To develop a pharmacokinetic model able to describe the nonlinear pharmacokinetics of paroxetine (PRX) and to predict the drug-drug interaction between PRX and metoprolol under various dosage regimens.A pharmacokinetic model of PRX incorporating mechanism-based inhibition was developed. This model was fitted to the drug concentration profiles obtained after single and repeated administrations of PRX to estimate the pharmacokinetic parameters of PRX and degradation rate constant of cytochrome P450 (CYP) 2D6. It was also fitted to the time profile of S-metoprolol after coadministration of metoprolol and PRX, and the fractional contribution of CYP2D6 to overall clearance of S-metoprolol was estimated. Using the developed model and estimated parameters, an optimal dosage regimen for metoprolol during withdrawal of PRX was simulated.The developed model well described the time profiles of both PRX and metoprolol concentration during concomitant administration. The estimated parameters were consistent with reported values. The nonlinear and accumulation properties of PRX could be explained by mechanism-based inhibition of CYP2D6 by PRX. Upon tapering PRX from 20 mg/ day to 10 mg/day for 14 days then 5 mg/day for 14 days until cessation, the optimal dosage regimen to resume 120 mg/day of metoprolol based on the developed model was as follows: 30 mg/day during concomitant administration, 40 mg/day for the next 14 days, 60 mg/day for the next 14 days, and finally 120 mg/day.The developed model enabled us to quantitatively estimate drug-drug interactions of PRX and CYP2D6 substrate drugs, and to predict optimal dosage regimens.