REVIEW

PharmVar GeneFocus: CYP2C19

Mariana R. Botton1, Michelle Whirl-Carrillo2, Andria L. Del Tredici3, Katrin Sangkuhl2,
Larisa H. Cavallari4, José A. G. Agúndez5, Jorge Duconge6, Ming Ta Michael Lee7, Erica L. Woodahl8,
Karla Claudio-Campos4, Ann K. Daly9, Teri E. Klein2, Victoria M. Pratt10, Stuart A. Scott11,12,†
and Andrea Gaedigk13,*,†

The Pharmacogene Variation Consortium (PharmVar) catalogues star (*) allele nomenclature for the polymorphic human
CYP2C19 gene. CYP2C19 genetic variation impacts the metabolism of many drugs and has been associated with both
efficacy and safety issues for several commonly prescribed medications. This GeneFocus provides a comprehensive
overview and summary of CYP2C19 and describes how haplotype information catalogued by PharmVar is utilized by the
Pharmacogenomics Knowledgebase and the Clinical Pharmacogenetics Implementation Consortium (CPIC).

CYP2C19 BRIEF HISTORY Cytochrome P450 Allele Nomenclature Database,12 which was
A novel drug hydroxylation polymorphism in CYP2C19 was first widely accepted by the pharmacogenetics community as the cen-
described in 1984 after deficient metabolism of the anticonvulsant tral resource for cataloguing CYP variation. Thirty-five CYP2C19
drug mephenytoin was identified in healthy volunteers and in fam- haplotypes, CYP2C19*1 through *35, were submitted by investiga-
ily studies.1 Mephenytoin was subsequently used as a probe drug to tors to this database before it was transitioned to the Pharmacogene
distinguish between normal and poor metabolizers, which ultimately Variation (PharmVar) Consortium in 2017.13 Although only exons
was determined to be due to variable expression of cytochrome were required to be sequenced for submission to the original CYP
P450 2C19 (CYP2C19).2,3 The CYP2C19 gene was first character- nomenclature site, upstream gene regions were eventually also in-
ized in 1991,4 and is one of four CYP2C subfamily genes, located terrogated, submitted, and included in haplotype definitions as
on chromosome 10q23.33 (CYP2C8, CYP2C9, CYP2C18, and investigators searched for variants in regulatory elements. Several
CYP2C19).5 The first CYP2C19 variant allele, implicated in poor submissions also contained intronic variants located near exon-intron
metabolism (CYP2C19*2), was reported in 1994.6 Numerous vari- boundaries, which were included in haplotype definitions, regardless
ants were subsequently reported and new allelic variants continue of whether they were experimentally shown to impact CYP2C19
to be identified.7,8 Although many CYP2C19 variants have been function. So-called “suballeles” (e.g., CYP2C19*2A-J), were initially
reported as having no function, a notable discovery in CYP2C19 also catalogued, but eventually no longer considered for independent
pharmacogenetics was the identification of the CYP2C19*17 allele, naming.
which is associated with increased enzyme activity and more rapid The information provided by the CYP nomenclature site was
drug clearance.9 A systematic nomenclature system for CYP2C19 utilized by the pharmacogenomics research communty, knowledge
alleles was first proposed in 200010 and continues to be updated resources (e.g., Pharmacogenomics Knowledgebase (PharmGKB))
as new alleles are described. Another two milestones in CYP2C19 as well as the pharmacogenetics testing and implementation com-
history are the US Food and Drug Administration (FDA) revision munities, including clinical genetic testing laboratories and the
of drug labeling (e.g., clopidogrel with boxed warning added) based CPIC.14 As the genomics field progressed with remarkable ad-
on CYP2C19 pharmacogenetics and the evidence-based Clinical vances, more human genetic variants were discovered, including
Pharmacogenetics Implementation Consortium (CPIC) guidelines pharmacogenetic variants, which ultimately resulted in challenges
issued for various CYP2C19 gene-drug pairs. with reconciling CYP variants into the star (*) allele system. This
also identified a need for a more standardized allele submission sys-
STATUS OF NOMENCLATURE BEFORE PHARMVAR tem and rigorous nomenclature protocol, as well as a more open-ac-
Star (*) nomenclature was first applied to CYP2D6 in the 1990s,11 cess haplotype definition database.
which then expanded to include CYP2C19 and other CYP genes.10 In this review, Human Genome Variation Society guide-
CYP star (*) allele nomenclature was maintained by the Human lines (Table 1) are used to report star (*) allele sequence variants

1
Hospital de Clínicas de Porto Alegre, Porto Alegre, RS, Brazil; 2Department of Biomedical Data Science, Stanford University, Stanford, California,
USA; 3Millennium Health LLC, San Diego, California, USA; 4University of Florida, Gainsville, Florida, USA; 5UNEx, ARADyAL, Instituto de Salud
Carlos III, University Institute of Molecular Pathology Biomarkers, Cáceres, Spain; 6School of Pharmacy, University of Puerto Rico, San Juan, Puerto
Rico; 7Genomic Medicine Institute, Geisinger, Danville, Pennsylvania, USA; 8Department of Biomedical and Pharmaceutical Sciences, University
of Montana, Missoula, Montana, USA; 9Newcastle University, Newcastle upon Tyne, UK; 10Indiana University School of Medicine, Indianapolis,
Indiana, USA; 11Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, New York, New York, USA; 12Sema4,
Stamford, Connecticut, USA; 13Division of Clinical Pharmacology, Toxicology & Therapeutic Innovation, Children’s Mercy, Kansas City, Missouri, USA.
*Correspondence: Andrea Gaedigk ([email protected])
†
These authors should be considered co-senior authors.
Received April 23, 2020; accepted June 15, 2020. doi:10.1002/cpt.1973

352 CLINICAL PHARMACOLOGY & THERAPEUTICS | VOLUME 109 NUMBER 2 | February 2021
 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

Table 1 Online CYP2C19 resources - links to sites and online in Drug Labeling,16 the FDA Table of Pharmacogenetic
resources referenced throughout the review Associations,17 and the CPIC drug-gene pairs.18 Links to re-
Sources and references References sources mentioned here and throughout this GeneFocus are
PharmVar
summarized in Table 1.
83,93,94 One of the most common clinical applications of CYP2C19
CYP2C19 important gene information
• Read Me Document genotyping is to predict response to clopidogrel and guide se-
• Change Log Document lection of antiplatelet therapy after percutaneous coronary in-
• Structural Variation Document tervention among patients with acute coronary syndromes.19
Standards 82
Data on outcomes with this genotype-guided approach to anti-
Allele Designation and Evidence Level 78 platelet therapy are beginning to emerge.20,21 Moreover, studies
Document in multi-ethnic, more diverse populations—including some in
CYP2C19 Gene Expert Panel Roster 79
minority, under-represented groups—are also underway.22 It is
P450 Nomenclature Site – Archive 95 expected that results from these studies will contribute signifi-
PharmGKB cantly to closing the current knowledge gap in the field regard-
CYP2C19 gene page 43 ing the identification of novel ethno-specific variants of clinical
44 value, as well as validating previous findings from early studies in
Gene-Specific Information Tables for CYP2C19
• Allele Definition Table diverse populations at large.
• Allele Functionality Table Other clinical uses of CYP2C19 genetic testing are to guide
• Frequency Table prescribing of selective serotonin reuptake inhibitors in the
• Diplotype-Phenotype Table treatment of depression, dosing of proton pump inhibitors for
• Gene Resource Mappings
15
gastroesophageal reflux disease, and dosing of voriconazole for
• CYP2C19 Drug Label Annotations
prophylaxis of invasive fungal infection among others. A recent
CPIC study of CYP2C19-guided voriconazole dosing for antifungal
96
Guidelines prophylaxis in adult allogeneic hematopoietic cell transplant
Gene/Drug Pairs 18
recipients demonstrated a decrease in the percent of patients
• Process for assigning CPIC levels with subtherapeutic concentrations and improved treatment
• Levels for gene/drug pairs
• Process for prioritizing CPIC guidelines
success rates with the genotype-guided approach compared with
historical controls, with an estimated cost savings of $4,700 per
Other resources
30
patient.23
Drug Interactions Flockhart Table
Given the clinical significance of CYP2C19 genotype information,
16
FDA Pharmacogenomic Biomarkers in Drug it is imperative that test information is accurate and that providers
Labeling
17
understand the limitations of a genotype test. Recommendations for
FDA table of pharmacogenetic associations
clinical CYP2C19 genotyping allele selection have been published24
80
NCBI reference sequences database and test options are discussed in detail elsewhere.25,26
81
Locus Reference Genomic Project
97
10X Genomics (Linked-Reads Genomics) OTHER FACTORS THAT CAN INFLUENCE CYP2C19 ACTIVITY
HGVS Nomenclature 98 A patient’s metabolic profile may be profoundly impacted by one or
multiple comedications. The magnitude of drug-drug interaction
depends on the individual metabolizer phenotype and the relative
according to their relative position in the CYP2C19 NM_000769.1 contribution of CYP2C19 to the metabolism of a drug(s). For ex-
transcript, with the ATG start codon being 1; corresponding pro- ample, fluconazole, fluoxetine, and fluvoxamine are known inhibi-
tein coordinates are also provided. For example, the CYP2C19*3 tors of CYP2C19, and may phenoconvert individuals with a normal
allele defining variant (rs4986893) is referred to as c.636G>A CYP2C19 genotype to intermediate or poor metabolizers, whereas
(p.W212X), indicating the variant affects the nucleotide at cDNA rifampin, efavirenz, and St. John’s wort are known inducers.27,28 Of
position 636 and causes a tryptophan to stop codon change at note, drug interactions are of no consequence for CYP2C19 poor
amino acid position 212. metabolizers who lack enzyme activity. Drug-drug interactions are
of particular concern in the elderly as polypharmacy is common in
CLINICAL SIGNIFICANCE this population.29 For more information on CYP2C19 inhibitors
The CYP2C19 enzyme is part of the CYP450 superfamily and inducers we refer to the Drug Interactions Flockhart Table.30
contributing to the metabolism of many clinically used drugs, It has also been suggested that inflammation may influence ef-
including clopidogrel, voriconazole, proton pump inhibi- ficacy. As shown in recent studies, chronic inflammation was as-
tors, antidepressants, carisoprodol, and diazepam. Moreover, sociated with significantly reduced CYP2C19 activity by 46% in
CYP2C19 also metabolizes endogenous substances, such as mel- patients with type 2 diabetes.31 In the OPTIMUS study, activity
atonin and progesterone. 8 More information on drugs metabo- was reported to be reduced by 60% and patients with type 2 di-
lized by CYP2C19 can be found in the PharmGKB drug label abetes remained poor responders although maintenance doses of
annotations,15 the FDA table on Pharmacogenomic Biomarkers clopidogrel were doubled.32

