Urine and Serum Sex Steroid Profile in Testosterone-Treated Transgender and Hypogonadal and Healthy Control Men

The Journal of Clinical Endocrinology & Metabolism, Jun 2018

The impact of testosterone (T) treatment on antidoping detection tests in female-to-male (F2M) transgender men is unknown. We investigated urine and serum sex steroid and luteinizing hormone (LH) profiles in T-treated F2M men to determine whether and, if so, how they differed from hypogonadal and healthy control men.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://academic.oup.com/jcem/article-pdf/103/6/2277/25011469/jc.2018-00054.pdf

Urine and Serum Sex Steroid Profile in Testosterone-Treated Transgender and Hypogonadal and Healthy Control Men

J Clin Endocrinol Metab, June Urine and Serum Sex Steroid Profile in Testosterone-Treated Transgender and Hypogonadal and Healthy Control Men Sasha Savkovic 1 Sarina Lim 1 Veena Jayadev 1 Ann Conway 1 Leo Turner 1 Douglas Curtis 0 Catrin Goebel 0 David J. Handelsman 1 0 Australian Sports Drug Testing Laboratory, National Measurement Institute , Sydney, New South Wales 2113 , Australia 1 Andrology Department, Concord Hospital, ANZAC Research Institute , Sydney, New South Wales 2139 , Australia Background: The impact of testosterone (T) treatment on antidoping detection tests in female-tomale (F2M) transgender men is unknown. We investigated urine and serum sex steroid and luteinizing hormone (LH) profiles in T-treated F2M men to determine whether and, if so, how they differed from hypogonadal and healthy control men. Method: Healthy transgender (n = 23) and hypogonadal (n = 24) men aged 18 to 50 years treated with 1000 mg injectable T undecanoate provided trough urine and blood samples and an additional earlier postinjection sample (n = 21). Healthy control men (n = 20) provided a single blood and urine sample. Steroids were measured by mass spectrometry-based methods in urine and serum, LH by immunoassay, and uridine 50-diphospho-glucuronosyltransferase 2B17 genotype by polymerase chain reaction. Results: Urine LH, human chorionic gonadotropin, T, epitestosterone (EpiT), androsterone (A), etiocholanolone (Etio), A/Etio ratio, dehydroepiandrosterone (DHEA), dihydrotestosterone (DHT), and 5a,3a- and 5b,3a-androstanediols did not differ between groups or by time since last T injection. Urine T/EpiT ratio was ,4 in all controls and 12/68 (18%) samples from T-treated men, but there was no difference between T-treated groups. Serum estradiol, estrone, and DHEA were higher in transgender men, and serum T and DHT were higher in earlier compared with trough blood samples, but serum LH, follicle-stimulating hormone, and 3a- and 3b,5a-diols did not differ between groups. Conclusion: Urine antidoping detection tests in T-treated transgender men can be interpreted like those of T-treated hypogonadal men and are unaffected by time since last T dose. Serum steroids are more sensitive to detect exogenous T administration early but not later after the last T dose. (J Clin Endocrinol Metab 103: 2277-2283, 2018) - Tof transgender people includes recognizing the dehe growing awareness of the health and welfare needs sirability of participation in common social pursuits such as sports. Participation of transgender athletes in elite sports provides positive role models and reinforcement for successful integration into society for people who have previously felt marginalized. The major international sports bodies, such as the International Olympic Committee and World Anti-Doping Agency (WADA), have led efforts to facilitate the inclusion of transgender athletes in major elite sporting competitions governed by WADA antidoping rules that also prohibit Abbreviations: 5a/5b, 5a,3a-androstanediol/5b,3a-androstanediol ratio; A, androsterone; ABP, Athlete Biological Passport; ANOVA, analysis of variance; BMI, body mass index; BSA, body surface area; CC, homozygous wild-type; CJ, heterozygous; DHEA, dehydroepiandrosterone; DHT, dihydrotestosterone; E1, estrone; E2, estradiol; EpiT, epitestosterone; Etio, etiocholanolone; FSH, follicle-stimulating hormone; hCG, human chorionic gonadotropin; JJ, homozygous deletor; LH, luteinizing hormone; T, testosterone; UGT2B17, uridine 50-diphospho-glucuronosyltransferase 2B17; WADA, World Anti-Doping Agency. doi: 10.1210/jc.2018-00054 https://academic.oup.com/jcem use of testosterone (T). Under the WADA Code, T use is permitted with strictly regulated approval by a Therapeutic Use Exemption for athletes with well-defined medical conditions, notably men with hypogonadism due to reproductive system pathology ( 1 ) or T treatment of female-to-male transgender men ( 2 ). Yet T-treated transgender and hypogonadal men must submit urine antidoping tests when competing in WADA-compliant elite sports and nonsporting occupational urine drug testing (e.g., security-related occupations), which may detect the use of exogenous T. However, the patterns of urine and serum sex steroids and related gonadotropin profiles in T-treated transgender men are not reported. Therefore, this observational study aimed to provide urine and serum sex steroid and gonadotropin profiles in transgender and hypogonadal men treated with standard replacement doses of T as well as healthy, untreated control men to determine whether and, if so, how they differed. Methods The study recruited transgender and hypogonadal men being treated with injectable T undecanoate (1000 mg in 4 mL arachis oil vehicle, Reandron; Bayer, Leverkusen, Germany) at the Andrology Department, Concord Hospital clinic. They were 18 to 50 years old and did not have chronic medical illness, use regular prescribed medication, have major psychiatric disease or psychological condition that limited understanding and compliance with the protocol, or have a history of androgen or other drug abuse. The T-treated men recruited for the study were not paid for their participation, which usually coincided with their regular clinic visits. Healthy, age-matched controls were recruited by local advertising and offered $40 for their time and travel costs for a single visit. The study was approved by the Sydney Local Health District (Concord Hospital) Human Ethics Committee, and all participants were required to provide written, informed consent before participation. Participants provided a urine (50 mL) and blood (15 mL) sample at time of the next T dose (trough), and some volunteered to provide an additional urine and blood sample at an earlier (i.e., nontrough) time after a T dose. Urine and serum samples were stored frozen (220°C) until run in single batches for all analytes. Serum T, dihydrotestosterone (DHT), estrone (E1), estradiol (E2), dehydroepiandrosterone (DHEA), 3a,5a-androstanediol, and 3b,5a-androstanediol were measured by validated liquid chromatography-mass spectrometry methods as originally described ( 3 ) and modified to include ultrapressure liquid chromatography and different liquid/liquid extraction ( 4 ), with extensive longitudinal quality control ( 5 ) reported previously. Urine T, epitestosterone (EpiT), androsterone (A), etiocholanolone (Etio), DHEA, DHT, 5a,3a-androstanediol, 5b,3aandrostanediol, T/EpiT, A/Etio, and 5a,3a-androstanediol/ 5b,3a-androstanediol (5a/5b) ratios were measured by gas chromatography–mass spectrometry methods in a WADAaccredited antidoping laboratory that used their routine gas chromatography-mass spectrometry steroid profiling ( 6 ) according to WADA standard requirements for measurement of endogenous androgens ( 7 ) and maintained ongoing accreditation by participating in WADA’s External Quality Assessment Scheme ( 8 ). Serum luteinizing hormone (LH) and folliclestimulating hormone (FSH) and urine LH and human chorionic gonadotropin (hCG) were measured by human gonadotropin immunoassays. Uridine 50-diphospho-glucuronosyltransferase 2B17 (UGT2B17) genotyping was performed by polymerase chain reaction ( 9 ) classifying genotypes into homozygous wildtype (CC), heterozygous (CJ), and homozygous deletor (JJ). Data were analyzed by analysis of variance (ANOVA) and t test via NCSS 11 Statistical Software (NCSS, Kaysville, UT). When there were overall significant differences between groups, post hoc differences were located by suitable linear contrast to determine whether T-treated groups differed from each other or from the healthy controls. The effects of time since last injection as analyzed according to either the visit (as a categorical variable, trough vs earlier postinjection visit) or time since last injection as a covariate. The potential confounding influence of anthropometric differences between groups was evaluated by considering the differences in age, height, weight, body mass index (BMI, in kg/m2), and body surface