P. Ghahramani, W.W. Yeo, G.T. Tucker
Dept. Medicine & Pharmacology, Floor L, Royal Hallamshire Hospital, Sheffield Univ., UK.
Introduction: About 8% of Caucasians have genetic variants of CYP2D6 that express inactive enzyme and are classed as “poor metabolisers”. Amitriptyline (AT) is N-demethylated to nortriptyline (NT), an active metabolite with similar clinical effects. It has been shown in vitro that CYP2D6 contributes to this reaction. We have examined its role in vivo in the pharmacokinetics of AT and NT.
Methods: Twelve extensive metabolisers and six poor metabolisers were given oral doses of AT (50 mg) and NT (50 mg) on separate occasions. Plasma samples were taken over 24 hours and assayed for both compounds. Kinetic analysis was performed using NONMEM (version 2.1). One- and two- compartment models were fitted to the data assuming first-order absorption with various error models. Relationships between individual clearances (CL) and CYP2D6 phenotype (assessed from the urinary debrisoquine drug/metabolite ratio (DMR)), weight and age were tested graphically and by linear regression. If there was a relationship, the factor was incorporated into the model as a covariate. A decrease in the objective function value (OFV) of 3.84 (i.e. p= 0.05) was considered as a significant improvement of the model.
Results: The NONMEM analysis showed that kinetics were best described by a two-compartment model for both AT (p<0.0005) and NT (p=0.05). There were no significant relationships between the CL of AT and the covariates but there was a trend for DMR. There was a significant relationship (r-0.48, p<0.05) between the CL of NT and DMR. There was also a trend relating the CL of NT to both weight and age. For AT, only DMR and for NT DMR, weight and age were sequentially added to the structural model in different forms (e.g. additive, multiplicative, etc.). When DMR was added, no significant improvement in OFV was observed for AT whereas there was a significant improvement for NT (p-0.05). For NT, further improvement was obtained by adding weight (p=0.05) as a second covariate but this was not the case for age. For NT produced from AT, a four-compartment model (Fig. 1) was fitted to the NT concentrations and kinetic parameters for AT (i.e. Ka, Kl2, K21) were fixed to the valuss obtained for the AT run. There was no significant relationship between Kl3 and DMR. Similar results were obtained from conventional data analysis.
Conclusions: We have not detected a significant effect of CYP2D6 on the plasma concentrations or the N-demethylation of AT, but the plasma concentrations of NT after an oral dose of NT were signiScantly influenced by CYP2D6 phenotype. Extensive first-pass metabolism by non-CYP2D6 enzymes in the gut or the liver may be involved in amitriptyline N-demethylation. This minimises the role of CYP2D6 mediated N-demethylation after absorption. With conventional methods of kinetic analysis, the parameters are calculated for each phenotype and then the means are compared. Using NONMEM, CYP2D6 activity (i.e. DMR) can be incorporated as a covariate in the pharmacokinetic model allowing use of the whole data set. This is particularly important as studies involving poor metabolisers often have a small sample size.
Reference: PAGE 7 () Abstr 691 [www.page-meeting.org/?abstract=691]
Poster: poster