Does ethnicity impact CYP2D6 genotype–phenotype relationships?

Abstract

Polymorphism of the CYP2D6 gene leads to substantial interindividual variability in CYP2D6 enzyme activity. Despite improvements in prediction of CYP2D6 activity based on genotype information, large interindividual variability within CYP2D6 genotypes remains and ethnicity could be a contributing factor. The aim of this study was to investigate interethnic differences in CYP2D6 activity using clinical datasets of three CYP2D6 substrates: (i) brexpiprazole (N = 476), (ii) tedatioxetine (N = 500), and (iii) vortioxetine (N = 1073). The CYP2D6 activity of all individuals in the dataset was estimated through population pharmacokinetic analyses as previously reported. Individuals were assigned a CYP2D6 phenotype and CYP2D6 genotype group based on their CYP2D6 genotype and interethnic differences were investigated within each group. Among individuals categorized as CYP2D6 normal metabolizers, African Americans had a lower CYP2D6 activity compared to Asians (p < 0.01) and in the tedatioxetine and vortioxetine analyses also compared to Whites (p < 0.01). Among CYP2D6 intermediate metabolizers, interethnic differences were also observed, but the findings were not consistent across the substrates. Asian carriers of CYP2D6 decreased function alleles tended to exhibit higher CYP2D6 activity compared to Whites and African Americans. The observed interethnic differences within the CYP2D6 phenotype and genotype groups appeared to be driven by differences in CYP2D6 allele frequencies across ethnicities rather than interethnic differences in enzyme activity for individuals carrying identical CYP2D6 genotypes.

Study Highlights.

WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC?

Substantial interindividual variability in CYP2D6 activity exists which is largely explained by genetic polymorphism. It has been hypothesized that other factors, such as ethnicity, may also contribute to the variability.

WHAT QUESTION DID THIS STUDY ADDRESS?

The objective of this study was to investigate potential interactions of ethnicity on CYP2D6 genotype–phenotype relationships.

WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE?

Ethnicity did not appear to affect CYP2D6 genotype–phenotype relationships. Interethnic differences in CYP2D6 activity were found within CYP2D6 phenotype groups, which seemed to be driven by differences in CYP2D6 allele frequencies across ethnicities.

HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE?

Knowledge of specific CYP2D6 allele variants appears to be more important than ethnicity to accurately predict CYP2D6 phenotype.

INTRODUCTION

Cytochrome P450 2D6 (CYP2D6) is involved in the metabolism of numerous commonly prescribed drugs, including antidepressants, antipsychotics, opioids, anti‐arrhythmics, beta‐blockers, and chemotherapeutics. ¹ The CYP2D6 gene encoding the enzyme is highly polymorphic, which causes large interindividual differences in CYP2D6 enzyme activity, which in turn leads to substantial interindividual variability in the exposure of CYP2D6 substrates.

The frequency of CYP2D6 allele variants varies considerably across world populations and consequently the distribution of predicted CYP2D6 phenotypes (i.e., poor metabolizers [PMs], intermediate metabolizers [IMs], normal metabolizers [NMs], and ultrarapid metabolizers [UMs]) differs between ethnic groups, as summarized by Gaedigk et al. ²

When comparing subjects carrying identical CYP2D6 genotypes, large interindividual differences in enzyme activity is often observed, which could suggest that nongenetic sources of variability exist. Reports of interethnic differences in CYP2D6 activity have led leading researchers in the field to warrant further investigation into the contribution of ethnicity on CYP2D6 genotype–phenotype relationships. ³ , ⁴ , ⁵

However, investigations of interethnic differences in CYP2D6 activity within CYP2D6 genotypes are often hampered by varying allele frequencies across ethnicities making clinical studies of adequate sample size difficult to conduct. ⁶ For example, the decreased function allele CYP2D6*10 occurs at a frequency of ~43% in Asian populations, whereas the allele frequency in other populations is only ~2%–7%. ² Similarly, the decreased function allele CYP2D6*17 is very common in individuals of African ancestry (allele frequency ~18%–20%), whereas other ethnic groups only carry the allele at a frequency of 0%–3%. ²

As the majority of pharmacogenetic studies are limited to specific geographic regions and thus only represent one or a few ethnic groups, interethnic comparisons are often underpowered due to inadequate sample sizes. ⁷ Datasets with subjects of different ethnicities carrying identical CYP2D6 genotypes would therefore be highly valuable to further elucidate the interaction between ethnicity and CYP2D6 genotype–phenotype relationships.

