athletic Foundation (IAF), 1988: 53-80 123 Bowers LD. Analytical advances in detection of performanceenhancing compounds. Clin Chem 1997; 43: 1299-304 124 Horning S, Donike M. High resolution GC/MS. In: Donike M, Geyer H, Gotzmann A, Mareck-Engelke U, Rauth S, eds. Recent

Anabolic steroids in sport

125

126

127

128

129

130

131

132

133

134 135

136

137

138

139

Advances in Doping Analysis. Proceedings of the 11th Cologne Workshop on Dope Analysis 7th to 12th March 1993. Koln: Sport und Buch Strauss, 1994: 155-61 Mueller RK, Grosse J, Lang R, Thieme D. Chromatographic techniques: the basis of doping control. J Chromatogr B 1995; 674: 1-11 Kokkonen J, Leinonen A, Tuominen J, Seppala T. Comparison of sensitivity between gas chromatography-low-resolution mass spectrometry and gas chromatography-high-resolution mass spectrometry for determining metandienone metabolites in urine. J Chromatogr B 1999; 734: 179-89 Schanzer W, Delahaut P, Geyer H, Machnik M, Horning S. Longterm detection and identiŽ cation of metandienone and stanozolol abuse in athletes by gas chromatography-high-resolution mass spectrometry. J Chromatogr B 1996; 687: 93-108 Horning S, Schanzer W. Steroid screening using GC/HRMS. In: Schanzer W, Geyer H, Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis (4). Proceedings of the 14th Cologne Workshop on Dope Analysis 17th to 22nd March 1993. Koln: Sport und Buch Strauss, 1997: 261-74 Bowers LD, Borts DJ. Evaluation of selected-ion storage ion-trap mass spectrometry for detecting urinary anabolic agents. Clin Chem 1997; 43: 1033-9 Bowers LD, Borts DJ. Separation and conŽ rmation of anabolic steroids with quadrupole ion trap tandem mass spectrometry. J Chromatogr B 1996; 687: 69-78 Kicman AT, Cowan DA, Myhre L, Nilsson S, Tomten S, Oftebro H. Effect on sports drug tests of ingesting meat from steroid (methenolone)-treated livestock. Clin Chem 1994; 40: 2084-7 Goudreault D, Masse R. Studies on anabolic steroids 4. IdentiŽ cation of new urinary metabolites of methenolone acetate (Primobolan) in human by gas chromatography/mass spectrometry. J Steroid Biochem Mol Biol 1990; 37: 137-54 Schanzer W, Opfermann G, Donike M. 17-Epimerization of 17amethyl anabolic steroids in humans: metabolism and synthesis of 17a-hydroxy-17b-methyl steroids. Steroids 1992; 57: 537-50 Expert Committee on Nandrolone. Nandrolone Review. Report to UK Sport, January 2000. Available from UK Sport, London Thevis M, Opfermann G, Schmickler H, Schanzer W. Mass spectrometry of steroid glucuronide conjugates. I. Electron impact fragmentation of 5a-/5b-androstan-3a-ol-17-one glucuronides, 5a-estran-3a-ol-17-one glucuronide and deuterium-labelled analogues. J Mass Spectrom 2001; 36: 159-68 Thevis M, Opfermann G, Schmickler H, Schanzer W. Mass spectrometry of steroid glucuronide conjugates. II: Electron impact fragmentation of 3-keto-4-en- and 3-keto-5alpha-steroid-17-O-beta glucuronides and 5alpha-steroid-3alpha,17beta-diol 3- and 17glucuronides. J Mass Spectrom 2001; 36: 998-1012 Kuuranne T, Vahermo M, Leinonen A, Kostiainen R. Electrospray and atmospheric pressure chemical ionization tandem mass spectrometric behavior of eight anabolic steroid glucuronides. J Am Soc Mass Spectrom 2000; 11: 722-30 Choi MH, Kim KR, Chung BC. Simultaneous determination of urinary androgen glucuronides by high temperature gas chromatography-mass spectrometry with selected ion monitoring. Steroids 2000; 65: 54-9 Borts DJ, Bowers LD. Direct measurement of urinary testosterone and epitestosterone conjugates using high-performance liquid chromatography/tandem mass spectrometry. J Mass Spectrom 2000; 35: 50-61