CLINICAL PHARMACOLOGY & THERAPEUTICS | VOLUME 109 NUMBER 2 | February 2021 353
 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

SEX AND AGE-RELATED DIFFERENCES Annotation Tool is currently under development by PharmGKB
Although gender-specific CYP2C19 gene expression has been to facilitate the interpretation and reporting of pharmacogenom-
observed in mice (higher in men than women in the liver and ic-based dosing recommendations.45
kidneys), this remains controversial in human studies.33 A num-
ber of investigations found no differences between genders,34,35 CYP2C19 AND THE CLINICAL PHARMACOGENETICS
however, these were contradicted by additional investigations IMPLEMENTATION CONSORTIUM
that reported higher CYP2C19 activity in women36 than men.37 The CPIC develops structured, evidence-based, clinical prac-
Age-related differences in CYP2C19 gene and protein expres- tice guidelines for drugs affected by PGx variation.46,47 Several
sion, as well as metabolic activity, have been reported. CYP2C19 CYP2C19 gene-drug pairs have been prioritized through consid-
gene and protein expression were found to be the lowest in fetal eration of multiple factors, such as the available body of literature,
and the highest in pediatric samples.38,39 CYP2C19 protein ex- severity of clinical consequences, availability of alternative ther-
pression appears to linearly increase during the first 5 months of apies, and whether a prescribing change (drug choice or dose) is
life but is variable from this age to 10 year of age (21-fold).40 It was warranted. To date, four CPIC guidelines have been published that
also shown that CYP2C19 mRNA levels were lower in adult com- include CYP2C19: tricyclic antidepressants,48 selective serotonin
pared with pediatric samples suggesting that CYP2C19 expression reuptake inhibitors,49 clopidogrel,50 and voriconazole,51 and a
may peak during childhood.41 CYP2C19 guideline on proton pump inhibitors (e.g., esomeprazole
and omeprazole) is forthcoming. Each guideline has multiple com-
CYP2C19 AND THE PHARMGKB ponents, including CYP2C19 phenotype-specific therapeutic rec-
The PharmGKB collects, curates, and disseminates knowledge ommendations, systematic evidence reviews, and implementation
about the impact of human genetic variation on drug response.42 resources to support the translation of the guideline into electronic
The CYP2C19 gene page allows structured access to gene-specific health records (EHRs) with example clinical decision support text.
pharmacogenomic knowledge.43 Information is presented in sec- To facilitate clinical implementation of PGx information and
tions, including prescribing information, drug label annotations, portability of PGx data across institutions, CPIC established
clinical annotations, variant annotations, and curated pathways. standard terminology for allele functional status and phenotypes
Prescribing information encompasses, (1) annotations of clini- for pharmacogenes, including CYP2C19. CPIC specifically rec-
cal guidelines from sources such as the CPIC, the Royal Dutch ommends the terms “no function,” “decreased function,” “normal
Association for the Advancement of Pharmacy - Pharmacogenetics function,” and “increased function” to describe alleles, and “poor,
Working Group (DPWG), and the Canadian Pharmacogenomic intermediate, normal, rapid, and ultrarapid” to describe inferred
Network for Drug Safety (CPNDS), and (2) “Rx study annotations" drug metabolizing enzyme phenotypes.52
that provide genotype-based drug dosing, reported in individual
journal articles. Nine CPIC, 19 DPWG clinical guideline annota- GENOTYPE TO PHENOTYPE TRANSLATION
tions (10 with recommendations and 9 providing no drug-specific An individual has two CYP2C19 haplotypes, one on each chro-
recommendations), and 12 “Rx study annotations” are available as mosome, which constitutes his/her diplotype. For example, a
of early 2020 for CYP2C19 with overlapping CYP2C19-drug pairs. CYP2C19*2/*3 diplotype assignment indicates that one chromo-
PharmGKB extracts pharmacogenomic-relevant information some (or allele) has single nucleotide variations (SNVs) defining the
from agency-approved drug labels and applies a “pharmacogenetics CYP2C19*2 haplotype and the second chromosome (or allele) has
(PGx) level” category, based on the level of action implied in each SNVs defining the CYP2C19*3 haplotype. The term “genotype” can
label (e.g., Testing required, Actionable PGx, Testing recommended, refer either to the sum of all detected SNVs or to a person’s diplotype.
or Informative PGx). On the CYP2C19 page, annotations can be For functional classification, individuals are categorized into
accessed (as of early 2020) for 23 FDA approved labels, 9 European the five CPIC-recommended phenotype categories: poor (PM),
Medicines Agency (EMA) approved labels, 6 Pharmaceuticals intermediate (IM), normal (NM; formerly EM, extensive), rapid
and Medical Devices Agency, Japan (PMDA) approved labels, 9 (RM), and ultrarapid (UM) metabolizers.52 Unlike the CYP2D6
Health Canada (HCSC) approved labels, and 12 Swiss Agency gene, which utilizes the Activity Score system to facilitate genotype
of Therapeutic Products (Swissmedic) approved labels. Currently, to phenotype translation,53 the CYP2C19 gene translation process
PharmGKB contains 82 CYP2C19-related clinical annotations, is based on the combination of increased, normal, decreased, and
which are evidence-rated genotype/diplotype level summaries for no function alleles, as provided in the Diplotype-Phenotype-Table
specific variant/allele-drug combinations based on curated single by the PharmGKB and CPIC.44 The table provides all possible
literature entries (variant annotations). Pharmacokinetic pathways CYP2C19 diplotypes and their respective phenotypes. An indi-
depicting CYP2C19 in drug metabolism are available for 30 drugs, vidual’s predicted phenotype is based on the function of the two
although the significance for CYP2C19 involvement varies by drug. alleles present in an individual.
PharmGKB and CPIC work collaboratively to develop gene-spe-
cific resources that accompany each CPIC guideline, including NEED FOR STANDARDIZED GENETIC VARIATION DEFINITIONS
allele definition mapping, allele functionality, allele frequency, AND REPORTING OF FUNCTIONAL/CLINICAL IMPACT
and diplotype to phenotype mapping files in a standardized for- In order to guide drug therapy, it is imperative to understand
mat. Gene-specific information tables for CYP2C19 are available CYP2C19 allelic variation as well as allele and genotype function.
from PharmGKB.44 In addition, the Pharmacogenomics Clinical Although many alleles have been observed in phenotypic PMs and

