Testosterone is performance enhancing in women but challenging to detect in doping analyses

Abstract

BACKGROUND: There is an ongoing debate whether women with testosterone (T) within the male range shall be allowed to compete in the female class in sports. In men, T has an ergogenic effect, but the effect of T in young women has not been investigated previously. We hypothesised that increased levels of T would enhance physical performance in exercising women. T is classified as a doping agent by the World Anti-Doping agency (WADA). However, the existing doping analysing methods seems to be insufficient to detect exogenous use of T in women. In addition, the menstrual cycle and the use of hormonal contraception might aggravate the interpretation of the doping test results. Accordingly, the percentage of positive doping tests are higher in men than in women. The aim was to study the effect of T in female athletic performance, as well as endocrine influence on the steroid profile of relevance for anti-doping testing in women.

METHODS: The studies in this thesis are based on two double-blind randomised, placebo-controlled trials. Forty-eight young healthy women were randomised to 10 weeks of 10mg daily T-cream or placebo to study the effects of moderately increased T concentration on physical performance, body composition, psychological well-being and self-confidence, and the steroid profile in urine and blood. Physical performance was measured by performance tests at the Swedish School of Sports and Health Sciences. Body composition was measured with DEXA. Well-being and self-confidence were measured by questionnaires. We also studied the effect of combined oral contraceptives (COC) and the menstrual cycle phases and genetic factors on the steroid profile using serum and urine samples from a trial where 348 women were randomised to three months of COC or placebo. Genotyping, immunoassays, LC-MS/MS and GC-MS/MS were performed.

MAIN RESULTS AND CONCLUSIONS: Our findings support a causal effect of T on physical performance as measured by an increase in running time to exhaustion of 8.5% in young healthy women. T also promotes an increase in lean mass and might influence phycological well-being and confidence. The results are of importance for the understanding of the sex difference in athletic performance, as well as for regulations of hyperandrogenism in women’s sports. Standard anti-doping urine analyses detected T-administration in only two out of 24 participants. Individual thresholds increased the detection rate to 40%. Furthermore, it was shown that COC have great impact on the urinary steroid biomarkers included in anti-doping analyses. COC suppressed the serum steroids and phase II metabolites while the ratio of T and androstenedione (A4) remained stable. Moreover, the results confirm that T/A4 remained stable throughout the menstrual cycle. T-administration increased the T/A4-ratio in serum 3.5 times compared to placebo. We can conclude that the urine analyses in use today is insufficient to detect T-doping in women. Instead, we suggest serum testing, including the ratio of T/A4.