353

140 Bowers LD, Sanaullah. Direct measurement of steroid sulfate and glucuronide conjugates with high-performance liquid chromatography mass spectrometry. J Chromatogr B 1996; 687: 61-8 141 Kruskemper HL. Creatine Metabolism. Anabolic Steroids. New York: Academic Press, 1968: 58-60 142 Townsley JD, Brodie HJ. Studies on the mechanism of estrogen biosynthesis. IV. Ovarian metabolism of estr-4-ene-3,17-dione. Biochim Biophys Acta 1967; 144: 440-5 143 Farnsworth WE. Metabolism of 19-nortestosterone by human prostate. Steroids 1966; 8: 825-44 144 Cole PA, Robinson CH. Mechanism and inhibition of cytochrome P-450 aromatase. J Med Chem 1990; 33: 2933-42 145 Beusen DD, Carrell HL, Covey DF. Metabolism of 19-methylsubstituted steroids by human placental aromatase. Biochemistry 1987; 26: 7833-41 146 Silberzahn P, Dehennin L, Zwain I, Reiffsteck A. Gas chromatography-mass spectrometry of androgens in equine ovarian follicles at ultrastructurally deŽ ned stages of development. IdentiŽ cation of 19-nortestosterone in follicular  uid. Endocrinology 1985; 117: 2176-81 147 Reznik Y, Herrou M, Dehennin L, Lemaire M, Leymarie P. Rising plasma levels of 19-nortestosterone throughout pregnancy: determination by radioimmunoassay and validation by gas chromatography-mass spectrometry. J Clin Endocrinol Metab 1987; 64: 1086-8 148 de Boer D, Bensink SN, Borggreve AR, VanOoijen RD, Maes RAA. ProŽ ling 19-norsteroids III: GC/MS/MS analysis of 19norsteroids during pregnancy. In: Donike M, Geyer H, Gotzmann A, Mareck-Engelke U, Rauth S, eds. 10th Cologne Workshop on Dope Analysis: Proceedings, 7th to 12th June 1992. Koln: Sport und Buch Strauss, 1993: 121-2 149 Mareck-Engelke U, Geyer H, Schanzer W. The interpretation of female steroid proŽ les. In: Schanzer W, Geyer H, Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis (5). Proceedings of the Manfred Donike Workshop: 15th Cologne Workshop on Dope Analysis 23rd to 28th February 1997. Koln: Sport und Buch Strauss, 1998: 51-70 150 Van Eenoo P, Delbecke FT, de Jong FH, De Backer P. Urinary metabolites of endogenous nandrolone in women: a case study. In: Schanzer W, Geyer H, Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis (6). Proceedings of the Manfred Donike Workshop: 16th Cologne Workshop on Dope Analysis 15th to 20th March 1998. Koln: Sport und Buch Strauss, 1999: 105-17 151 Mareck-Engelke U, Schultze G, Geyer H, Schanzer W. The appearance of urinary 19-norandrosterone during pregnancy. Eur J Sports Sci 2002; 2: 1-7 152 Dehennin L, Bonnaire Y, Plou P. Urinary excretion of 19norandrosterone of endogenous origin in man: quantitative analysis by gas chromatography mass spectrometry. J Chromatogr B 1999; 721: 301-7 153 Le Bizec B, Monteau F, Gaudin I, Andre F. Evidence for the presence of endogenous 19-norandrosterone in human urine. J Chromatogr B 1999; 723: 157-72 154 Ciardi M, Ciccoli R, Barbarulo MV, Nicoletti R. Presence of norandrosterone in ‘normal’ urine samples. In: Schanzer W, Geyer H, Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis (6). Proceedings of the Manfred Donike Workshop: 16th Cologne Workshop on Dope Analysis 15th to 20th March 1998. Koln: Sport und Buch Strauss, 1999: 97-104 Ann Clin Biochem 2003; 40: 321–356