354 VOLUME 109 NUMBER 2 | February 2021 | www.cpt-journal.com

 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

their causative genetic variations described (e.g., CYP2C19*2, *3, includes CYP2C19*2, *3, and *17, and tier 2 (extended panel of
*4, etc.), the function of an allele may be unknown (currently none variant alleles) that includes CYP2C19*4.001 and *4.002 (previ-
for CYP2C19) or uncertain. In vitro characterization of allelic ously defined as *4A and *4B), *5, *6, *7, *8, *9, *10, and *35. The
variants39,54–58 often produce results that are inconsistent among utilization of star allele nomenclature as provided by PharmVar14
test systems, which may be attributed to differences in the sub- will not only ensure that each star allele represents a unique and
strates used and between the experimental variables (see CYP2C19 fully defined haplotype but also minimize “mis-interpretation” of a
functionality table for a detailed summary44). Although in silico genotype result and its clinical implication(s).
prediction tools are improving, in vivo validation is still the gold The two end-user groups benefiting the most from standard-
standard. Therefore, for any given allele (except for those shown ized allele designations are clinicians and patients. Standardized
to completely abolish activity) caution should be taken when ex- terms and language will help clinicians to convey and explain re-
trapolating functional data from one drug or substrate to another. sults and patients to understand what the results mean for them.
Ideally, one would be able to assess the in vivo function of each Consistent nomenclature is a must for the integration of PGx into
individual CYP2C19 haplotype with each individual CYP2C19 EHRs, as well as for the establishment of clinical decision support
substrate, which would refine the phenotype predicting capacity algorithms and the design of clinical support tools, such as inter-
of CYP2C19 genetic testing. In addition, comedications (drug- ruptive alerts.63,64 For example, drug/allele combinations for alerts
drug interactions) may not affect all CYP2C19 variants equally, require systematic annotations within the EHR, using standard-
and there is still limited or no information regarding genetic vari- ized nomenclature and terms. Finally, the analysis of PGx clinical
ability for many minority populations.59,60 correlations will benefit from harmonized nomenclature of the
For many drug metabolizing enzymes, the combination of se- gene variants, as well as support consistent test interpretation by
quence variations that define haplotypes is critical to precisely providers across healthcare systems.
predict enzyme function. A notable example for CYP2C19 is c.-
806C>T, which is the defining core SNV of the CYP2C19*17 THE CYP2C19 GENE LOCUS
increased function allele but can also be found with c.1A>G on The CYP2C19 gene is composed of 9 exons, which encode a
the no function CYP2C19*4.002 suballele. Given that both ge- protein of 490 amino acids. The gene is located on chromosome
notyping and short-read sequencing are unable to determine the 10q23.33 within a cluster of genes comprising the CYP2C gene
phase of variants like c.-806C>T and c.1A>G, precisely inferring family.8 The CYP2C gene locus contains four genes, CYP2C8,
CYP2C19 haplotypes remains a challenge. CYP2C9, CYP2C18, and CYP2C19 (Figure 1) spanning a total
As previously reported in the CYP2D6 GeneFocus,61 clinical of 486 kb. CYP2C19 is the longest of the four genes with 93.9 kb
PGx programs have successfully been implemented over the past of full-gene sequence. Due to high similarities among these genes,
years, but numerous challenges remain to accelerate adoption.62 genotyping assay design requires particular attention to ensure
Standardization is a key area that continues to represent an opportu- specific amplification from the targeted gene to avoid false-pos-
nity for all PGx stakeholders to improve upon, including laboratory itive test results. CYP2C19 genotyping assays are commercially
processes, test ordering, result reporting, and data representation. available and commonly used for testing. However, this (and
This is in alignment with recent reports, which emphasized that clin- similar methods) require(s) the amplification of relatively short
ically actionable PGx information must be accurately represented polymerase chain reaction (PCR) fragments, which can make
in EHRs by using a harmonized system for genotype and pheno- CYP2C19-specific primer design difficult for regions with high
type information.63,64 Although many PGx laboratories utilize star similarity with the other CYP2C genes, especially CYP2C9.
nomenclature as recommended by PharmVar and CPIC, interlab- Genetic variation, contributing to interindividual variability
oratory differences in testing approaches and reporting remain. in CYP2C19 activity, is caused by SNVs within coding regions.
Clinical PGx testing for CYP2C19 can be performed on a variety Gene copy number variations (CNVs) may also affect activity in
of platforms using different methodologies and genotyping data can rare cases. Numerous deletion and duplication events, affecting the
be reported in different ways, such as chromosomal or genomic po- CYP2C gene locus, have been described.69,70 Many of the gene de-
sition on a reference sequence (RefSeq), as an amino acid change, letion and duplication events involve CYP2C19, as well as other
database SNP rsID, and/or using star (*) allele nomenclature.12,65,66 CYP2C gene(s) and, in some instances also other genes within the
A recent study performed by the Genetic Testing Reference chromosomal region. PGx genotyping tests typically do not inter-
Material Program concluded that many PGx variants interrogated rogate CYP2C19 CNVs; these are more commonly interrogated
were not consistent across platforms.67 Similar findings were re- by copy number arrays and other quantitative molecular assays
ported by Moyer et al., when they surveyed laboratories offering (e.g., quantitative PCR multiplex ligation-dependent probe am-
PGx services for CYP2D6 and CYP2C19 genotyping.68 To ad- plification).70 PharmVar has recently catalogued the CYP2C19*36
dress this, the Association for Molecular Pathology (AMP) and and *37 alleles, which represent full and partial (with including at
College of American Pathologists (CAP) have recently published least exon 1) CYP2C19 gene deletions, respectively.
recommendations for alleles to be included in clinical CYP2C19
testing.24 In brief, the recommendations are based on allele func- CYP2C19 ALLELE, GENOTYPE, AND PHENOTYPE
tion, allele frequencies across populations and ethnicities, and the FREQUENCIES ACROSS POPULATIONS
availability of reference materials. The AMP working group rec- The CYP2C19 frequency table available at PharmGKB,44 sum-
ommended tier 1 alleles (minimum panel of variant alleles), which marizes population-based allele frequencies reported in the

CLINICAL PHARMACOLOGY & THERAPEUTICS | VOLUME 109 NUMBER 2 | February 2021 355
 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

(a)

(b)

Figure 1 Overview of the gene locus and allelic variation. (a) Provides a graphical overview of the CYP2C gene locus containing CYP2C18,
CYP2C19, CYP2C9, and CYP2C8. The latter is encoded on the reverse strand (arrow), whereas the other three genes are encoded on the forward
strand. CYP2C19 is composed of nine exons, covering 92.9 kb. Rare whole and partial gene deletion events that include at least exon 1 of the
CYP2C19 gene have been designated by Pharmacogene Variation Consortium as CYP2C19*36 and *37, respectively. Whole gene deletions may
include the neighboring CYP2C gene(s) and even extend beyond the CYP2C gene locus. Partial deletions affect different regions of CYP2C19 and
may involve CYP2C18 and/or 2C9 as well. (b) Summarizes the core sequence variations defining CYP2C19*2 through*10 and *17.