area [BSA, calculated with the Gehan George equation ( 10 )] as covariates. The impact of ovariectomy on transgender men was evaluated according to whether they had undergone the surgery or the time since ovariectomy. Results The descriptive data of the transgender (n = 23), hypogonadal (n = 24), and healthy controls (n = 20) are in Table 1. T-treated men provided the trough samples at a median time of 84 days since last T injection, and the 21 additional nontrough blood and urine samples at a median time of 29 days after the last T injection and the time since last injection did not differ between T-treated groups. Transgender men had been treated with T for a median of 3 years (range 1 to 11 years), with eight having undergone oophorectomy at a median time of 6.5 (range 2 to 17) years previously. The transgender men were younger, smaller, and lighter than the hypogonadal and control men, but BMI did not differ between groups (Table 1). The hypogonadal men had primary (hypergonadotropic) hypogonadism in 16 (Klinefelter syndrome 6, orchiectomy 5, postcancer 4, cryptorchidism 2, unknown 1) and secondary (hypogonadotropic) hypogonadism in 8 men (pituitary tumor 3, isolated hypogonadotropic hypogonadism 5). Differences between groups Serum E1, E2, and DHEA were higher in transgender than in hypogonadal or control men, whereas serum T was higher and serum LH lower in T-treated than in control men (Table 2). These significant differences persisted after adjustment for significant effects of body size for serum E1 (BSA, P = 0.013; BMI, P = 0.005; weight, P = 0.009; but not height, P = 0.64) and serum DHEA (BSA, P = 0.004; BMI, P = 0.057; weight, P = 0.01; height, P = 0.03) but not E2 (BSA, P = 0.13; BMI, P = 0.056; weight, P = 0.11; height, P = 0.98). Serum DHT, 3a- and 3b,5a diols, and serum FSH Anthropometric Data Pa <0.001 <0.001 Data presented as mean 6 standard deviation, with median and range (minimum, maximum) below. Boldface indicates statistically significant differences. aOverall P value from one-way ANOVA for comparison between three groups. bP value for linear contrast comparing both T-treated groups against controls. cP value for linear contrast comparing the two T-treated groups against each other. Pc 0.003 <0.001 <0.001 0.23 <0.001 Pc 0.19 0.97 0.038 <0.001 0.003 0.15 0.17 did not differ between groups. No covariates (age, height, weight, BMI, BSA) were significantly correlated with any serum hormone except for serum DHEA, which was inversely correlated with age (r = 20.27, P = 0.01). Urine LH and EpiT were lower and T/EpiT ratio higher in T-treated groups compared with control men, but there were no significant differences in these variables between transgender and hypogonadal men (Table 3). Urine T, A, Etio, DHEA, DHT, 5a,3a and 5b,3a diols, A/Etio and 5a/5b ratios, and hCG did not differ between groups. No covariates (age, height, weight, BMI, BSA) were significantly correlated with any urine hormone except for T/EpiT ratio, which was weakly correlated with age (r = 20.22, P = 0.04). In the Serum Hormone Data transgender men, the minority who had undergone ovariectomy had significantly lower serum E2 (21 vs 37 pg/mL, P = 0.006) and E1 (126 vs 169 pg/mL, P = 0.024) and higher serum DHEA (5.7 vs 3.1 mg/mL, P = 0.06). Time since ovariectomy was significantly correlated (inversely) only with serum DHEA (r = 20.65, P = 0.016). Serum and urine DHEA were correlated (r = 0.30, P = 0.005) overall, and the correlation did not differ between groups. Time since last injection When time since last injection was analyzed according to the visit date (trough vs earlier), serum T and DHT cP value for linear contrast comparing the two T-treated groups against each other. Urine Hormone Data Data presented as mean 6 standard deviation, with median and range (minimum, maximum) below. Boldface indicates statistically significant differences. aOverall P value from one-way ANOVA for comparison between three groups. bP value for linear contrast comparing both T-treated groups against controls. cP value for linear contrast comparing the two-T treated groups against each other. dRatio of urine T to urine EpiT. eRatio of urine A to urine Etio. fRatio of urine 3a,5a-androstanediol to 3b,5a-androstanediol. were higher on earlier samples compared with trough samples, but there were no other significant differences between other