354

Kicman and Gower

155 Reznik Y, Dehennin L, CofŽ n C, Mahoudeau J, Leymarie P. Urinary nandrolone metabolites of endogenous origin in man: a conŽ rmation by output regulation under human chorionic gonadotropin stimulation. J Clin Endocrinol Metab 2001; 86: 146-50 156 Saugy M, Robinson N, Cardis C, et al. Nandrolone metabolites in football players: utility for in and out of competition testing. In: Schanzer W, Geyer H, Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis (7). Proceedings of the Manfred Donike Workshop. Ko¨ln: Sport und Buch Strauss, 1999 157 Robinson N, Taroni F, Saugy M, Ayotte C, Mangin P, Dvorak J. Detection of nandrolone metabolites in urine after a football game in professional and amateur players: a Bayesian comparison. Forensic Sci Int 2001; 122: 130-5 158 Le Bizec B, Bryand F, Gaudin I, Monteau F, Poulain F, Andre F. Endogenous nandrolone metabolites in human urine. Two-year monitoring of male professional soccer players. J Anal Toxicol 2002; 26: 43-7 159 Schmitt N, Flament MM, Goubault C, Legros P, Grenier-Loustalot MF, Denjean A. Nandrolone excretion is not increased by exhaustive exercise in trained athletes. Med Sci Sports Exerc 2002; 34: 1436-9 160 Le Bizec B, Bryand F, Gaudin I, Monteau F, Poulain F, Andre F. Endogenous nandrolone metabolites in human urine: preliminary results to discriminate between endogenous and exogenous origin. Steroids 2002; 67: 105-10 161 Mathurin JC, Herrou V, Bourgogne E, Pascaud L, de Ceaurriz J. Gas chromatography-combustion-isotope ratio mass spectrometry analysis of 19-norsteroids: application to the detection of a nandrolone metabolite in urine. J Chromatogr B 2001; 759: 26775 162 Desroches MC, Mathurin JC, Richard Y, Delahaut P, de Ceaurriz J. Urinary 19-norandrosterone puriŽ cation by immunoafŽ nity chromatography: application to gas chromatography/combustion/ isotope ratio mass spectrometric analysis. Rapid Commun Mass Spectrom 2002; 16: 370-4 163 Le Bizec B, Gaudin I, Monteau F, et al. Consequence of boar edible tissue consumption on urinary proŽ les of nandrolone metabolites. I. Mass spectrometric detection and quantiŽ cation of 19-norandrosterone and 19-noretiocholanolone in human urine. Rapid Commun Mass Spectrom 2000; 14: 1058-65 164 Machnik M, Scrader Y, Schanzer W. Plasma levels of 19norsteroids after oral and buccal administration of norandrostenedione. In: Schanzer W, Geyer H, Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis (9). Proceedings of the Manfred Donike Workshop: 19th Cologne Workshop on Dope Analysis, 18th to 23rd March 2001. Koln: Sport und Buch Strauss, 2001: 125-31 165 Belkien L, Schurmeyer T, Hano R, Gunnarsson PO, Nieschlag E. Pharmacokinetics of 19-nortestosterone esters in normal men. J Steroid Biochem Mol Biol 1985; 22: 623-9 166 Wijnand HP, Bosch AMG, Donker CW. Pharmacokinetic parameters of nandrolone (19-nortestosterone) after intramuscular administration of nandrolone decanoate (deca-durabolin) to healthy-volunteers. Acta Endocrinol 1985; 110: 19-30 167 Minto CF, Howe C, Wishart S, Conway AJ, Handelsman DJ. Pharmacokinetics and pharmacodynamics of nandrolone esters in oil vehicle: effects of ester, injection site and injection volume. J Pharmacol Exp Ther 1997; 281: 93-102 168 Cowan DA, Kicman AT. Doping in sport: misuse, analytical tests, and legal aspects [editorial]. Clin Chem 1997; 43: 1110-3 Ann Clin Biochem 2003; 40: 321–356

