Method development for analysis of testosterone/epitestosterone in human samples using liquid chromatography tandem mass spectrometry

Title from first page of PDF file (viewed January 6, 2011)Includes bibliographical references (p. 66-69)Testosterone (T) has an epimeric chemical form called epitestosterone (E), which differs not in chemical composition but in configuration of one carbon stereocenter with an alcohol group. Urinal testosterone and epitestosterone naturally exist in a ratio of approximately 1:1. The world Anti-Doping Code Prohibited List specifies that a borderline case exists if the T/E ratio is [greater than or equal to] 4 or if the testosterone level is higher than 200 ng/mL. If above the borderline, samples are subjected to a carbon isotope ratio test for a determination of sources of testosterone since synthetic have a different carbon isotope ratio than those produced within the human body Most of the testosterone and epitestosterone in urine are conjugated with glucuronide and only a very small portion is in a free form. Steroid levels determined for doping analysis are glucuronide equivalent concentration in urine. We have developed a rapid electrospray ionization method for LC-MS/MS to quantify both free and conjugated forms of testosterone and epitestosterone using isotope dilution mass spectrometry and tested it for sensitivity, accuracy, precision, linearity, recovery and repeatability. With a methanol/water mobile phase on a C8 reverse phase column, a resolution of 2.84 was achieved between the epimers. Both T and E had a LOD of 0.01 ng/mL and LOQ of 0.05 ng/mL. The calibration curves were linear from 0.05 ng/mL to 50 ng/mL with R?? of 0.999. To determine the concentration of conjugated forms, urine samples were hydrolyzed with [lower case beta]-glucuronidase and extracted with SPE columns, followed by LC-MS/MS analysis. Free testosterone concentrations were determined by the same procedure but without the hydrolysis step. Accuracy was determined by two spiked aliquots (0.5 ng/mL and 5 ng/mL) of the same urine sample, and they were recovered to within with 6% and 3% of the spiked amounts, respectively. The precision was evaluated with five aliquots of one female urine sample hydrolyzed and extracted separately, and %RSDs of 3.8% (T), 20% (E), and 16% (T/E ratio) were obtained. The T/E ratios of urine from eight non-doping subjects were determined, and the values were found to be in the range of 0.1 to 2.3, all below the cutoff value for dopant screening. The poor consistency of internal standard peak areas suggests that ion suppression by matrix compounds is an underlying reason for errors in measurement, especially the epitestosterone levels. Full scan MS spectra of samples shows that co-eluting compounds contain 500 times greater signal intensity than those of epitestosterone, suggesting that ion suppression of epitestosterone is very likely. The primary interfering component could not be identified, but it produced an ion at m/z 229. The signals from testosterone and testosterone d??? were not as suppressed by co-eluting compounds. Sample hydrolysis produced significantly stronger interference peaks and suppression of the internal standard peaks by tenfold in some samples. A study of the interferences and means to minimize their effects are required to produce reliable T/E ratios by LC-MS/MS