In three previously reported studies, we performed population pharmacokinetic (PopPK) meta‐analyses to quantify the CYP2D6‐mediated metabolism of brexpiprazole, tedatioxetine, and vortioxetine. ⁸ , ⁹ , ¹⁰ All three analyses involved large datasets originating from multiple clinical studies performed across several geographic regions (North America, Asia, and Europe). Therefore, these datasets represent an opportunity to study interethnic differences in CYP2D6 activity.

The aim of the current analysis was to investigate interethnic differences in CYP2D6 activity using the datasets from the three studies of brexpiprazole, tedatioxetine, and vortioxetine.

METHODS

The analysis presented in the current report is based on three PopPK analyses of brexpiprazole, tedatioxetine, and vortioxetine, respectively, previously reported. ⁸ , ⁹ , ¹⁰ In each of the three analyses, PopPK modeling was used to describe the joint pharmacokinetics (PK) of parent compounds and their CYP2D6‐dependent metabolites, thereby enabling quantification of the CYP2D6‐mediated metabolism. The CYP2D6 activity was estimated for all individuals in the datasets using individual parameter estimates from the final PopPK models. A brief summary of the three PopPK studies is presented below.

Brexpiprazole

Brexpiprazole is a drug indicated for the treatment of schizophrenia and as adjunctive treatment of major depressive disorder (MDD). ¹¹ CYP3A4 and CYP2D6 are the primary enzymes involved in the metabolism of brexpiprazole. The major metabolite, DM‐3411, is formed via both CYP3A4 and CYP2D6, whereas formation of the minor metabolite, DM‐3412, is almost exclusively mediated via CYP2D6. ¹² , ¹³ Neither DM‐3411 nor DM‐3412 are believed to contribute to the pharmacological effect of brexpiprazole. ¹³

The brexpiprazole dataset included 826 subjects from the United States, Europe, and Asia. The data originated from nine phase I clinical studies including healthy volunteers and four phase II studies, including patients diagnosed with MDD, attention deficit hyperactivity disorder, or schizophrenia. CYP2D6 genotype information was available for 496 subjects. For details on the CYP2D6 genotyping assay, please refer to the original report. ¹⁰

A PopPK model describing the joint PK of brexpiprazole and two of its metabolites, DM‐3411 (major metabolite) and DM‐3412 (minor metabolite), was developed based on more than 8000 plasma samples. The model included central and peripheral compartments for brexpiprazole and DM‐3412 and a central compartment for DM‐3411. As the formation of DM‐3412 (minor metabolite) is almost exclusively mediated via CYP2D6, the metabolic ratio of DM‐3412:brexpiprazole calculated from individual parameter estimates from the final PopPK model was used as an estimate of CYP2D6 activity.

Tedatioxetine

Tedatioxetine is a compound that was in development as an antidepressant drug candidate. Although development of tedatioxetine has now been terminated, its property as a sensitive CYP2D6 substrate makes it excellent for studying CYP2D6 genotype–phenotype relationships. Approximately 80% of tedatioxetine's clearance is mediated by CYP2D6 resulting in the major (inactive) metabolite Lu AA370208 (internal data, H. Lundbeck A/S).

The tedatioxetine dataset included 578 subjects from the United States, Canada, Europe, and Asia. The data originated from six phase I clinical studies, including healthy volunteers, and one phase II study, including patients diagnosed with MDD. CYP2D6 genotype information was available from 534 subjects. For details on the CYP2D6 genotyping assay, please refer to the original report. ⁹

A PopPK model describing the joint PK of tedatioxetine and its CYP2D6‐mediated metabolite, Lu AA37208, was developed based on more than 5400 plasma samples. The model consisted of six compartments, including central and peripheral compartments, for both tedatioxetine and Lu AA37208 and a hypothetical absorption compartment to account for presystemic formation of the metabolite. The extent of presystemic metabolite formation was quantified as a secondary parameter, F_MET, whereas the systemic metabolite formation was quantified as the clearance from the tedatioxetine central compartment to the Lu AA37208 central compartment (CL_CYP2D6). The total CYP2D6 activity for each individual in the dataset was calculated using individual estimates of F_MET and CL_CYP2D6.

