Urinary steroid profiles in doping testing : in relation to natural variation and drug administration

Abstract

To detect doping with endogenous steroids, such as testosterone, biomarkers in urine are measured. These biomarkers include testosterone and some metabolically related steroids. The measured steroids are combined into ratios and together they make up the steroid profile. This steroid profile is followed over time in the Steroid Module of the Athlete Biological Passport. The software used for the passport, calculates individual reference ranges based on the previous results and gives atypical findings if one or more biomarker goes outside of the reference ranges.

All passports are evaluated by experts and all atypical findings are assessed. Evaluating steroidal passports is however difficult since factors, other than doping, can affect the biomarkers of the steroid profile. In this thesis, we evaluated natural variations of the steroid profile, including variations during the menstrual cycle and pregnancy, as well as how certain drugs, such as hormonal contraceptives and testosterone, affect the steroid profile.

In a study of over 11 000 steroid profiles we have seen that intra-individual variation in the steroid profile is large (16-27% in men and 23-40% in women), but that inter-individual variation is larger (49-76% in men and 55-84% in women). Some of this variation could be explained by annual and diurnal variation, with time of day having a larger impact on makers of the steroid profile. Another confounder to consider when evaluating passports is if the urine was collected in competition or not, a factor that could explain over 6% of the total interindividual variation of some ratios. We have also seen that the menstrual cycle affect biomarkers of the steroid profile and that hormonal contraceptives can give patterns on the steroid passports similar to micro-doping with T. Pregnant women also show great differences in their steroid profiles as compared to non-pregnant women. We have seen that doping with as low as 125 mg T enanthate and 100 mg T gel can be detected with the ABP. However, it is possible that the large natural variation as well as confounding factors, such as permitted drugs, will conceal the effect of doping.

The goal of studying confounding factors in steroid profiling is to provide the scientists evaluating the passport with sharper tools, not only to select the profiles suspicious of doping, but also to be able to reject and not spend unnecessary time and resources on profiles showing atypical results due to natural causes. The ultimate goal is to be able to proceed with a passport case, where the steroidal passport is the only evidence of doping.