literature. Studies were considered for inclusion if (1) the ethnicity Regardless, there is considerable variation among the estimated
of the population was clearly indicated, (2) either allele frequen- frequencies for individual alleles across biogeographical groups.
cies or genotype frequencies were reported, (3) the methodology The no function CYP2C19*2 allele has been found at high fre-
by which the genes were genotyped was indicated, (4) the sample quencies in Oceanian (60%) and South and East Asian (28%)
population consisted of at least 50 individuals, and (5) the study populations. This allele is less frequent, albeit still common, in Sub-
represented an original publication. The ethnicities/locations Saharan African populations and their descendants (16–18%), and
reported in the articles are mapped into seven geographically Europeans (15%). CYP2C19*3, also a no function allele, is more
defined groups (American, Central/South Asian, East Asian, frequent in Oceanians (15%), compared with East Asians (7%) and
European, Near Eastern, Oceanian, and Sub-Saharan African) Near Eastern populations (2%), and is rarely found in other pop-
and two admixed groups (African American/Afro-Caribbean ulations around the world (e.g., 0.3% in African-American, Afro-
and Latino), using the biogeographical grouping system developed Caribbean, and Sub-Saharan Africans, and 0.2% in Europeans).
by PharmGKB.71 The CYP2C19 frequency table is periodically Other alleles, including CYP2C19*9 (decreased function), *15
updated and contains multiple tabs summarizing “allele frequen- (normal function), and *35 (no function), are more prevalent in
cies by biogeographical group,” “diplotype frequencies by biogeo- African populations. In contrast, the CYP2C19*17 increased
graphical group,” “phenotype frequency,” and “references”; the function allele is frequently observed in Europeans (22%), Near
latter describes allele frequencies for each publication included Eastern populations (19%), and Sub-Saharan Africans (17%), but
in the listing, which also allows the user to customize allele fre- rarely in East Asians (2%). Consequently, the highest frequency
quencies as needed. There are, however, limitations regarding the of CYP2C19 PMs are found in Oceanic and East Asian popula-
accuracy of allele frequencies as follows: (1) frequencies are based tions. In addition, the frequency of genetic variants as well as their
on published allele data (limited for some populations), (2) most distribution and combination within a haplotype may differ in ad-
studies test for a limited number of allelic variants, which may lead mixed populations when compared with parental populations. For
to an underestimation of certain alleles. For example, c.-806C>T example, CYP2C19 has a higher number of rare genetic variants
is often defaulted to a CYP2C19*17 assignment, although this in Hispanics (4%), a highly admixed population, when compared
SNV is also present on CYP2C19*4.002 (Figure 2). Likewise, if with Europeans (0.05%), and CYP2C19*2 and *17 were found at
no SNVs are found, CYP2C19*1 is assigned, which inflates the a higher frequency in self-reported Afro-Caribbean subjects from
frequency of this allele; (3) limited data for CNVs, and (4) errors Costa Rico than mestizos of the same country.72,73 In addition,
translating SNVs into haplotypes, which consequently may lead to frequencies may also profoundly differ among indigenous popula-
over-reporting or under-reporting of certain alleles, or some alleles tions, including those in North, South, and Central America.74–76
not being detected at all. Therefore, all calculations based on allele Considering CYP2C19*1 through *35 (excluding the
frequencies are estimates at best and should be used with caution. CYP2C19*36 and *37 deletion alleles), hundreds of allele

356 VOLUME 109 NUMBER 2 | February 2021 | www.cpt-journal.com

 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

g.1A>G g.80161A>G
c.1A>G (p.M1V) c.991A>G (p.I331V)
PharmVar CAVE tool
*4.001

*4.002

*17
c.-806C>T
g.12662A>G g.19154G>A g.80161A>G
c.332-23A>G c.681G>A c.991A>G (p.I331V)

*2

*35
Figure 2 Allele default assignment strategy used by many testing platforms. Many pharmacogenetic test platforms only comprise the
more commonly observed single nucleotide variations (SNVs) and do not include all known SNVs or star (*) alleles that are defined by the
Pharmacogene Variation Consortium (PharmVar). Consequently, some alleles may not be identified or receive an assignment by “default.”
Therefore, it is important to know which SNVs are tested in order to fully understand how alleles were called and translated into phenotype,
as well as to appreciate a test’s limitations. (a) Visualizes the two known CYP2C19*4 subvariants, which are nonfunctional due to c.1A>G
(p.M1V). The CYP2C19*4.002 subvariant also carries c.-806C>T, which is the core SNV defining the increased function *17 allele. Thus, if
testing only includes c.-806C>T and not c.1A>G, a CYP2C19*4 allele will be called as a *17 allele potentially leading to an incorrect phenotype
assignment. (b) illustrates that g.12662A>G (c.332-23A>G), which causes aberrant splicing, occurs on both, CYP2C19*2 and *35. If not
tested, the latter will be defaulted to a CYP2C19*1 assignment that may also lead to an incorrect phenotype assignment. (c) Shows the
comparison of CYP2C19*2, *4, *17, and *35, created with the PharmVar Comparative Allele ViewEr (CAVE) tool. Blue boxes indicate the
presence of a core SNV on all suballeles, whereas the gray box indicates that the core SNV is not present on all suballeles. The function (
) symbol indicates that a core SNV alters function and the PharmVar ( ) symbol highlights that a core SNV is unique to a star allele. SNV
positions refer to genomic coordinates on the NG_008384.3 reference sequence (with the ATG start codon being +1).

combinations are possible, and thus, the number of genotypes in a document describes the nomenclature system and provides ex-
population or patient cohort can be quite large. The actual number amples.78 For example, a new star number is only issued if a hap-
of combinations that occur in each population may be significantly lotype contains a sequence variant that: (1) results in an amino
less, however, depending on the number of alleles and their fre- acid change (e.g., CYP2C19*5 harbors an arginine to tryptophan
quencies. Phenotype frequencies across populations are provided change (c.1297C>T, p.R433W)), (2) abolishes a splice site (e.g.,
in the “Calculated phenotype frequency” tab in the PharmGKB/ c.681G>A in exon 5 of CYP2C19*2 alters a splice site that leads
CPIC CYP2C19 Frequency Table.44 We stress, however, to view to a frameshift and premature translation termination), or (3)
all phenotype group frequencies (including those shown in the changes expression levels causing decreased or increased function,
PharmGKB/CPIC table) with caution due to the limitations re- which is exemplified by the CYP2C19*17 defining variant (c.-
garding the accuracy of allele frequencies, as well as the method 806C>T) that increases expression. Importantly, new haplotypes
used to translate genotype into phenotype and inconsistencies in that contain previously characterized variants that obliterate
the classification of “population,” “ethnicity,” or “race.”77 function are catalogued under the original star (*) allele number
as a suballele. For example, any allele having a novel variant and
CYP2C19 ALLELE FUNCTION c.1297C>T (p.R433W), will be designated as a CYP2C19*5 sub-
PharmVar displays CPIC allele clinical function, using respective allele and considered having no function, regardless of the func-
CPIC terms (increased, normal, decreased, or no function, or un- tional status of the novel variant.
certain or unknown function). Alleles that have not been assigned
function by CPIC are shown as “not available.” The filter option THE PHARMVAR CYP2C19 GENE EXPERT PANEL
allows the user to sort alleles by functional status. See sections International experts representing CYP2C19 research, clini-
above regarding details of CPIC function assignment. cal testing, and implementation interests were recruited from
PharmVar members to serve on the CYP2C19 expert panel and
PHARMVAR NOMENCLATURE AND CYP2C19 ALLELE tasked with reviewing the current nomenclature and any new
DESIGNATION submissions. The panel also includes PharmGKB/CPIC represen-
PharmVar stores and displays allelic data consistently across genes, tation to ensure that the nomenclature is consistent with CPIC
relying on public standards and data sources wherever possible. guidelines and to facilitate dissemination to a greater audience
The standardized nomenclature follows criteria developed by gene through PharmGKB as well as other databases, such as ClinGen.
experts. The “Allele Designation and Evidence Level Criteria” The CYP2C19 expert panel first convened in January 2018 and