Vortioxetine

Vortioxetine is an antidepressant indicated for the treatment of MDD. ¹⁴ Approximately 50% of vortioxetine's clearance is mediated via CYP2D6 resulting in the major (inactive) metabolite Lu AA34443. ¹⁵ , ¹⁶

The vortioxetine dataset included more than 28,000 plasma samples from 1140 subjects participating in 29 clinical studies performed in the United States, Europe, and Asia. All the clinical studies included healthy volunteers and CYP2D6 genotype data were available for 1112 of the subjects. For details on the CYP2D6 genotyping assay, please refer to the original report. ⁸

The joint PK of vortioxetine and its CYP2D6‐dependent metabolite, Lu AA34443, was characterized by a seven‐compartment model, including central and peripheral compartments for both vortioxetine and Lu AA34443. The model included two clearance parameters for vortioxetine: one reflecting the CYP2D6‐mediated metabolism (CL_CYP2D6) and one reflecting the non‐CYP2D6‐mediated metabolism (CL_other). To account for presystemic formation of the metabolite, Lu AA34443, the model included a hypothetical absorption compartment, where the extent of presystemic metabolite formation was quantified by the parameter F_MET.

The total CYP2D6 activity was estimated for each individual in the dataset using individual parameter estimates of the presystemic metabolite formation (F_MET) and the systemic metabolite formation (CL_CYP2D6).

Interethnic differences

CYP2D6 activity estimates from the three analyses were used to investigate interethnic differences. Individuals from the three datasets who were African American, Asian, or White and who had been CYP2D6 genotyped were included in the analysis. Individuals carrying heterozygous genotypes with duplications (e.g., CYP2D6*1/*4XN) were excluded from the analysis as the genotyping assays did not allow identification of which allele was duplicated. In total, the merged dataset included 476 subjects from the brexpiprazole analysis, 500 subjects from the tedatioxetine analysis, and 1073 subjects from the vortioxetine analysis (Table 1).

TABLE 1.

Summary of subject characteristics.

	African American	Asian	White
Categorical characteristics	Frequency	Frequency	Frequency
Compound (brexpiprazole/tedatioxetine/vortioxetine)	104/25/113	54/54/222	318/421/738
CYP2D6 phenotype a PM/IM/NM/UM	8/61/160/13	1/102/222/5	82/524/839/32
Gender, male/female	167/75	209/121	915/562
Study population, healthy subjects/patients b	152/90	281/49	945/532

Continuous characteristics	N	Median	IQR	Range	N	Median	IQR	Range	N	Median	IQR	Range
Age, years	242	35	26–43	7–72	330	25	23–33	18–78	1477	34	26–45	8–80
Body weight, kg	242	79	70–89	30–159	312	60	54–65	42–87	1477	76	66–86	23–152
Height, cm	242	171	165–179	128–191	312	166	160–173	144–191	1477	172	165–178	124–204
BMI, kg/m2	242	27	24–29	17–57	312	21	20–23	17–32	1477	25	23–28	14–49

Abbreviations: BMI, body mass index; IM, intermediate metabolizer; IQR, interquartile range; NM, normal metabolizer; PM, poor metabolizer; UM, ultrarapid metabolizer.

^a CYP2D6 phenotype assigned based on CYP2D6 genotype according to current guidelines. ¹⁷

^b Patients were diagnosed with major depressive disorder, schizophrenia, or attention deficit hyperactivity disorder.

The CYP2D6 activity of subjects carrying identical CYP2D6 genotypes was compared using nonparametric tests. Kruskal‐Wallis tests were used for multiple group comparisons followed by pairwise comparisons using Wilcoxon rank sum tests correcting for multiple testing (Bonferroni‐Holm procedure). To ensure sufficient sample sizes for these comparisons, the null function alleles were pooled and denoted as “*null.”

Furthermore, interethnic differences within CYP2D6 genotype groups were assessed. In this analysis, subjects' CYP2D6 genotypes were grouped according to the functionality of the individual alleles (i.e., full function [CYP2D6*1, CYP2D6*2, denoted “FF”]), decreased function (CYP2D6*9, CYP2D6*10, CYP2D6*14, CYP2D6*17, CYP2D6*29, CYP2D6*41, denoted “DF”), or null function (CYP2D6*3, CYP2D6*4, CYP2D6*5, CYP2D6*6, denoted “NF”).

