DISCUSSION
In this retrospective study, we found a significant effect ofCYP2D6 genotype on vortioxetine exposure measured in routine clinical practice. The systemic exposure of vortioxetine observed in the CYP2D6 PM group was 3-fold higher than that observed in the NM group, while the exposure in the IM group was increased 1.5-fold compared to the NM group. A population pharmacokinetic study of vortioxetine has previously shown a significant effect of CYP2D6 phenotype on oral clearance of vortioxetine with average estimated values being 53 L/h for UMs, 34 L/h for NMs, 27 L/h for IMs and 18 L/h for PM [7]. This is reflected in the vortioxetine drug label, where it is recommended that CYP2D6 PMs should be treated with a maximum dose of 10 mg vortioxetine per day [8]. The present study confirms this recommendation based on data from a naturalistic setting, which is important from a clinical point of view. Although the vortioxetine exposure for CYP2D6 UMs is expected to be reduced compared to NMs, clinical studies have shown a significant overlap in exposures between UMs and NMs and therefore dose adjustment for UMs is not recommended [3]. This is in line with the results from the current study where no significant difference in vortioxetine exposure was found between CYP2D6 UMs and NMs. However, it should be noted that this finding was based on a limited number of UM patients (N=8).
The current study showed that CYP2D6 PMs and UMs, as compared with NMs, had an increased frequency of switching to another antidepressant within the expected time frame of a depressive episode (three months). As PMs exhibited significantly higher exposures compared to the other CYP2D6 phenotype groups, the increased switch rate in PMs is likely to be driven by a higher frequency of adverse events caused by supratherapeutic drug concentrations. By decreasing the dose among the CYP2D6 PMs, patients could have achieved lower vortioxetine concentrations, which may have increased the probability of staying within the therapeutic window and reduced their risk of concentration-dependent adverse drug reactions. Although no significant difference in vortioxetine exposure was found between CYP2D6 UMs and NMs, the median concentration observed among the UMs was lower than 10 ng/mL, which has been reported as the lower limit for efficacy of vortioxetine [9]. As the UM patients generally had exposure levels close to the lower limit of the therapeutic window, the increased frequency of antidepressant switch may be related to insufficient clinical response. However, these findings are based on a very limited number of patients and larger studies would be needed to adequately address this hypothesis.
Overall, the findings from this study are in line with previous reports on the effect of CYP2D6 and CYP2C19 genotypes on risperidone and escitalopram treatment outcomes, respectively [5,6]. In a study of 725 patients treated with risperidone, CYP2D6 PMs and IMs were found to have a significant increase in serum levels of risperidone active moiety compared to NMs (P<0.0001). Furthermore, the incidence of treatment switch from risperidone to another antipsychotic was significantly increased in PMs (P=0.015) and UMs (P=0.003) compared to NMs[6]. Similarly, a study of more than 2,000 escitalopram-treated patients showed a 3.3-fold increase in escitalopram exposure among CYP2C19 PMs and a 10% reduction in exposure in CYP2C19 UMs compared to NMs. Paralleled by the differences in exposure, the CYP2C19 PMs and UMs showed significantly higher frequencies of switching from escitalopram to another antidepressant compared to CYP2C19 NMs (P<0.001) [5].
The main limitations of the current study were the limited number of UM patients and lack of clinical information retrieved from the patients’ medical records due to privacy issues. Although switch of antidepressant treatment within three months may indicate therapeutic failure, exact information on treatment outcomes (including the reasons for treatment switch) was not obtainable from the TDM database. Furthermore, lack of knowledge on covariates that may affect drug exposure, such as comedications, renal function and body size represent additional limitations. However, the use of TDM data also represents several advantages, e.g. exact information on drug use and replacement (switch), as the analytical method allows for the detection of serum concentration levels of multiple antidepressants simultaneously. Furthermore, the availability of serum concentration levels enables identification and exclusion of non-compliant patients, where the drug levels are undetectable.
In conclusion, this study confirms the significant effect ofCYP2D6 genotype on vortioxetine exposure and provides novel data on the association between CYP2D6 genotype and therapeutic failure in a naturalistic, clinical setting. Together with previous studies, these results underline the importance of variability in CYP metabolism on treatment outcomes of psychiatric medications and support the value of routine TDM and CYP genotyping in personalised medicine in psychiatry.