CLINICAL PHARMACOLOGY & THERAPEUTICS | VOLUME 109 NUMBER 2 | February 2021 357
 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

meets monthly via teleconferences and communicates regularly by also the option of two count modes (i.e., counting from the first
email as needed. The composition of the panel can be found on nucleotide in the reference sequence or the ATG translation start
the PharmVar website.79 Table 2 summarizes the alleles received codon being 1).
and accepted by the panel.
CYP2C19 HAPLOTYPE EVIDENCE LEVELS
THE PHARMVAR CYP2C19 GENE PAGE PharmVar designates the “Haplotype Evidence Level” for each
PharmVar maps sequence variations for each gene to its latest of the star alleles reported on the CYP2C19 gene page. Evidence
genomic and cDNA RefSeqs, issued by the National Center levels are displayed as symbols indicating “definitive” (Def),
for Biotechnology Information (NCBI) Reference Sequences “moderate” (Mod), or “limited” (Lim) levels of support for a
database,80and the GRCh37 (NC_000010.10) and GRCh38 given haplotype. This three-tiered system represents a modi-
(NC_000010.11) genome builds. For CYP2C19, the cur- fied ClinVar classification system; more detailed information is
rent genomic and transcript RefSeqs are NG_008384.3 and provided in the “Allele Designation and Evidence Level” docu-
NM_000769.1, respectively. CYP2C19 was transitioned from the ment.78 This type of information (e.g., whether an allele was se-
original nomenclature site to PharmVar in September 26, 2017 quenced across the gene, how a haplotype was determined) was
(original content is available through the “Archive” link on the not always systematically captured prior to PharmVar. For ex-
PharmVar homepage). isting haplotype definitions, a literature review was conducted
It is important to note that GRCh37 differs from GRCh38 in in order to assign evidence levels. Many alleles are currently
three positions, 96521422A>G (c.-1041G>A), 96522561T>C labeled as “Lim” because their definitions do not include any,
(c.99C>T), and 96602623G>A (c.991A>G). Therefore, the num- or the complete 2 kb of upstream sequence required for each
ber of variants displayed for GRCh37 differs from those shown for submission (as set by the CYP2C expert panel). This was the
GRCh38, which corresponds to the variant that was originally des- case for many allelic variants, including CYP2C19*2.001 and
ignated as CYP2C19*27 and was reclassified as *1.006. *3.001. Other alleles were labeled as “Mod,” although the re-
A Locus Reference Genomic (LRG) record was requested by quired upstream region was sequenced, but the phase of se-
PharmVar from the LRG Project, an NCBI (RefSeq) and European quence variants was computationally inferred and has not been
Molecular Biology Laboratory-European Bioinformatics Institute validated (e.g., CYP2C19*9-*14). The value of evidence levels is
(EMBL-EBI; Ensembl/GENCODE) initiative.81LRGs are uni- centered on providing users with as much information on hap-
versally accepted reference standards that are created specifically lotype reliability as possible and enabling users to quickly parse
for clinical reporting by manual curation. LRGs are stable enti- haplotypes based on robust, high evidence, vs. other haplotypes
ties that never change or version after their release. The LRG for with “Lim” or “Mod” evidence levels.
CYP2C19 (LRG_584, released 6-3-2020) matches 100% with PharmVar solicits submissions for all alleles labeled “Lim” and
the NG_008384.3 RefSeq and will be used by PharmVar as the “Mod” to ultimately raise their evidence levels to “Def.” Moreover,
“gold-standard” reference moving forward. PharmVar also encourages encore submissions for alleles with sin-
PharmVar is left-aligning, i.e., insertions and deletions of nu- gle citations and shown as “Def ” to further corroborate a haplo-
cleotides in a repeat or homopolymer sequence are listed using type definition.
the 3′ rule recommended by Human Genome Variation Society
(see PharmVar “Standards” for more details82 or the CYP2D6 PHARMVAR IDENTIFIERS
GeneFocus61 for examples). Due to different alignment meth- Each characterized haplotype receives a PharmVar ID (PVID).
ods, coordinates for insertion/deletion polymorphisms may differ The PVID is a unique numeric identifier analogous to database
among databases. No such variants have, however, been described SNP rsIDs. Star allele names are driven by functional grouping
for CYP2C19 to date. (i.e., they are not guaranteed to be permanent and can be subject to
On the PharmVar CYP2C19 gene page, the user can easily change (for details, see “corrections revisions, new alleles, updates”
cross-reference genomic and cDNA position(s) by choosing the section below)). Additional changes may be necessary in the future
respective reference sequence or genome build of interest; there is as more information becomes available. If an allele’s star designa-
tion is updated to a new star number, the PVID of the haplotype
Table 2 Novel allele(s) and confirmed suballele(s) remains constant and does not change. In contrast, if a haplotype
Core allele designation Novel alleles/suballeles
definition changes (e.g., through the addition or removal of vari-
ants) a new PVID will be assigned. An example is the above-men-
*1 *1.004, *1.005 tioned CYP2C19*27 allele. Its initial PVID (PV00073) remained
*2 *2.0010 the same. In contrast, CYP2C19*28 received a new PVID
*36 Complete deletion of CYP2C19 gene (PV00622) after its haplotype was revised (i.e., two SNVs were
*37 Partial deletion of CYP2C19 gene (i.e., removed). Original PVIDs and their haplotype definitions can be
deletions that include at least the entire tracked in the database via the PVID Lookup function.
exon 1)
Submissions for known alleles; original haplotype was confirmed, CURATION EFFORTS
and evidence level raised to “Def”
Extensive curation efforts were part of the content transfer from
*1.002, *2.002, *15.001, *17.001, and *35.001
the P450 nomenclature webpage into the PharmVar database to

358 VOLUME 109 NUMBER 2 | February 2021 | www.cpt-journal.com

 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

standardize the annotations to the above-mentioned conventions The expert panel recommended to reassign CYP2C19*27 as a
(Table 3). The following sections describe general and specific ef- CYP2C19*1 suballele (CYP2C19*1.006), due to limited informa-
forts undertaken. tion regarding c.-1041G>A. This allele was originally given its own
star designation based on in vitro data, suggesting that c.-1041G>A
Gene Region mapped/required for Allele Definition may decrease expression levels.84 There is no in vivo evidence, how-
Haplotype definitions for CYP2C19 allelic variation is based on ever, supporting the initial findings.
the sequence of the coding regions (counting from the start and As of February 2020, the CYP2C19 expert panel has desig-
stop codons, respectively), as well as upstream and downstream nated two novel alleles (*36 and *37) and three new suballeles.
regions. Partial or incomplete gene sequences are not accepted by In addition, five alleles that were based on partial information are
PharmVar. Many of the allelic variants that have been transitioned now fulfilling PharmVar allele definition requirements and their
from the original nomenclature site to PharmVar have limited evidence level was raised to “Def ” (Table 2). There are still nu-
information, such as: (1) upstream and downstream regions were merous allele definitions with “Lim” and “Mod” evidence levels for
not, or only partially sequenced, (2) exons were not sequenced, but which PharmVar seeks submissions. The “Change Log” document
analyzed through different methodologies, and (3) the haplotype also tracks submissions and indicates the star alleles that have been
definition was inferred and, therefore, may have SNVs in exons, updated.
flanking intronic, and/or upstream and downstream regions that
have not been detected. Challenges of allele designation
Despite detailed allele definition criteria, assigning star allele
Corrections, revisions, new alleles, and other updates definitions can be challenging, as illustrated by two indepen-
Coordinates were mapped to NG_008384.2 by the nomenclature dent submissions. The haplotype of the first submission was
site and updated to the current CYP2C19 RefSeq NG_008384.3 designated as a CYP2C19*1.005 suballele. The second sub-
as the gene was transitioned to PharmVar. This upgrade did not mission, accompanied by unpublished in vitro data, was by the
change position annotations of SNVs. author who previously published this haplotype, 85 requesting
As CYP2C19 was transitioned into the PharmVar database, com- that this allele receive its own star number. Upon careful review
ments and footnotes were removed and errors corrected. References and communication with the submitter, the panel unanimously
in support of allele definitions have been updated and those solely agreed that the published and unpublished data were inconsis-
describing function removed (references for function are pro- tent and not sufficient for issuing this allelic variant its own star
vided in the PharmGKB/CPIC CYP2C19 Allele Functionality name.
table44). A number of descriptors, such as “g.12662A>G is likely Upon reviewing the limited data available for CYP2C19*27,
part of all *2 alleles” have been removed and the intronic SNV the panel recommended to reclassify this allelic variant as a
at g.12662A>G (c.332-23A>G), which also causes a splicing de- CYP2C19*1 suballele (noting that this haplotype should not have
fect, has been added to all CYP2C19*2 haplotype definitions. It received a new star number when it was first accepted by the P450
remains unknown, however, whether all CYP2C19*2 haplotypes Nomenclature Committee, given the limited in vitro functional
indeed have this SNV in addition to c.681G>A, which is shared data). The re-assignment of CYP2C19*27 as CYP2C19*1.006 is
among all CYP2C19*2 alleles. Changes and revisions are detailed consistent with the PharmVar criteria, as well as consistent of how
in the “Change Log” document on the CYP2C19 gene page.83 the criteria were applied to the allele designated CYP2C19*1.005.
These examples illustrate that allele designation is not always
straightforward and needs to be a dynamic process to accommo-
Table 3 Summary of edits and changes during the
date new findings.
transitioned into the PharmVar database and notable
changes made thereafter
Update and reassignment of CYP2C19*1 (CYP2C19*1.001)
Reason Change Affected alleles
Since the inception of PharmVar, the CYP2C19 panel and
Standardization Intronic SNVs were *2C, *2D, *3B, *18, Steering Committee discussed that the previously issued
removed and *19 CYP2C19*1 suballele definitions do not conform to formal
Retired star alleles *20, *21 PharmVar allele designation criteria.78 The CYP2C19*1.001
removed
lacks c.991A>G (p.I331V), whereas this amino acid changing
Comment “g.12662A>G *2A, *2C, *2B, *2E, SNV is present in the CYP2C19*1.002-*1.006 suballeles. The
is likely part of all *2 *2F, *2G, *2H, and
alleles” was removed *2J original designation of the latter as CYP2C19*1 suballeles was
g.12662A>G was most likely due to the interpretation of p.I331V having no ef-
added to allele fect on enzyme activity. The common c.991A>G (p.I331V)
definition variant is also found on many other CYP2C19 haplotypes
Comments removed *3B, *11, *16, and and thus represents the major allele across most populations
*30
(c.991A>G, G = 0.9386 per the gnomAD exome database).
Other Reassigned to *1.006 *27 This prompted the question of whether the haplotype repre-
Correction *27 (99T>C was senting CYP2C19*1.002 would be a more appropriate reference
revised to 99C>T)
sequence for allele definition (i.e., if allele definitions would