Last, subjects were assigned a CYP2D6 phenotype (i.e., PM, IM, NM, or UM) based on their CYP2D6 genotype according to current guidelines ¹⁷ and CYP2D6 activity estimates were compared within each phenotype.

RESULTS

Figures 1, 2, 3 illustrate the CYP2D6 activity according to ethnic group for individual CYP2D6 genotypes (Figure 1), CYP2D6 genotype groups (Figure 2), and CYP2D6 phenotypes (Figure 3). Mean and standard deviation (SD) of each group is presented Table S1, which also summarizes the comparisons showing statistically significant differences between the ethnic groups.

FIGURE 1.

Box‐ and scatterplots of estimated CYP2D6 activity according to ethnic group and CYP2D6 genotype. Boxes represent interquartile ranges, lines within boxes indicate median values. Colors indicate ethnic group, blue: African American, red: Asian, and green: White. *Significant interethnic difference, p < 0.05.

FIGURE 2.

Box‐ and scatterplots of estimated CYP2D6 activity according to ethnic group and CYP2D6 genotype grouped by the functionality of the alleles. Boxes represent interquartile ranges, lines within boxes indicate median values. Colors indicate ethnic group, blue: African American, red: Asian, and green: White. DF, decreased function; FF, full function; IM, intermediate metabolizer; NF, null function; NM, normal metabolizer; PM, poor metabolizer; UM, ultrarapid metabolizer. ***Significant interethnic difference, p < 0.05.

FIGURE 3.

Box‐ and scatterplots of estimated CYP2D6 activity according to ethnic group and CYP2D6 phenotype. Boxes represent interquartile ranges, lines within boxes indicate median values. Colors indicate ethnic group, blue: African American, red: Asian, and green: White. IM, intermediate metabolizer; NM, normal metabolizer; PM, poor metabolizer; UM, ultrarapid metabolizer. ***Significant interethnic difference, p < 0.05.

When comparing African Americans, Asians, and Whites carrying identical CYP2D6 genotypes, significant differences were found for CYP2D6*1/*1 (p = 0.02), CYP2D6*1/*null (p = 0.04), and CYP2D6*null/*41 (p = 0.03) carriers in the tedatioxetine analysis and CYP2D6*2/*2 (p = 0.002) and CYP2D6*2/*10 (p = 0.007) carriers in the vortioxetine analysis (see Tables S2, S3, and Figure 1). Among the CYP2D6*2/*2 and CYP2D6*2/*10 carriers in the vortioxetine analysis and CYP2D6*1/*1 carriers in the tedatioxetine analysis, Asians exhibited the highest CYP2D6 activity. For the CYP2D6*1/*null and CYP2D6*null/*41 carriers in the tedatioxetine analysis, Asian subjects had lower CYP2D6 activity compared to Whites.

When assessing interethnic differences within CYP2D6 genotype groups, significant differences were found in the DF/DF (p = 0.0001) and FF/DF (p = 0.0001) groups in the brexpiprazole analysis with Asians exhibiting the highest enzyme activity within the groups (see Figure 2 and Table S4). In the tedatioxetine analysis, Asian FF/NF carriers had a significantly higher enzyme activity compared to White FF/NF carriers (p = 0.04), and in the FF/FF group, African Americans exhibited a significantly lower CYP2D6 activity compared to both Asians and Whites (p = 0.01; see Figure 2 and Table S4).

In the vortioxetine analysis, Asian carriers of NF/DF, DF/DF, FF/NF, and FF/DF had a significantly higher CYP2D6 activity compared to Whites belonging to the same genotype groups. In the FF/NF and FF/DF groups, the difference between Asians and African Americans was also statistically significant with Asians exhibiting the highest enzyme activity (see Table S2 and Figure 2).

For subjects categorized as CYP2D6 NMs, a significant difference was found among the ethnic groups across the three compounds. In the brexpiprazole analysis, Asian CYP2D6 NMs had a significantly higher CYP2D6 activity compared to White NMs (p = 0.0001) and African American NMs (p = 0.0001). In the tedatioxetine analysis, African American NMs had a lower CYP2D6 activity compared to Asian NMs (p = 0.005) and White NMs (p = 0.009). In the vortioxetine analysis, African American NMs had a lower CYP2D6 activity compared to Asian NMs (p = 0.02) and White NMs (p = 0.002).

