Methemoglobin (MHb) formation can be a clinically significant and dose-limiting side effect of 8-aminoquinoline antimalarials. MHb may also protect against cyanide poisoning. A two-compartment pharmacokinetic model, linked to a sigmoid Emax pharmacodynamic model, was developed to predict the MHb levels after administration of 8-[(4-amino-1-methylbutyl)amino]-2,6-dimethoxy-4-methyl-5-[(3- trifluoromethyl)phenoxy] quinoline succinate (WR 238605 succinate), a primaquine analogue. Six healthy male beagle dogs received four daily doses of 6.0 mg/kg (base) orally. Forty plasma drug concentrations and 19 MHb levels (effect) were determined over 7 weeks on each dog. Compartmental and noncompartmental pharmacokinetic and parametric and nonparametric pharmacodynamic analyses were performed. Model parameters (mean +/- SD) included a Vss/f of 18.5 +/- 2.8 L/kg, CL/f of 83 +/- 24 ml/hr/kg, terminal elimination t1/2 of 169.7 +/- 52.0 hr, t1/2keo of 123.0 +/- 22.4 hr, an Emax of 31.3 +/- 15.9% MHb, an EC50 of 596 +/- 128 ng/ml, and a sigmoidicity coefficient (n) of 1.94 +/- 0.47. The model was then validated in three additional dogs given three different dosing regimens. It predicted the peak plasma concentrations and MHb levels and the times of their occurrence well. This model could be useful for dose and sampling time selection in further animal studies and initial human phase I clinical testing.