CLINICAL PHARMACOLOGY & THERAPEUTICS | VOLUME 109 NUMBER 2 | February 2021 359
 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

be reported against CYP2C19*1.002 as reference, the majority changing expression levels or interfere with splicing and are present
of allelic variants would not have c.991A>G, nor c.99C>T, a in all suballeles within an allele group. With this rule-based system,
synonymous SNV that is present on the majority of alleles as suballeles are collapsed into a single “core” definition representing
well). Considering a current massive standardization effort un- all suballeles categorized under a star (*) number. For example, the
dertaken by the NCBI and EMBL-EBI, neither the CYP2C19 growing number of CYP2C19*2 suballeles share two SNVs that
RefSeq, nor the LRG will be revised to accommodate star no- fulfill this rule (i.e., c.332-23A>G (g.12662A>G) and c.681G>A
menclature. Based on a majority consensus, the allele originally (g.19154G>A). Both SNVs have been shown to cause aberrant
catalogued as CYP2C19*1.001 (i.e., *1A) has now received its splicing, preventing the formation of a functional gene product.
own star designation (i.e., CYP2C19*38). The core allele defi- Thus, both SNVs constitute the CYP2C19*2 core allele definition
nition of this variant does not have any SNVs because this (Figure 3). For CYP2C19*4, only c.1A>G (p.M1V) that abolishes
haplotype matches the NG_008384.3 RefSeq. In contrast, the translation initiation is shared among all subvariants and, there-
CYP2C19*1 allele definition, representing the most frequently fore, is the sole variant of the *4 core allele definition (Figure 2).
observed CYP2C19 allele, gained a core SNP because all vari- Of importance, a sequence variant found in a core allele defini-
ants catalogued under this designation have p.I331V. tion is not necessarily unique to that haplotype as illustrated by the
RefSeqs are generally representing a haplotype that encodes two SNVs of the CYP2C19*2 core definition (Figure 2). Indeed,
functional protein and is commonly found across populations, this, c.332-23A>G is also present in CYP2C19*35 and constitutes
however, is not always the case (the CYP3A5 RefSeq, for example, the sole core SNV for this allele. Another example is c.-806C>T,
represents functional protein, but the nonfunctional CYP3A5*3 which increases mRNA expression and, hence, fulfills core SNV
variant is the most common across many populations). In addition, definition requirements for CYP2C19*17. This SNV is, however,
a RefSeq may have changed over the past two decades, as it has for also part of one of the two known CYP2C19*4 haplotypes.
CYP2C19 (the genomic sequences among GRCh37 and GRCh38 One challenge with core allele definitions is that a definition
differ in three positions, i.e., c.-1041, c.99, and c.991) making con- may change over time as new information becomes available. For
sistent reporting of alleles difficult. As the NCBI is standardizing instance, based on the rules described above, the current core al-
reference sequences and releases LRGs, PharmVar has not only a lele definition of CYP2C19*2 harbors c.332-23A>G, c.681G>A,
unique opportunity, but also an obligation to follow suit, and align and c.990C>T (p.I331V). According to the CYP2C19 allele
allele definitions with globally accepted RefSeqs as well as update designation criteria, all new alleles having c.681G>A, and, there-
definitions that are inconsistent. Although changes, such as the fore, are nonfunctional due to aberrant splicing, will be assigned
CYP2C19*1.001 to *38 update may be viewed as “interruptive,” as CYP2C19*2 suballeles, regardless of the nature of the other
having a standardized nomenclature system moving forward is pre- SNV(s) present. Consequently, the CYP2C19*2 core allele defini-
ferred as reflected by the expert consensus vote. tion may change, should a sequence containing c.681G>A, but not
What does this update mean for the interpretation of genotyp- c.332-23A>G, and/or c.990C>T (p.I331V) be identified.
ing test results? If no SNVs are found, and c.991A>G (p.I331V) is The core alleles are the basis of the CYP2C19 allele defini-
not tested, a CYP2C19*1 will be assigned based on defaulting to tion table used in CPIC guidelines and by PharmGKB44). The
the more common allele having c.991G. Clinical genotyping tests CYP2C19 core allele definitions are also utilized for clinical anno-
typically do not include c.991A>G (p.I331V) (it is also not recom- tations in PharmGKB.
mended by AMP for tier 1 or 2 testing24) and thus CYP2C19*38
alleles will be called as *1 by default. If genotype is based on se- THE PHARMVAR COMPARATIVE ALLELE VIEWER
quencing and information for c.991A>G (p.I331V) is available, a The Comparative Allele ViewEr (CAVE) tool was developed by
small fraction of alleles will be called as CYP2C19*38. The intro- PharmVar to easily compare core alleles.86 This tool can be ac-
duction of CYP2C19*38 will not impact phenotype assignments cessed using the “Compare View” button on the CYP2C19 gene
as it will be designated as having normal function. page. Figure 2c exemplifies the utility of this tool on CYP2C19*2,
*4, *17, and *35. In this display mode, it is easy to see which core
CORE ALLELE DEFINITIONS SNVs are shared among haplotypes, whether they alter function
For many alleles, there is a growing number of suballeles that share and/or are unique to a haplotype.
one or more “key” defining sequence variant(s), which are consid-
ered “core” SNVs by PharmVar. Although suballele information REPORTING GENOTYPE AND TRANSLATION INTO
can be valuable for the design of test platforms (sequence or gen- PHENOTYPE
otype-based alike) and interpretation of genotyping test results, PharmVar and PharmGKB have also collaboratively developed
there is no need to distinguish suballeles for phenotype prediction templates to facilitate more consistent and transparent reporting
because all alleles under a star number are defined per criteria as of genotype details and how genotype is translated into phenotype
functionally equivalent. Thus, even if a test is capable of distin- to be used by the community to include more detailed informa-
guishing suballeles, these can be simply reported as the defining tion as part of the submission of research findings to publishers.
allele (e.g., CYP2C19*1, *2, etc.) This information can be provided as supplemental materials of a
PharmVar and the PharmGKB have collaboratively developed publication to facilitate access to important data for subsequent
core allele definitions. A core allele is defined only by sequence curation. The first template file (Supplementary Materials S1)
variations that cause an amino acid change or impact function by collects information, including methods or platforms used for

360 VOLUME 109 NUMBER 2 | February 2021 | www.cpt-journal.com

 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

(a)

c.332-23A>G c.681G>A c.991A>G

(b) (splice defect) (splice defect) (p.I331V)

1 2 3 4 5 6 7 8 9

core alleles
*2

*11 1 2 3 4 5 6 7 8 9

c.449G>A c.991A>G
(p.R150H) (p.I331V)
c.332-23A>G

(c)
(splice defect)

(splice defect)

c.991A>G
c.449G>A
c.276G>C

c.6981>A

c.990C>T
c.-98T>C
c.99C>T

(p.R150H)

(p.I331V)
(p.E92D)