Among the subjects categorized as CYP2D6 IMs, significant differences among the ethnic groups were found in the brexpiprazole and tedatioxetine analysis. In the brexpiprazole analysis, African American IMs had a significantly lower CYP2D6 activity compared to Whites (p = 0.003). In the tedatioxetine analysis, Asian IMs had a significantly lower CYP2D6 activity compared to White IMs (p = 0.0004).

DISCUSSION

Using extensive data material from three CYP2D6 genotype–phenotype analyses, statistically significant interethnic differences were only found for very few genotypes and the findings were not consistent across the analyses. In the tedatioxetine analysis, Asian CYP2D6*1/*1 carriers had a significantly higher CYP2D6 activity compared to both White and African American CYP2D6*1/*1 carriers. Furthermore, Asian carriers of CYP2D6*2/*2 and CYP2D6*2/*10 had significantly higher CYP2D6 activity compared to White carriers of the same genotypes in the vortioxetine analysis. It should be acknowledged that comparisons at the genotype level were challenged by a low number of each ethnicity‐genotype combination.

Significant interethnic differences in CYP2D6 activity of CYP2D6*2 carriers have previously been reported. ⁵ A study by Gaedigk et al. ⁵ showed pronounced differences in CYP2D6 enzyme activity (measured by the dextromethorphan/dextrorphan urinary metabolic ratio) between White and African American CYP2D6*2 carriers: African American subjects carrying the genotypes CYP2D6*1/*2 and *2/*2 had a significantly lower CYP2D6 activity compared with White subjects carrying the same genotypes, whereas no difference was found among CYP2D6*1/*null and *2/*null carriers. In our analysis, we did not find evidence of differences between Whites and African Americans carrying CYP2D6*2, but the observation of higher enzyme activity among Asian CYP2D6*2 carriers in the vortioxetine analysis is noteworthy.

When dividing the subjects into CYP2D6 genotype groups, a slight tendency was found of Asians exhibiting a higher CYP2D6 activity within some genotype groups, particularly those involving decreased function alleles. In the three original CYP2D6 genotype–phenotype analyses of brexpiprazole, tedatioxetine, and vortioxetine, considerable differences were found in the enzyme activity associated with different CYP2D6 decreased function alleles. These analyses showed consistently lower CYP2D6 activity for the CYP2D6*17 and *41 alleles compared to CYP2D6*10. ⁸ , ⁹ , ¹⁰ Therefore, when assessing genotype groups involving decreased function alleles, the interethnic differences observed might be a consequence of allele‐specific differences in CYP2D6 activity, as allele frequencies vary between ethnic groups. Specifically, the CYP2D6*10 allele is most common among Asian populations which provides a possible reason for the higher CYP2D6 activity observed for the Asian subjects belonging to the NF/DF, DF/DF, and FF/DF genotype groups.

In the comparisons within CYP2D6 phenotypes, we found interethnic differences in CYP2D6 activity within subjects classified as CYP2D6 NMs and IMs, respectively. Among CYP2D6 NMs, African Americans tended to have the lowest CYP2D6 activity among the three ethnic groups, whereas Asian NMs generally had the highest CYP2D6 activity. Within the IM phenotype, Whites tended to have a higher CYP2D6 activity, although findings were not consistent across the three analyses.

Previous studies have shown that CYP2D6 NMs and IMs exhibit distinct CYP2D6 activity depending on the functionality of the alleles comprising the genotypes. ⁸ , ¹⁸ As frequencies of null, decreased, and fully functional alleles vary across ethnic groups, ² this may explain some of the differences observed within the phenotype categories.

Nonetheless, as prescribers are often only informed of patients' CYP2D6 phenotype (and not the underlying genotype), the observed interethnic differences within the phenotype groups could be clinically relevant and may need to be taken into consideration when individualizing treatment based on CYP2D6 phenotype.

