Characterization of CYP2D6 Pharmacogenetic Variation in Sub-Saharan African Populations

Abstract

Cytochrome P450 2D6 (CYP2D6) is a key enzyme in drug response owing to its involvement in the metabolism of ~ 25% of clinically prescribed medications. The encoding CYP2D6 gene is highly polymorphic, and many pharmacogenetics studies have been performed worldwide to investigate the distribution of CYP2D6 star alleles (haplotypes); however, African populations have been relatively understudied to date. In this study, the distributions of CYP2D6 star alleles and predicted drug metabolizer phenotypes-derived from activity scores-were examined across multiple sub-Saharan African populations based on bioinformatics analysis of 961 high-depth whole genome sequences. This was followed by characterization of novel star alleles and suballeles in a subset of the participants via targeted high-fidelity Single-Molecule Real-Time resequencing (Pacific Biosciences). This study revealed varying frequencies of known CYP2D6 alleles and predicted phenotypes across different African ethnolinguistic groups. Twenty-seven novel CYP2D6 star alleles were predicted computationally and two of them were further validated. This study highlights the importance of studying variation in key pharmacogenes such as CYP2D6 in the African context to better understand population-specific allele frequencies. This will aid in the development of better genotyping panels and star allele detection approaches with a view toward supporting effective implementation of precision medicine strategies in Africa and across the African diaspora.

Figure 1

African geographical regions represented in the study. The size of the circles (not drawn to scale) reflect the relative number of participants from each population group. AWI‐Gen, Africa Wits‐INDEPTH partnership for Genomics studies; CBRL, Cell Biology Research Laboratory (National Institute of Communicable Diseases, Wits University, Johannesburg); DRC, Democratic Republic of the Congo; H3Africa, Human Heredity and Health in Africa Consortium; SAHGP, Southern African Human Genome Programme.

Figure 2

Frequency distribution of CYP2D6 star alleles grouped by function across African and global populations. SSA populations had a higher frequency of decreased‐function CYP2D6 star alleles compared with European, admixed American, and South Asian populations in part due to a higher frequency of CYP2D6*17 and *29. Conversely, CYP2D6*10 accounts for the high proportion of decreased‐function alleles in East Asian populations. SSA and South Asian populations had a relatively higher proportion of alleles with unknown or uncertain function which warrants more functional assays to determine the CPIC function of these alleles. The sample sizes were as follows: sub‐Saharan African = 961 participants; African American/Afro‐Caribbean = 157 participants; European = 503 participants; Admixed American = 347 participants; South Asian = 489 participants; East Asian = 504 participants. CPIC, Clinical Pharmacogenetics Implementation Consortium.

Figure 3

Distribution of CYP2D6 phenotypes predicted from diplotypes called from high‐coverage WGS data. IM, intermediate metabolizer; n, sample size; NM, normal metabolizer; PM, poor metabolizer; UM, ultrarapid metabolizer; WGS, whole genome sequence.

Figure 4

CYP2D6 phenotype distribution across the SSA populations in the study. BFA, participants from Burkina Faso; BOT, participants from Botswana; BRN, Berom in Nigeria; BSZ, Bantu‐speakers in Zambia; CAM, Cameroonian participants; ESN, Esan in Nigeria; FNB, Fon in Benin; GHA, Ghanaian participants; GWD, Gambian (Mandinka) in Western Divisions  of the Gambia; LWK, Luhya in Webuye (Kenya); MSL, Mende in Sierra Leone; SA, South African participants (south eastern Bantu‐speakers); SSA, sub‐Saharan African; YRI, Yoruba in Ibadan, Nigeria.

Figure 5

Novel CYP2D6 star alleles in SSA characterized via XL‐PCR and HiFi sequencing. Panel (a) shows the two haplotypes in a South African south eastern Bantu‐speaking participant with the novel CYP2D6*70+rs16947 (R296C) haplotype. The submission of this allele for consideration by PharmVar is ongoing. The variant phasing and frequency information from our analysis suggests that CYP2D6*70, which has a “moderate” PharmVar level of evidence, should be redefined to include rs16947 but the final decision on this rests with the PharmVar CYP2D6 expert panel. Panel (b) shows the two alleles identified in a South African participant with the novel CYP2D6*41+rs141824015 (I339L) allele. Two suballeles of this haplotype have been predicted in the Luhya in Webuye, Kenya (LWK) from the 1000 Genomes Project data set and they are being characterized in a collaborative project with 3 labs led by PharmVar members. SNVs (single‐nucleotide variations) indicated in black text are from the “backbone” haplotype. HiFi, high fidelity; PharmVar, Pharmacogene Variation Consortium; SSA, sub‐Saharan Africa; XL‐PCR, long‐range polymerase chain reaction.