Trine Frederiksen (1,2), Prof. Kim Brøsen (1), Tore Bjerregaard Stage (1), Ellen Schmidt (2), Johan Areberg (2)
(1) University of Southern Denmark (2) H. Lundbeck A/S
Introduction: The CYP2D6 enzyme is involved in the metabolism of numerous drugs, but the polymorphic nature of the CYP2D6 gene causes substantial interindividual differences in enzymatic activity. Therefore, CYP2D6 genotype tests can be a useful tool to guide pharmacotherapy involving CYP2D6 substrates. However, there is a lack of consensus on how to translate CYP2D6 genotypes into phenotypes and consequently CYP2D6-based dose recommendations can be inconsistent. Tedatioxetine is metabolized to its major metabolite, Lu AA37208, almost exclusively (80%) via CYP2D6, thereby making tedatioxetine a good compound to study CYP2D6 activity.
Objectives:
- to test the ability of different popPK modelling approaches to quantify CYP2D6 activity (i.e. phenotype) based on tedatioxetine data
- to investigate CYP2D6 genotype-phenotype relationships based on individual estimates from the popPK models
Methods: Plasma concentration data of tedatioxetine and its metabolite, Lu AA37208, was available from six densely sampled phase I studies (8.939 samples from 220 subjects) and one sparsely sampled phase II study (1.883 samples from 358 MDD patients). Two different modelling approaches were used to quantify the CYP2D6-mediated metabolism of tedatioxetine to Lu AA37208:
(i) metabolic ratios based on separate popPK models of tedatioxetine and Lu AA37208, and
(ii) a joint popPK model of tedatioxetine and Lu AA37208 estimating the CYP2D6-mediated metabolism.
The subjects carried 52 different CYP2D6 genotypes that are normally translated into four phenotypic groups: poor, intermediate, normal and ultra-rapid metabolizers (PM, IM, NM, UM). A multiple linear regression analysis was performed on each measure of CYP2D6 activity to investigate CYP2D6 genotype-phenotype relationships. The CYP2D6 activity associated with different CYP2D6 genotypes was compared using non-parametric tests.
Results: The PK of the two compounds individually was best described by two compartment models with lag-time absorption. A metabolic ratio (MR) was calculated for each individual using normalized (covariate adjusted) clearance estimates from the two final models.
The joint popPK model comprised central and peripheral compartments for both tedatioxetine and Lu AA37208 plus a hypothetical absorption compartment for Lu AA37208 to account for pre-systemic metabolism. The extent of pre-systemic metabolite formation was estimated by a bioavailability parameter, Fmet, while the systemic metabolite formation was estimated as the clearance from the tedatioxetine central compartment to the Lu AA37208 central compartment (CLCYP2D6). The total CYP2D6-mediated metabolism was estimated as the product between Fmet and CLCYP2D6.
Both modelling approaches provided estimates of CYP2D6 activity that correlated well with the expected CYP2D6 phenotype groups (UM/NM/IM/PM). The MR performed slightly better than the estimates from the joint popPK model. For the CYP2D6 activity estimates based on the densely sampled studies, the CYP2D6 predicted phenotypes explained 68% and 63% of the variability, respectively. For the sparsely sampled studies, 48% and 38% of the variability was explained by the CYP2D6 phenotypes.
The results showed that carriers of one fully functional CYP2D6 allele in combination with a null function CYP2D6 allele had a significantly higher level of CYP2D6 activity compared to carriers of two decreased function CYP2D6 alleles (p<0.0001). This finding stands in contrast to current recommendations where both CYP2D6 allele combinations are translated to the same phenotype. Furthermore, the CYP2D6*10 allele was associated with a significantly higher CYP2D6 activity compared to CYP2D6*41 although both alleles are normally assigned the same functional level.
Conclusions: Both popPK modelling approaches successfully described CYP2D6 activity levels across a range of different CYP2D6 genotypes. The findings from this study indicate that current recommendations for CYP2D6 genotype-phenotype translation could be optimized, particularly with regards to decreased function alleles and the intermediate metabolizer phenotype. Similar analyses on other CYP2D6 substrates could preferably be performed to facilitate discussions on how to optimally translate CYP2D6 genotypes to phenotypes for improved personalized medicine.
Reference: PAGE () Abstr 9468 [www.page-meeting.org/?abstract=9468]
Poster: Clinical Applications