*2.001

*2.002

suballeles
*2.010

*11.001

Figure 3 Overview of core alleles, suballeles, and the graphical Core Allele ViewEr (CAVE). (a) Shows the CYP2C19*2 core allele definition
(gray bar). Core single nucleotide variations (SNVs), Pharmacogene Variation (PharmVar) ID (PVID), and evidence level is shown for each
allele. All currently defined suballeles are displayed. Legacy allele designations are cross-referenced (e.g., *2.001 corresponds to *2A).
Note that g.12662A>G (c.332-23A>G) was added to all suballeles, when the gene was transferred to PharmVar (Table 3). (b) Is a graphical
representation of the CYP2C19*2 and *11 core alleles and their core SNVs. c.991A>G (p.I331V) is present on both alleles, whereas c.332-
23A>G and c.681G>A (splice defects) are the CYP2C19*2 core SNVs and c.449G>A (p.R150H) is the core SNV defining CYP2C19*11. Core
SNVs are shown by red lines; gray boxes represent the nine exons (scale is approximated). (c) Shows three of the ten CYP2C19*2 suballeles
defined to date and the only CYP2C19*11 haplotype. Of note, the core SNV of the latter (c.449G>A (p.R150H)) is also present on the
CYP2C19*2.010 suballele. Core SNVs (causing an amino acid change or aberrant splicing) are shown in red, all other SNVs are highlighted in
blue. c.449G>A (p.R150H) is not a *2 core SNP because it is not found on all suballeles.

genotyping and which SNVs are interrogated; the template also as well as other groups, to use their standardized translation
provides a standardized set-up for reporting genotype results for method, not every investigator or laboratory adopts this method.
individual subjects, as well as allele frequencies. The second tem- Too often, papers reference previous work stating that “geno-
plate file (Supplementary Materials S2) facilitates the reporting typing was performed as previously described” or indicate that
of how genotype was translated into phenotype in a study, as well “CYP2C19 phenotype was correlated with the metabolism of a
as genotype frequencies. Although it is recommended by CPIC, drug” without specifying which SNVs or alleles were genotyped or

CLINICAL PHARMACOLOGY & THERAPEUTICS | VOLUME 109 NUMBER 2 | February 2021 361
 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

>A >A >T

1G 1G 6C
c.6
8 8 80
(a) (b) c.6 (c) c.-

>G P >G P P
3A SN 3A SN SN
-2 v el -2 v el v el
32 no 32 no no
c.3 c.3

pedigree informs specific amplifica on single molecule sequencing, Long Read

SNP linkage informs SNP linkage mapping or ddPCR linkage analysis can
inform linkage over larger distances
*2/*new *1/*2 *2/*new *1/*new

2C19*2 2C19*1 2C19*2 new * allele

new * allele 2C19*2 new * allele

2C19*1
parent parent
OR OR

*2/*new *1/new *4 suballele *17*new

new *2 CYP2C19*17
2C19*2 suballele

child G

new * allele 2C19*1 new * allele

(d)

*17.001 *1.002
*17.001 *1.002
*17.001 *17.001 *35.001
*35.001
father mother

*17.001
*35.001
child

*17.001

*35.001

Figure 4 Experimental approaches for phasing single nucleotide variations (SNVs) to establish haplotype. (a, b) Depict a subject who is
heterozygous for three SNVs of which two designate the CYP2C19*2 allele. Whether the novel SNV is in cis or trans with the CYP2C19*2 SNVs
can be informed by, for example, inheritance (a), or experimentally determined using, for example, allele-specific (long-range) polymerase
chain reaction (PCR), followed by sequencing (b), given that the distance among the novel SNV is not exceeding PCR amplification limitations.
(c) Provides an example of a subject who is heterozygous for two SNPs, one of which designates the CYP2C19*17 allele. In this instance, the
phase of the novel SNV may be inferred by inheritance (not shown) or experimentally determined by single molecule sequencing or utilizing
digital droplet PCR-based SNP-linkage analysis. (d) Details a submission for the CYP2C19*35 allele. The haplotype of the CYP2C19*35
suballele was determined from whole genome sequence data of each member of the trio. This data revealed an additional SNV on the
CYP2C19*35 haplotype, as well as provided evidence to support changing the allele’s evidence level to “Def.”

362 VOLUME 109 NUMBER 2 | February 2021 | www.cpt-journal.com

 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

how phenotype was assigned. The lack of such information makes characterize a novel haplotype. Figure 4a demonstrates that ped-
it extremely difficult, if not impossible, for curators to compare igree analysis can be used to determine whether a novel SNV is in
findings or extract information for CPIC guideline development. cis or trans with those defining the CYP2C19*2 allele. Figure 4b
Colleagues are, therefore, strongly encouraged to utilize the pro- illustrates, that allele-specific PCR using primers, that only am-
vided templates, or revised versions thereof, for publication of plify when SNVs are either in cis or trans, can be utilized to estab-
these types of information as supplemental materials. lish SNV linkage; this approach, however, only works if the SNVs
to be linked are within a distance that can reliably be amplified
CYP2C19 REFERENCE MATERIALS by PCR. Figure 4c illustrates the utility of single molecule real
The Genetic Testing Reference Material Program is a combined ef- time sequencing methods (e.g., Pacific Biosciences, Menlo Park,
fort among the Centers for Disease Control and Prevention–based CA or Oxford Nanopore Technologies, Oxford, UK, and 10X
Genetic Testing Reference Material Coordination Program, Coriell Genomics Long-Read mapping, Pleasanton, CA) to establish
Institute for Medical Research, and members of the PGx testing com- haplotypes over a range of distances. SNVs can also be phased
munity. Considering the growing use of PGx testing, established sets over long distances of up to 200 kb using digital droplet PCR
of well-characterized reference materials are needed for assay devel- to establish haplotypes.91,92 In Figure 4d, we offer an example of
opment, validation, quality control, and proficiency testing. To ad- a PharmVar submission for CYP2C19*35. The haplotype of the
dress the increasing need for reference materials, a set of 137 genomic CYP2C19*35 was determined from whole genome sequence data
DNA samples were characterized for 28 pharmacogenes, including of each member of the family trio (father, mother, and child). The
CYP2C19 and “consensus” genotypes established.67 Although the father is homozygous for the CYP2C19*17.001 allele and, thus,
most common variants were assayed, many rare alleles were not iden- passes these to his offspring. All SNVs encompassing the child’s
tified among the samples tested; efforts are ongoing to find additional CYP2C19*35 allele must have been inherited from the mother.
samples to complement the existing materials for CYP2C19. Testing This analysis revealed that the allele has an additional SNV, c.-
and research laboratories can acquire these materials from the Coriell 913G>A, which was not part of the initial haplotype definition.
Institute (Camden, NJ), as they are publicly available. Based on this information, the CYP2C19*35 allele definition was
updated, and its evidence level revised from “Lim” to “Def.”
INFERRING CYP2C19 HAPLOTYPE FROM NEXT-GENERATION To detect CNVs, methods, such as multiplex ligation-depen-
SEQUENCE DATA AND PUBLIC DATABASES dent probe amplification, TaqMan copy number assays, or digital
Bioinformatic tools have been developed to call star (*) alleles from droplet PCR, may be used to evaluate or confirm CYP2C19 gene
next-generation sequencing panels and/or exome/genome-based copy number status. CYP2C19 CNVs are rare and are most often
sequence data for several CYP genes. Available tools include detected as incidental findings by array platforms, such as array
Aldy,87 Stargazer,88 Astrolabe,89,90 and Pharmacogenomics comparative genomic hybridization, performed for unrelated diag-
Clinical Annotation Tool.36 These tools use algorithms to predict nostic purposes.
the individual diplotype, based on the known alleles catalogued
by PharmVar. These programs have not yet been systematically CONCLUSIONS
validated, however. In addition, high quality data are needed in This is the second of a series of gene-centric review articles, focus-
order to generate reliable genotype calls with next-generation se- ing on important pharmacogenes. The CYP2C19 GeneFocus pro-
quencing. Finally, CNVs affecting the CYP2C19 gene locus may vides essential information for the understanding of this highly
also lead to misinterpretation in rare cases.70 polymorphic gene, complementing clinically relevant information
provided by CPIC guidelines and other PGx resources. We are
Methods for CYP2C19 allele characterization highlighting PharmVar efforts and challenges of systematically
As mentioned above, the CYP2C19 gene spans almost 93 kb, which cataloging CYP2C19 allelic variation as well as collaborative ef-
makes haplotype analysis nontrivial, especially in cases when the forts with the PharmGKB to make the information useful and
SNVs for which linkage needs to be established exceed the dis- easily accessible to the entire PGx community.
tance that can be covered by XL-PCR amplification (typically
under 20 kb). If a subject is heterozygous for a single SNV or ho- SUPPORTING INFORMATION
Supplementary information accompanies this paper on the Clinical
mozygous for one or more SNVs, the haplotype is unambiguous. If Pharmacology & Therapeutics website (www.cpt-journal.com).
a novel SNV (or SNVs) is detected along a known SNV (or SNVs),
of which all are heterozygous, it remains unknown, whether the ACKNOWLEDGMENTS
novel SNV(s) is/are in cis or trans with the other SNV(s). Function The authors thank Dr Mia Wadelius, Uppsala University, Sweden, for
may depend on which SNVs are occurring together on the same serving on the PharmVar CYP2C expert panel since its inception and
critical comments on the manuscript. We also like to sincerely thank Dr
chromosome; the latter will also dictate whether the novel hap- Roger Gaedigk for his assistance with graphical artwork and critically
lotype will be cataloged as a suballele or receive its own star num- reading the manuscript.
ber. Haplotypes may also be computationally inferred, which may,
however, not result in high-confidence predictions, especially if FUNDING
This work was funded by the National Institutes of Health for the
the novel haplotype was observed in a single individual. Pharmacogene Variation Consortium (R24 GM123930; P.I., A.G.)
Thus, the gold standard for allele definition is experimental val- and PharmGKB (U24 HG010615; P.I., T.E.K.). V.M.P. is supported
idation. Figure 4 provides an overview of methods/approaches to by the Implementing Genomics in Practice (IGNITE) project grant

