Abdul Basit , University of Washington, Seattle, WA
Hay Young Zhang , University of Washington, Seattle, WA
Deepak Kumar Bhatt , University of Washington, Seattle, WA
John K Amory , University of Washington, Seattle, WA
Bhagwat Prasad , University of Washington, Seattle, WA

Background and objectives: Testosterone (T) undergoes extensive and highly variable first-past metabolism with an oral bioavailability of 2-8% (1). While the major fraction of the oral dose is excreted as a glucuronide metabolite (2), its complete metabolic fate and subsequent effect on the endogenous steroid metabolomics is not well characterized. The aim of this study was to elucidate the effect of oral testosterone on the plasma steroid metabolome using a fully validated LC-MS/MS method.

Methods: Seven healthy male volunteers with mean age 24.2 ± 8.7 years were recruited. T emulsion in sesame oil, mixed with large volume of milk, was administered orally to human subjects with experimental hypogonadism as described previously (3). The blood samples were collected and the plasma was isolated and stored in -80 ºC until LC-MS/MS analysis. Steroids and their metabolites were extracted from plasma using protein precipitation followed by solid-phase extraction. Steroids were analyzed by AB SCIEX 6500 MS/MS. Other than T, following metabolites were measured: testosterone glucuronide (TG), androstenedione (AED), androsterone (A), androsterone glucuronide (AG) and dihydrotestosterone (DHT) (Fig 1a). Pharmacokinetics data were analyzed using the non-compartment method (Phoenix, Certara, Princeton, NJ). A UGT2B17 genotype-guided in vitro analysis of metabolic clearance of T and DHT was also performed in the human liver microsomes.



The mean area under the curve (AUC) with SD for T after oral dosing was 178 ± 74.95 nmol.h/L, with a huge interindividual variability ranging from 35.57 to 246.83 nmol.h/L. The mean AUC of conjugated steroids (TG and AG) was 80-and 376- fold higher than T AUC. A modest increase was also observed in AUC of all unconjugated T metabolites, i.e., AED, A and DHT. However, 6-β OH-T, which is formed in vitro by CYP3A4, was below limit of quantification (0.1 ng/ml). Significant interindividual variability in Cmax and Tmax values was observed for T (16.75 ± 11.08 nmol/L and 4 hr) and for TG (1937.26 ± 1257.42 nmol/L, 1 hr) (Fig. 1a). Complementary in vitro data confirmed UGT2B17 as a major contributor to T-glucuronidation.


Our results confirmed the importance of glucuronidation in the first-pass metabolism of T. In addition to TG formation, we found that T to AG conversion (Fig. 1b) is an important mechanism of the low oral bioavailability of T. These data are important for the development of oral testosterone therapy as well as for doping control regulations. Testosterone PK and steroid metabolomics data analyses with 200 and 400 mg oral dose (with and without dutasteride, a dual inhibitor of 5α reductase) are ongoing and will be presented.


1)    Tauber, U. Schroder, K. Dusterberg, B. Matthes, H. Absolute bioavailability of testosterone after oral administration of testosterone-undecanoate and testosterone. Eur J Drug Metab Pharmacokinet. 1986 Apr-Jun;11(2):145-9.

2)    Korenman, S. G. Lipsett, M. B. Is Testosterone Glucuronoside Uniquely Derived from Plasma Testosterone? J Clin Invest. 1964. Vol. 43, No. 11, 2125-31

3)    Amory, J. K. Bremner, W. J. Oral testosterone in oil plus dutasteride in men: a pharmacokinetic study. J Clin Endocrinol Metab, May 2005, 90(5):2610–2617.