serum or any urine concentrations of steroids or gonadotropins (Figs. 1 and 2, Supplemental Fig. 1). When time since last injection was investigated as a continuous variable by covariance analysis, time since injection was a significant covariate for analysis of serum T, DHT, E1, E2, urine T, and T/EpiT ratio but not for other serum or any urine steroids or gonadotropins. Effects of UGT2B17 genotype The distribution of UGT2B17 genotypes was similar between transgender (CC 9, CJ 9, JJ 5), hypogonadal (CC 8, CJ 11, JJ 5), and control (CC 7, CJ 13, JJ 0) men (P = 0.16, extended Fisher test), with an overall prevalence of 15% for the homozygous deletion genotype. Urine T/EpiT ratio was ,4 (the threshold for the WADA-approved urine antidoping screening test to detect exogenous T exposure) in all controls and 12/68 (18%) samples from T-treated men, but there was no significant difference between T-treated groups in the proportions .4 or ,4. Up to 77 days after a T injection, urine T/EpiT ratio was .4 in all samples from T-treated individuals (Fig. 3). From 77 days onward, among the 28 samples 11 had a T/EpiT ratio ,4 (39%) and 17 had a T/EpiT ratio .4 (61%). Similarly, after 77 days among the 6 samples from T-treated men with the deletion genotype (JJ), 5 had a T/EpiT ratio ,4 (83%), whereas among 22 samples from men without deletion genotype (CC or CJ), the T/EpiT ratio was ,4 in 6 (27%) and .4 in 16 (73%). The deletion genotype was also associated with significantly lower urine T and 5b,3a-androstanediol, with a higher urine 5a/5b ratio, but there were no other significant differences between those with and without the deletion genotype for any other serum and urine steroids. Discussion The current study indicates that the sex steroid and gonadotropin profiles of T-treated transgender men are very similar to those of T-treated hypogonadal men while also confirming that both T-treated groups have distinctive differences from untreated healthy eugonadal control men. Differences between the T-treated groups in the urine or serum sex steroid profiles were limited to higher serum E1, E2, DHEA, and LH in transgender men compared with hypogonadal men. These differences were attributable mostly to transgender men who had undergone ovariectomy. The higher E1, E2, and DHEA concentrations in transgender men may be related partly to the effects of their smaller body size, represented by BSA, BMI, or weight in this study, as reported in postmenopausal women ( 11 ). Another possible explanation for the unexpectedly higher serum (but not urine) DHEA in transgender men is use of exogenous DHEA, available as an over-the-counter product. However, no participants admitted to using DHEA, which would be expected to increase both serum and urine measurements; however, undisclosed use of nonprescription drugs such as DHEA cannot be excluded. The present findings are consistent with the experience reported elsewhere for T treatment of transgender men ( 12 ), including the use of injectable T undecanoate ( 13 ). However, we were unable to locate previous reports comparing T treatment in hypogonadal and transgender men or urine steroid findings in transgender individuals. The major differences attributable to T treatment (i.e., compared with untreated controls) are higher serum T, E1, and E2 and lower serum LH as well as lower urinary EpiT and LH and higher urine T/EpiT ratio. These differences due to T treatment were reflected in time-dependent changes in the analytes according to time since last T dose, whereby the changes were distinctive #11 weeks after T undecanoate injection and most evident in the first month after treatment but with the differences gradually dissipating with time since last T injection. Time since injection was an important predictor of serum and urine T, serum DHT, and urine T/EpiT ratio but not of other serum or urine sex steroids. For antidoping purposes, these differences indicate the potential advantages and limitations of the greater sensitivity of the serum steroid profile for detection of T doping. Although the sensitivity is greater than for urine testing early after the last injection, it drops with greater time since last injection, an essentially uncontrollable variable in the antidoping context. Therefore, although monitoring serum steroids increases the sensitivity of the steroidal module of the Athlete