The woodchuck (Marmota monax) has proven to be a suitable animal model for studying hepatitis B virus (HBV) infection owing to similarities in the course of infection between woodchuck hepatitis virus (WHV) in woodchucks and HBV in humans. (-)-beta-L-2',3'-Dideoxy-3'-thiacytidine (3TC; lamivudine) is a nucleoside analog which has demonstrated antiviral activity against HBV as well as human immunodeficiency virus (HIV). The purpose of the present investigation was to characterize the pharmacokinetics of 3TC following intravenous and oral administration of 20 mg of 3TC per kg of body weight to woodchucks. Following intravenous administration, the concentrations of 3TC in plasma declined, with a terminal half-life of 2.84 +/- 0.85 h (mean +/- standard deviation). The systemic clearance and steady-state volume of distribution of 3TC were 0.22 +/- 0.078 liters/h/kg and 0.75 +/- 0.13 liters/kg, respectively. The renal clearance of the nucleoside analog was 0.063 +/- 0.016 liters/h/kg. The oral bioavailability of 3TC ranged from 18 to 54%. Allometric relationships between pharmacokinetic parameters and body weight developed by Hussey et al. (E.K. Hussey, K.H. Donn, M.J. Daniel, S.T. Hall, A.J. Harker, and G.L. Evans, J. Clin. Pharmacol. 34:975-977, 1994) were augmented by including data from woodchucks, monkeys (S.M. Blaney, M.J. Daniel, A.J. Harker, K. Godwin, and F.M. Balis, Antimicrob. Agents Chemother. 39:2779-2782, 1995), and additional data from rats (P. Rajagopalan, L. Moore, C.K. Chu, R.F. Schinazi, and F.D. Boudinot, submitted for publication). Interspecies scaling of the pharmacokinetic parameters of 3TC demonstrated a good correlation between clearance (0.74 . W0.76 [where W is body weight]; r = 0.93; P < 0.025), apparent volume of distribution (1.62 . W0.81; r = 0.98; P < 0.005), and steady-state volume of distribution (1.09 . W0.94; r = 0.99; P < 0.05) and species body weight. The allometric relationships for clearance and volume of distribution at steady state predicted the observed pharmacokinetic parameters in humans quite well; however, the apparent volume of distribution was underestimated in humans. Thus, the pharmacokinetic data obtained with the woodchuck HBV animal model should be useful for designing clinical trials.