Abstract

The determination of the exogenous origin of urinary metabolites through the measurement by GC-C-IRMS of their carbon isotopic signature (δ13C) is widely used to confirm the administration of steroids related to testosterone and prohibited in sports. The rigorous analytical conditions required to perform accurate measurements have often dictated the selection of the final most intense metabolites, androsterone (A) and etiocholanolone (Etio), potentially limiting its efficacy. In fact, their alteration is less marked and persistent than for testosterone (T) itself and 5β-androstane-3α,17β-diol (5α-Adiol) or 5α-androstane-3α,17β-diol (5α-Adiol), which on the other end are present in much lower amount and hence, more difficult to test reliably. We have developed a one-step HPLC purification of seven diagnostic urinary metabolites (target steroids: TS, hydrolyzed glucuronides): T, DHEA, 5β-Adiol, 5α-Adiol, epitestosterone (E), A, Etio and two endogenous reference compounds (ERC), pregnanediol (pgdiol) and 5α-androst-16-en-3β-ol (16-enol). These steroids are pooled in three fractions and analyzed without derivatization. With regards to the GC-C-IRMS analysis, a multi-level isotopic calibration using the ‘identical treatment’ principle was found to provide results for purified reference steroids with a precision ≤ 0.17 and accuracy of ≤ 0.30 ‰(between assay, n=26). Compared to other common endogenous reference compounds, those selected in this study have δ13C values close to the testosterone metabolites which, along with the proposed isotopic calibration, produced reference intervals within ± 3 ‰ for most diagnostic TS-ERC pairs, in compliance with the requirements of the World Anti-Doping Agency. Verf.-Referat