Testosterone esters in serum: application to an oral testosterone undecanoate administration study and implementation of an inter-laboratory comparison

Code: 21A07RN

The detection of testosterone misuse in elite sports is still one of the mayor challenges for anti-doping laboratories. Since the human body also produces testosterone and its metabolites, the presence of abnormal concentrations of endogenous androgens in urine sample(s) of an athlete is not sufficient to declare an Adverse Analytical Finding (AAF). Their concentrations and ratios are monitored longitudinally in the urinary steroidal module of the athlete biological passport (ABP). If a sample is outside of the individual reference limits for the athlete, it is considered as atypical, and the origin of testosterone has to be confirmed. Currently, the methodology to routinely distinguish between endogenous and exogenous testosterone is the gas chromatography–combustion-isotopic ratio mass spectrometry (GC/C/IRMS) analysis. This type of investigation is accompanied by labour-intensive sample preparation that is prone to deficiencies. The results can sometimes be inconclusive, most often due to limited sensitivity, which is the main problem associated with GC-C-IRMS method, and which requires also a significant volume of urine. In this context, in a recent study carried out at LAD, the use of blood matrix was proposed as a complementary tool for improving the urinary steroidal module of the ABP. Through the quantification of testosterone (T), androstenedione (AND) and 5-dihydrotestosterone (DHT) in serum samples, the development and consideration of a future longitudinal “blood steroidal profile (BSP)” was hypothesized. This approach already showed a particularly good detection windows for patch and oral T administrations [1] . Also WADA has recently established a working group to further investigate this BSP approach and the possibility of setting up future regulations for the concept. In addition to the BSP approach, the serum matrix will also be very useful to detect the testosterone esters. In fact, oral and/or intramuscular applications are usually administered not in the form of free testosterone, but its esters. This is to prevent a fast diurnal excretion and provide the body with a prolonged supply of free testosterone through the esters hydrolysis in blood. Those esters cannot be synthesised by the human body, therefore their presence in the blood matrix is a direct sign of illicit testosterone intake and an unambiguous proof of doping. In the case of testosterone ester positive findings, this would even negate the need for an IRMS analysis. With this in mind, and in compliance with the prevailing WADA TD2018EAAS, different gas or liquid chromatography-mass spectrometry methods have been developed by few anti-doping laboratories in the last years (see below). The combination of results obtained with the analysis of endogenous steroids and testosterone esters in serum could provide additional and complementary information to the urinary steroid profile and IRMS results. With this approach, we would also emphasize the usefulness of simultaneous collection of urine and blood samples within a sample collection session [2] .

The project is divided into four different sub-projects, which focus each on different aspects of the detection of testosterone via the analysis of the esters. The first part focuses on the method development and validation of an initial testing procedure (ITP, “screening”) for the majority of ester derivatives (e.g. propionate, cypionate, decanoate, undecanoate) at LAD, in the second part this method will be utilised to analyse serum samples from a previous clinical study performed at LAD which included an oral TU administration to 19 healthy men volunteers. In the third part of the project, GC/C/IRMS analyses will be performed on 50 urine samples of the same clinical study and results, in terms of detection window, will be compared to that of testosterone esters. Finally, a comparison between the different methods for the testosterone ester analysis will be performed through an inter-laboratory study, which is arranged and managed by the LAD.

Main findings

An LC-MS/MS method for the quantification of three endogenous steroids (T, A4 and DHT) as well as qualitative analysis of a selected menu of exogenous steroids and steroid esters with a single aliquot of human serum was developed and validated. With this approach it was possible to combine the direct detection of synthetic steroids and steroid esters with the indirect detection of steroid doping in the context of the Athlete Biological Passport (ABP). The method was then applied to the analysis of approximately 200 serum samples coming from a previously described testosterone undecanoate (TU) oral administration study. Results demonstrated that elevated TU serum concentrations had a profound impact on the endogenous steroids, allowing distinction between the control phase (no TU administration) and samples after oral TU administration. The two markers, which were found to be most influenced by the oral TU administration, are TU itself and DHT, which was elevated even at low TU serum concentrations.

In the second part of this study, urine and serum samples from the same administration study were analysed to compare the TU detection window with the urinary GC/C/IRMS analysis used in the routine environment. Intact TU was detected in serum samples for at least 4 h up to 24 h after the administration. While urinary IRMS analysis led to slightly longer overall detection windows (especially for 5α/5β-diols) from 12 to 48 hours, this analysis was associated with increased efforts needed in comparison with the serum analysis.

Finally, an inter-laboratory study on the detection of testosterone esters in serum samples was performed showing good results in terms of detection capabilities by the different participating laboratories. All substances spiked in the serum samples were detected correctly by different ITP and estimations of concentrations were judged satisfactory for qualitative methods in the low ng/mL range of concentrations.