LIMITATIONS

There are several limitations to the current study that need to be addressed. First of all, it is possible that the observed differences among CYP2D6*1/*1 carriers in the tedatioxetine analysis could have been affected by variant alleles not being detected in the CYP2D6 genotyping assay. In all the genotyping assays used across the clinical studies, the CYP2D6*1 allele was assigned by default when no allele variants were detected. It is thus possible that the individuals assigned CYP2D6*1/*1 harbored other allele variants not covered by the assay in the given study.

It is also possible that the interethnic differences observed for genotypes involving CYP2D6*2 could be caused by subvariants or polymorphisms in enhancer regions. For example, evidence suggests that the CYP2D6*2A allele is associated with increased levels of CYP2D6 expression. ¹⁹ Furthermore, an enhancer single‐nucleotide polymorphism (SNP; rs5758550) in linkage disequilibrium with one of the SNPs harbored by CYP2D6*2 (rs16947) may enhance transcription. ²⁰ Unfortunately, the genotyping assays used in the three studies did not test for the enhancer SNP or CYP2D6*2A, and it was therefore not possible to test these hypotheses on our data.

Several CYP enzymes contribute to the metabolism of vortioxetine including CYP2D6, CYP2C19, and CYP2C9. ²¹ Although the PopPK modeling approach aimed to isolate the CYP2D6 pathway, it is possible that variability in other enzymes may affect the estimated CYP2D6 activity. The two most common allele variants of CYP2C9 are CYP2C9*2 and *3, which cause reduced enzyme activity. Both of these variants occur at a lower frequency in Asians compared to Whites, ²² which may contribute to the greater (apparent) CYP2D6 activity estimated among Asians.

Several nongenetic factors may contribute to the interethnic differences in CYP2D6 activity observed in this study. First, subject characteristics, such as age, body weight, height, body mass index (BMI), and study population differed among the three ethnic groups. Noteworthy, Asian subjects had a significantly lower age, body weight, height, and BMI compared with White and African American subjects (p < 0.0001). These differences could impact the PK of the three compounds, for example, by affecting the volume of distribution or drug clearance. However, the effect of age, height, body weight, and BMI were all tested as covariates in the original PopPK analyses and included in the final models if found to impact the PK significantly. The PopPK models therefore accounted for these differences and the confounding potential is therefore considered limited.

Furthermore, the Asian cohort included more healthy subjects relative to patients compared to the other ethnic groups. Previous studies have indicated that different mood and cognitive disorders potentially could impact CYP2D6 gene expression through changes in the hypothalamic–pituitary–adrenal axis. ²³ , ²⁴ , ²⁵ The different proportion of healthy subjects versus patients could therefore potentially have confounded the results. However, in the original PopPK analyses, disease state (healthy subjects vs. patients) was not identified as a significant covariate affecting the PK of any of the compounds.

The CYP2D6 activity estimates were based on data originating from multiple clinical studies, and the majority of Asian subjects participated in studies exclusively performed in Asia. There is a possibility that systematic differences in the clinical studies conducted in Asia may have affected the PopPK parameter estimates and thereby the estimated CYP2D6 activity. However, in the covariate analyses of the three original PopPK studies, neither study nor region was identified as a significant covariate, which makes this hypothesis unlikely.

In conclusion, our study suggests that interethnic differences in enzyme activity within CYP2D6 phenotype groups exist, whereas we found limited evidence of interethnic differences within individual CYP2D6 genotypes. The differences observed within the CYP2D6 NMs and IMs appeared to be driven by interethnic differences in the prevalence of individual allele variants rather than differences in enzyme activity for subjects carrying identical CYP2D6 genotypes.

AUTHOR CONTRIBUTIONS

T.F., J.A., E.S., T.B.S., and K.B. wrote the manuscript. T.F., J.A., and E.S. designed the research. T.F and J.A. performed the research. T.F., J.A., T.B.S., and K.B. analyzed the data.

FUNDING INFORMATION

The research was funded by Innovation Fund Denmark (grant number: 8053‐00083B) and H. Lundbeck A/S.

CONFLICT OF INTEREST STATEMENT

T.F., J.A., and E.S. are employed by H. Lundbeck A/S. All other authors declared no competing interests for this work.

Supporting information

Table S1

Frederiksen T, Areberg J, Schmidt E, Stage TB, Brøsen K. Does ethnicity impact CYP2D6 genotype–phenotype relationships? Clin Transl Sci. 2023;16:1012‐1020. doi: 10.1111/cts.13506