169 Donike M, Barwald KR, Klostermann K, Schanzer W, Zimmermann J. Nachweis von exogenem Testosteron [Detection of endogenous testosterone]. In: Heck H, Hollmann W, Liesen H, et al., eds. Sport: Leistung und Gesundheit, Kongressbd. Dtsch. Sporta¨rztekongress. Ko¨ln: Deutscher A¨rtze-Verlag, 1983: 293-8 170 Dehennin L. Secretion by the human testis of epitestosterone, with its sulfoconjugate and precursor androgen 5-androstene3beta,17alpha-diol. J Steroid Biochem Mol Biol 1993; 44: 171-7 171 Kicman AT, Coutts SB, Cowan DA, et al. Adrenal and gonadal contributions to urinary excretion and plasma concentration of epitestosterone in men: effect of adrenal stimulation and implications for detection of testosterone abuse. Clin Endocrinol (Oxf) 1999; 50: 661-8 172 Dehennin L, Matsumoto AM. Long-term administration of testosterone enanthate to normal men: alterations of the urinary proŽ le of androgen metabolites potentially useful for detection of testosterone misuse in sport. J Steroid Biochem Mol Biol 1993; 44: 179-89 173 Palonek E, Gottlieb C, Garle M, Bjorkhem I, Carlstrom K. Serum and urinary markers of exogenous testosterone administration. J Steroid Biochem Mol Biol 1995; 55: 121-7 174 Anderson RA, Wallace AM, Kicman AT, Wu FC. Comparison between testosterone oenanthate-induced azoospermia and oligozoospermia in a male contraceptive study. IV. Suppression of endogenous testicular and adrenal androgens. Hum Reprod 1997; 12: 1657-62 175 Dehennin L, Jondet M, Scholler R. Androgen and 19-norsteroid proŽ les in human preovulatory follicles from stimulated cycles: an isotope dilution-mass spectrometric study. J Steroid Biochem 1987; 26: 399-405 176 Oftebro H. Evaluating an abnormal urinary steroid proŽ le. Lancet 1992; 339: 941-2 177 Kicman AT, Oftebro H, Walker C, Norman N, Cowan DA. Potential use of ketoconazole in a dynamic endocrine test to differentiate between biological outliers and testosterone use by athletes. Clin Chem 1993; 39: 1798-803 178 Oftebro H, Jensen J, Mowinckel P, Norli HR. Establishing a ketoconazole suppression test for verifying testosterone administration in the doping control of athletes. J Clin Endocrinol Metab 1994; 78: 973-7 179 Raynaud E, Audran M, Pages JC, et al. Urinary-excretion of testosterone and epitestosterone glucuronides in preadolescents and adolescents. Pathol Biol 1993; 41: 159-63 180 Raynaud E, Audran M, Pages JC, et al. Determination of urinary testosterone and epitestosterone during pubertal development: a cross-sectional study in 141 normal male subjects. Clin Endocrinol 1993; 38: 353-9 181 Garle M, Ocka R, Palonek E, Bjorkhem I. Increased urinary testosterone epitestosterone ratios found in Swedish athletes in connection with a national control program: evaluation of 28 cases. J Chromatogr B 1996; 687: 55-9 182 Donike M, Rauth S, Mareck-Engelke U, Geyer H, Nitschke R. Evaluation of longitudinal studies, the determination of subject based reference ranges of the testosterone/epitestosterone ratio. In: Donike M, Geyer H, Gotzmann A, Mareck-Engelke U, Rauth S, eds. Recent Advances in Doping Analysis. Proceedings of the 11th Cologne Workshop on Dope Analysis 7th to 12th March 1993. Koln: Sport und Buch Strauss, 1994: 33-9 183 Donike M, Mareck-Engelke U, Rauth S. Statistical evaluation of longitudinal studies, part 2: the usefulness of subject based reference ranges in steroid proŽ ling. In: Donike M, Geyer H,