Biological Passport (ABP), its advantages are limited by the (unknown) time since last T injection and is lost later, whereas urine testing appears less sensitive to time since last dose. The standard WADA-approved screening test to detect use of exogenous T is the urine T/EpiT ratio. Initially it was calibrated to a population-based threshold of 6, but the frequency of false negatives led to this populationbased norm being reduced to 4. If required, an elevated T/EpiT ratio was confirmed by the carbon isotope ratio test. A common population polymorphism in the UGT2B17 gene results in a deletor phenotype that produces a urine T/EpiT ratio with a mean of 0.1 instead of 1.0 ( 14 ), creating the risk of false negatives for T doping ( 15 ). The prevalence of this polymorphism varies widely between populations, ranging from 10% to 20% among Caucasians to 50% to 70% among Southeast Asian populations ( 16 ). More recently, urine T/EpiT ratio testing has been deployed in the steroidal module of the ABP, which monitors the T/EpiT ratio serially in individuals over time. This testing allows a Bayesian conversion of the population-based threshold into narrower, individual-specific reference limits. As expected, T injection increased urine T/EpiT ratio to above the population threshold of 4 for 11 weeks after an injection of 1000 mg T undecanoate, with the magnitude of the increase being inversely related to time since T injection. The increase was consistently above the population threshold of 4 for 11 weeks after the last T injection, but after that a growing minority of urine samples displayed a nondiagnostic (false negative) T/EpiT ratio of ,4. In this context, genotype increased the sensitivity and specificity of testing only to a limited degree. Therefore, reliance on serial urine samples in the steroidal module of the ABP is a more versatile and effective antidoping detection test for exogenous T. Limitations of this study include that we did not examine transdermal T gel, where the route-dependent pharmacokinetics of nonesterified T differ from those of injectable T esters and may alter the steroid profiles and consequently the utility and sensitivity of antidoping tests. The injectable T ester used in this study was 1000 mg T undecanoate in 4 mL oil vehicle ( 17 ), which is marketed worldwide except in the United States, where it is marketed as a 750-mg formulation in a 3-mL oil vehicle ( 18 ), so the findings may differ according to injection dose. However, for antidoping purposes the lack of substantial difference between the profiles of transgender and hypogonadal men in this study suggests that the urine T/EpiT ratio is most likely to operate equally effectively in both groups of T-treated men for other T products, including different transdermal gels, creams, or solutions, although this supposition warrants confirmation. We conclude that urine antidoping detection tests in T-treated transgender men can be interpreted like those of T-treated hypogonadal men and are largely unaffected by time since last T dose, unlike serum steroids, which are more sensitive to exogenous T in the early period after injection. Urine steroids may be considered a more timeintegrated measure of steroid exposure, production, and excretion that may buffer out fluctuations evident in serum steroid concentrations but at the expense of lesser sensitivity relative to blood testing. Acknowledgments Financial Support: The study was funded by the Partnership for Clean Competition. The sponsor had no role in the design, conduct, or interpretation of the study. Correspondence and Reprint Requests: David J. Handelsman, MBBS, PhD, ANZAC Research Institute, Sydney, New South Wales 2139, Australia. E-mail: . Disclosure Summary: D.J.H. declares that his institution has received grant funding support for investigator-initiated clinical studies in testosterone pharmacology from Besins Healthcare and Lawley and has provided expert testimony in testosterone litigation and to antidoping tribunals but has no other relevant disclosures. D.C. and C.G. are employees of the National Measurement Institute. The remaining authors have nothing to disclose. 1. World Anti-Doping Agency . Medical information to support the decisions of TUECs - male hypogonadism . Available at: https://www. wada-ama.org/en/resources/therapeutic-use -exemption-tue/medicalinformation-to-support-the-decisions-of-tuecs-male . Accessed 9 April 2018 . 