Detection of epitestosterone doping by isotope ratio mass spectrometry.

Background: Epitestosterone is prohibited by sport authorities because its administration will lower the urinary testosterone/epitestosterone ratio, a marker of testosterone administration. A definitive method for detecting epitestosterone administration is needed.

Methods: We developed a gas chromatography-combustion-isotope ratio mass spectrometry method for measuring the delta(13)C values for urinary epitestosterone. Sample preparation included deconjugation with beta-glucuronidase, solid-phase extraction, and semipreparative HPLC. Epitestosterone concentrations were determined by gas chromatography-mass spectrometry for urines obtained from a control group of 456 healthy males. Epitestosterone delta(13)C values were determined for 43 control urines with epitestosterone concentrations > or =40 microg/L (139 nmol/L) and 10 athletes' urines with epitestosterone concentrations > or =180 microg/L (624 nmol/L), respectively.

Results: The log epitestosterone concentration distribution was gaussian [mean, 3.30; SD, 0.706; geometric mean, 27.0 microg/L (93.6 nmol/L)]. The delta(13)C values for four synthetic epitestosterones were low (less than or equal to -30.3 per thousand) and differed significantly (P <0.0001). The SDs of between-assay precision studies were low (< or =0.73 per thousand). The mean delta(13)C values for urine samples obtained from 43 healthy males was -23.8 per thousand (SD, 0.93 per thousand). Nine of 10 athletes' urine samples with epitestosterone concentrations >180 microg/L (624 nmol/L) had delta(13)C values within +/- 3 SD of the control group. The delta(13)C value of epitestosterone in one sample was -32.6 per thousand (z-score, 9.4), suggesting that epitestosterone was administered. In addition, the likelihood of simultaneous testosterone administration was supported by low delta(13)C values for androsterone and etiocholanolone.

Conclusions: Determining delta(13)C values for urinary epitestosterone is useful for detecting cases of epitestosterone administration because the mean delta(13)C values for a control group is high (-23.8 per thousand) compared with the delta(13)C values for synthetic epitestosterones.