Anabolic steroids in sport

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis. Proceedings of the 12th Cologne Workshop on Dope Analysis 10-15th April 1994. Koln: Sport und Buch Strauss, 1995: 157-65 Harris EK. Some theory of reference values. I. StratiŽ ed (categorized) normal ranges and a method for following an individual’s clinical laboratory values. Clin Chem 1975; 21: 145764 Harris EK, Cooil BK, Shakarji G, Williams GZ. On the use of statistical models of within-person variation in long-term studies of healthy individuals. Clin Chem 1980; 26: 383-91 Catlin DH, Hatton CK, Starcevic SH. Issues in detecting abuse of xenobiotic anabolic steroids and testosterone by analysis of athletes’ urine. Clin Chem 1997; 43: 1280-8 Palonek E, Gottlieb C, Garle M, Bjorkhem I, Carlstrom K. Serum and urinary markers of exogenous testosterone administration. J Steroid Biochem Mol Biol 1995; 55: 121-7 Carlstrom K, Palonek E, Garle M, Oftebro H, Stanghelle J, Bjorkhem I. Detection of testosterone administration by increased ratio between serum concentrations of testosterone and 17alphahydroxyprogesterone. Clin Chem 1992; 38: 779-84 Dehennin L. On the origin of physiologically high ratios of urinary testosterone to epitestosterone: consequences for reliable detection of testosterone administration by male athletes. J Endocrinol 1994; 142: 353-60 Perry PJ, MacIndoe JH, Yates WR, Scott SD, Holman TL. Detection of anabolic steroid administration: ratio of urinary testosterone to epitestosterone vs the ratio of urinary testosterone to luteinizing hormone. Clin Chem 1997; 43: 731-5 Yap BK, Kazlauskas R, Elghazi K, Johnston GAR, Weatherby RP. ProŽ ling of urinary testosterone and luteinizing hormone in exercise-stressed male athletes, using gas chromatography mass spectrometry and enzyme immunoassay techniques. J Chromatogr B 1996; 687: 117-25 Southan GJ, Brooks RV, Cowan DA, Kicman AT, Unnadkat N, Walker CJ. Possible indices for the detection of the administration of dihydrotestosterone to athletes. J Steroid Biochem Mol Biol 1992; 42: 87-94 Kicman AT, Coutts SB, Walker CJ, Cowan DA. Proposed conŽ rmatory procedure for detecting 5alpha-dihydrotestosterone doping in male athletes. Clin Chem 1995; 41: 1617-27 Donike M, Ueki M, Kuroda Y, et al. Detection of dihydrotestosterone (DHT) doping: alterations in the steroid proŽ le and reference ranges for DHT and its 5alpha-metabolites [editorial]. J Sports Med Phys Fitness 1995; 35: 235-50 Coutts SB, Kicman AT, Hurst DT, Cowan DA. Intramuscular administration of 5alpha-dihydrotestosterone heptanoate: changes in urinary hormone proŽ le. Clin Chem 1997; 43: 2091-8 Dehennin L, Ferry M, Lafarge P, Peres G, Lafarge JP. Oral administration of dehydroepiandrosterone to healthy men: alteration of the urinary androgen proŽ le and consequences for the detection of abuse in sport by gas chromatography-mass spectrometry. Steroids 1998; 63: 80-7 Bosy TZ, Moore KA, Poklis A. The effect of oral dehydroepiandrosterone (DHEA) on the urine testosterone/epitestosterone (T/E) ratio in human male volunteers. J Anal Toxicol 1998; 22: 455-9 Bowers LD. Oral dehydroepiandrosterone supplementation can increase the testosterone/epitestosterone ratio. Clin Chem 1999; 45: 295-7 Levesque J-F, Ayotte C. Criteria for the detection of androstenedione oral administration. In: Schanzer W, Geyer H,

200

201

202

203

204

205

206

207

208

209

210

211 212

355

Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis (7). Proceedings of the Manfred Donike Workshop: 17th Cologne Workshop on Dope Analysis, 14th to 19th March 1999. Koln: Sport und Buch Strauss, 1999 Goudreault D, Bherer P, Levesque JF, Pourier D, Ayotte C. Androstenedione metabolism: end of the story. In: Schanzer W, Geyer H, Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis (9). Proceedings of the Manfred Donike Workshop: 19th Cologne Workshop on Dope Analysis, 18th to 23rd March 2001. Koln: Sport und Buch Strauss, 2001: 73-83 Van Eenoo P, Delbeke FT, Desmet N, De Backer P. Excretion studies with 4-androstene-3,17-dione. In: Schanzer W, Geyer H, Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis (6). Proceedings of the Manfred Donike Workshop: 16th Cologne Workshop on Dope Analysis 15th to 20th March 1998. Koln: Sport und Buch Strauss, 1999: 171-80 Laidler P, Cowan DA, Hider RC, Kicman AT. New decision limits and quality-control material for detecting human chorionic gonadotropin misuse in sports. Clin Chem 1994; 40: 1306-11 Becchi M, Aguilera R, Farizon Y, Flament MM, Casabianca H, James P. Gas chromatography/combustion/isotope-ratio mass spectrometry analysis of urinary steroids to detect misuse of testosterone in sport. Rapid Commun Mass Spectrom 1994; 8: 304-8 Aguilera R, Becchi M, Grenot C, Casabianca H, Hatton CK. Detection of testosterone misuse: comparison of two chromatographic sample preparation methods for gas chromatographiccombustion/isotope ratio mass spectrometric analysis. J Chromatogr B 1996; 687: 43-53 Aguilera R, Becchi M, Casabianca H, et al. Improved method of detection of testosterone abuse by gas chromatography/ combustion/isotope ratio mass spectrometry analysis of urinary steroids. J Mass Spectrom 1996; 31: 169-76 Aguilera R, Catlin DH, Becchi M, et al. Screening urine for exogenous testosterone by isotope ratio mass spectrometric analysis of one pregnanediol and two androstanediols. J Chromatogr B 1999; 727: 95-105 Shackleton CH, Phillips A, Chang T, Li Y. ConŽ rming testosterone administration by isotope ratio mass spectrometric analysis of urinary androstanediols. Steroids 1997; 62: 379-87 Horning S, Geyer H, Machnik M, Schanzer W, Hilkert A, Oesselmann J. Detection of exogenous testosterone by 13C/12C analysis. In: Schanzer W, Geyer H, Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis. Proceedings of the 14th Cologne Workshop on Dope Analysis. Koln: Sport und Buch Strauss, 1997: 275-83 Cowan DA, Kicman AT, Walker CJ, Wheeler MJ. Effect of administration of human chorionic gonadotrophin on criteria used to assess testosterone administration in athletes. J Endocrinol 1991; 131: 147-54 de Boer D, de Jong EG, van Rossum JM, Maes RA. Doping control of testosterone and human chorionic gonadotrophin: a case study [published erratum appears in Int J Sports Med 1991; 12: 430]. Int J Sports Med 1991; 12: 46-51 Kicman AT, Brooks RV, Cowan DA. Human chorionic gonadotrophin and sport. Br J Sports Med 1991; 25: 73-80 Brooks RV, Collyer SP, Kicman AT, Southan GJ, Wheeler MA. HCG doping in sport and methods for its detection. In: Bellotti P, Benzi G, Ljungqvist A, eds. IInd International Athletic Foundation World Symposium on Doping in Sport: OfŽ cial Proceedings, Monte Ann Clin Biochem 2003; 40: 321–356