2. World Anti-Doping Agency . Medical information to support the decisions of TUECs - transgender athletes . Available at: https://www.wada-ama.org/en/resources/science-medicine/ medicalinformation-to-support-the-decisions-of-tuecs-female-to-male . Accessed 9 April 2018 . 3. Harwood DT , Handelsman DJ . Development and validation of a sensitive liquid chromatography-tandem mass spectrometry assay to simultaneously measure androgens and estrogens in serum without derivatization . Clin Chim Acta . 2009 ; 409 ( 1-2 ): 78 - 84 . 4. Keski-Rahkonen P , Desai R , Jimenez M , Harwood DT , Handelsman DJ . Measurement of estradiol in human serum by LC-MS/MS using a novel estrogen-specific derivatization reagent . Anal Chem . 2015 ; 87 ( 14 ): 7180 - 7186 . 5. Hsu B , Cumming RG , Hirani V , Blyth FM , Naganathan V , Le Couteur DG , Seibel MJ , Waite LM , Handelsman DJ . Temporal trend in androgen status and androgen-sensitive outcomes in older men . J Clin Endocrinol Metab . 2016 ; 101 ( 4 ): 1836 - 1846 . 6. Goebel C , Howe CJ , Ho KK , Nelson A , Kazlauskas R , Trout GJ . Screening for testosterone abuse in male athletes using the measurement of urinary LH, a revision of the paradigm . Drug Test Anal . 2009 ; 1 ( 11 -12): 511 - 517 . 7. World Anti-Doping Agency . Technical document for endogenous anabolic androgenic steroids (TD2016EAAS) . Available at: https:// www.wada-ama.org/en/resources/science-medicine /td2016eaas. Accessed 9 April 2018 . 8. World Anti-Doping Agency . International Standard for Laboratories (ISL) . Available at: https://www.wada-ama.org/en/resources/ laboratories/international -standard-for-laboratories-isl . Accessed 9 April 2018 . 9. Handelsman DJ , Idan A , Grainger J , Goebel C , Turner L , Conway AJ . Detection and effects on serum and urine steroid and LH of repeated GnRH analog (leuprolide) stimulation . J Steroid Biochem Mol Biol . 2014 ; 141 : 113 - 120 . 10. Gehan EA , George SL . Estimation of human body surface area from height and weight . Cancer Chemother Rep . 1970 ; 54 ( 4 ): 225 - 235 . 11. Longcope C , Jaffee W , Griffing G. Production rates of androgens and oestrogens in post-menopausal women . Maturitas . 1981 ; 3 ( 3 -4): 215 - 223 . 12. Irwig MS . Testosterone therapy for transgender men . Lancet Diabetes Endocrinol . 2017 ; 5 ( 4 ): 301 - 311 . 13. Jacobeit JW , Gooren LJ , Schulte HM . Safety aspects of 36 months of administration of long-acting intramuscular testosterone undecanoate for treatment of female-to-male transgender individuals . Eur J Endocrinol . 2009 ; 161 ( 5 ): 795 - 798 . 14. Jakobsson J , Ekstro¨ m L , Inotsume N , Garle M , Lorentzon M , Ohlsson C , Roh HK , Carlstro¨ m K , Rane A . Large differences in testosterone excretion in Korean and Swedish men are strongly associated with a UDP-glucuronosyl transferase 2B17 polymorphism . J Clin Endocrinol Metab . 2006 ; 91 ( 2 ): 687 - 693 . 15. Schulze JJ , Lundmark J , Garle M , Skilving I , Ekstro¨ m L , Rane A . Doping test results dependent on genotype of uridine diphosphoglucuronosyl transferase 2B17, the major enzyme for testosterone glucuronidation . J Clin Endocrinol Metab . 2008 ; 93 ( 7 ): 2500 - 2506 . 16. Xue Y , Sun D , Daly A , Yang F , Zhou X , Zhao M , Huang N , Zerjal T , Lee C , Carter NP , Hurles ME , Tyler-Smith C . Adaptive evolution of UGT2B17 copy-number variation . Am J Hum Genet . 2008 ; 83 ( 3 ): 337 - 346 . 17. Nieschlag E , Bu¨ chter D , Von Eckardstein S , Abshagen K , Simoni M , Behre HM . Repeated intramuscular injections of testosterone undecanoate for substitution therapy in hypogonadal men . Clin Endocrinol (Oxf) . 1999 ; 51 ( 6 ): 757 - 763 . 18. Krakowsky Y , Conners W , Davidson E , Rawji A , Morgentaler A . Initial clinical experience with testosterone undecanoate therapy (AVEED) in men with testosterone deficiency in the United States . Urology . 2017 ; 109 : 27 - 31 .


This is a preview of a remote PDF: https://academic.oup.com/jcem/article-pdf/103/6/2277/25011469/jc.2018-00054.pdf

Savkovic, Sasha, Lim, Sarina, Jayadev, Veena, Conway, Ann, Turner, Leo, Curtis, Douglas, Goebel, Catrin, Handelsman, David J. Urine and Serum Sex Steroid Profile in Testosterone-Treated Transgender and Hypogonadal and Healthy Control Men, The Journal of Clinical Endocrinology & Metabolism, 2018, 2277-2283, DOI: 10.1210/jc.2018-00054