356

213

214

215

216 217

218

Kicman and Gower Carlo, 5-7 June 1989: International Athletic Foundation (IAF), 1990: 37-45 Stenman UH, Unkila-Kallio L, Korhonen J, Alfthan H. Immunoprocedures for detecting human chorionic gonadotropin: clinical aspects and doping control. Clin Chem 1997; 43: 1293-8 Laidler P, Cowan DA, Hider RC, Keane A, Kicman AT. Tryptic mapping of human chorionic gonadotropin by matrix-assisted laser desorption/ionization mass spectrometry. Rapid Commun Mass Spectrom 1995; 9: 1021-6 Liu CL, Bowers LD. ImmunoafŽ nity trapping of urinary human chorionic gonadotropin and its high-performance liquid chromatographic-mass spectrometric conŽ rmation. J Chromatogr B 1996; 687: 213-20 Kicman AT, Brooks RV, Collyer SC, et al. Criteria to indicate testosterone administration. Br J Sports Med 1990; 24: 253-64 Aguilera R, Hatton CK, Catlin DH. Detection of epitestosterone doping by isotope ratio mass spectrometry. Clin Chem 2002; 48: 629-36 Geyer H, Schanzer W, Donike M. Probenecid as masking agent in dope control: inhibition of the urinary excretion of steroid glucuronides. In: Donike M, Geyer H, Gotzmann A, Mareck-

Ann Clin Biochem 2003; 40: 321–356

219 220

221

222

Engelke U, Rauth S, eds. 10th Cologne Workshop on Dope Analysis: Proceedings, 7th to 12th June 1992. Koln: Sport und Buch Strauss, 1993: 141-51 Randall VA. Role of 5alpha-reductase in health and disease. Baillieres Clin Endocrinol Metab 1994; 8: 405-31 Bull HG, GarciaCalvo M, Andersson S, et al. Mechanism-based inhibition of human steroid 5alpha-reductase by Ž nasteride: enzyme-catalyzed formation of NADP-dihydroŽ nasteride, a potent bisubstrate analog inhibitor. J Am Chem Soc 1996; 118: 2359-65 Geyer H, Nolteernsting E, Schanzer W. Finasteride: a substance for manipulation in dope control? In: Schanzer W, Geyer H, Gotzmann A, Mareck-Engelke U, eds. Recent Advances in Doping Analysis (7). Proceedings of the Manfred Donike Workshop: 17th Cologne Workshop on Dope Analysis, 14th to 19th March 1999. Koln: Sport und Buch Strauss, 1999: 71-80 Carlin JR, Hoglund P, Eriksson LO, et al. Disposition and pharmacokinetics of [14C]Ž nasteride after oral administration in humans. Drug Metab Dispos 1992; 20: 148-55

Accepted for publication 9 February 2003