CLINICAL PHARMACOLOGY & THERAPEUTICS | VOLUME 109 NUMBER 2 | February 2021 363
 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

(U01 HG010245). J.D. is supported by grant U54 MD007600-33 (RCMI 15. PharmGKB CYP2C19 Drug Labels Annotations <https://www.
Program) from NIH/NIMHD. pharm​gkb.org/gene/PA124/​label​Annot​ation>. Accessed April 6,
2020.
CONFLICTS OF INTEREST 16. FDA Table of Pharmacogenomics Biomarkers in Drug Labeling
Indiana University School of Medicine Pharmacogenomics Laboratory, <https://www.fda.gov/drugs/​scien​ce-resea​rch-drugs/​table​- pharm​
Millennium Health, and Sema4 are fee-for-service clinical laboratories acoge​nomic ​- bioma​rkers​- drug-labeling>. Accessed April 6, 2020.
that offer clinical pharmacogenetic testing. A.L.D. is a paid employee 17. FDA Table of Pharmacogenetics Associations <https://www.fda.
of Millennium Health; S.A.S., is a paid employee of Sema4. All other gov/medic​al-devic​es/preci​sion-medic​ine/table ​- pharm​acoge​netic​
authors declared no competing interests for this work. -assoc​iations>. Accessed April 6, 2020.
18. CPIC Gene-drug pairs <https://cpicp​g x.org/genes​-drugs>.
AUTHOR CONTRIBUTIONS Accessed April 6, 2020.
All authors wrote the manuscript. 19. Cavallari, L.H. et al. The IGNITE pharmacogenetics working group:
an opportunity for building evidence with pharmacogenetic
implementation in a real-world setting. Clin. Transl. Sci. 10,
© 2020 The Author. Clinical Pharmacology & Therapeutics published 143–146 (2017).
by Wiley Periodicals LLC on behalf of American Society for Clinical 20. Cavallari, L.H. et al. Multisite investigation of outcomes with
Pharmacology and Therapeutics. implementation of CYP2C19 genotype-guided antiplatelet therapy
after percutaneous coronary intervention. JACC Cardiovasc. Interv.
This is an open access article under the terms of the Creat​ive Commo​ 11, 181–191 (2018).
ns Attri​b utio​n -NonCo​m merc​ial-NoDerivs License, which permits use 21. Claassens, D.M.F. et al. A genotype-guided strategy for oral
and distribution in any medium, provided the original work is properly P2Y12 inhibitors in primary PCI. N. Engl. J. Med. 381, 1621–1631
cited, the use is non-commercial and no modifications or adaptations (2019).
are made. 22. Bergmeijer, T.O. et al. Genome-wide and candidate gene
approaches of clopidogrel efficacy using pharmacodynamic
and clinical end points-rationale and design of the International
1. Kupfer, A. & Preisig, R. Pharmacogenetics of mephenytoin: a new Clopidogrel Pharmacogenomics Consortium (ICPC). Am. Heart J.
drug hydroxylation polymorphism in man. Eur. J. Clin. Pharmacol. 198, 152–159 (2018).
26, 753–759 (1984). 23. Patel, J.N. et al. Evaluation of CYP2C19 genotype-guided
2. Goldstein, J.A. et al. Evidence that CYP2C19 is the major voriconazole prophylaxis after allogeneic hematopoietic cell
(S)-mephenytoin 4’-hydroxylase in humans. Biochemistry 33, transplant. Clin. Pharmacol. Ther. 107, 571–579 (2020).
1743–1752 (1994). 24. Pratt, V.M. et al. Recommendations for clinical CYP2C19
3. Wrighton, S.A., Stevens, J.C., Becker, G.W. & VandenBranden, genotyping allele selection: a report of the association for
M. Isolation and characterization of human liver cytochrome molecular pathology. J. Mol. Diagn. 20, 269–276 (2018).
P450 2C19: correlation between 2C19 and S-mephenytoin 25. Bousman, C.A., Zierhut, H. & Muller, D.J. Navigating the labyrinth
4’-hydroxylation. Arch. Biochem. Biophys. 306, 240–245 (1993). of pharmacogenetic testing: a guide to test selection. Clin.
4. Romkes, M., Faletto, M.B., Blaisdell, J.A., Raucy, J.L. & Goldstein, Pharmacol. Ther. 106, 309–312 (2019).
J.A. Cloning and expression of complementary DNAs for 26. Bousman, C.A., Jaksa, P. & Pantelis, C. Systematic evaluation
multiple members of the human cytochrome P450IIC subfamily. of commercial pharmacogenetic testing in psychiatry: a focus
Biochemistry 30, 3247–3255 (1991). on CYP2D6 and CYP2C19 allele coverage and results reporting.
5. Goldstein, J.A. & de Morais, S.M. Biochemistry and molecular Pharmacogenet. Genom. 27, 387–393 (2017).
biology of the human CYP2C subfamily. Pharmacogenetics 4, 27. Malki, M.A. & Pearson, E.R. Drug-drug-gene interactions and
285–300 (1994). adverse drug reactions. Pharmacogenomics J. 20, 355–366
6. de Morais, S.M., Wilkinson, G.R., Blaisdell, J., Nakamura, (2020).
K., Meyer, U.A. & Goldstein, J.A. The major genetic defect 28. Zhou, Z.W. et al. Clinical association between pharmacogenomics
responsible for the polymorphism of S-mephenytoin metabolism and adverse drug reactions. Drugs 75, 589–631 (2015).
in humans. J. Biol. Chem. 269, 15419–15422 (1994). 29. Maher, R.L., Hanlon, J. & Hajjar, E.R. Clinical consequences of
7. Meyer, U.A. & Zanger, U.M. Molecular mechanisms of genetic polypharmacy in elderly. Expert Opin. Drug Saf. 13, 57–65 (2014).
polymorphisms of drug metabolism. Annu. Rev. Pharmacol. Toxicol. 30. Drug Interactions Flockhart TableTM <https://edswi.org/wp-conte​
37, 269–296 (1997). nt/uploa​ds/2018/02/Flock​hart-Table ​- Medic​ation ​- Metab​olism.
8. Scott, S.A. et al. PharmGKB summary: very important pdf>. Accessed April 6, 2020.
pharmacogene information for cytochrome P450, family 2, 31. Gravel, S., Chiasson, J.L., Turgeon, J., Grangeon, A. & Michaud, V.
subfamily C, polypeptide 19. Pharm. Genomics 22, 159–165 (2012). Modulation of CYP450 activities in patients with type 2 diabetes.
9. Sim, S.C. et al. A common novel CYP2C19 gene variant causes Clin. Pharmacol. Ther. 106, 1280–1289 (2019).
ultrarapid drug metabolism relevant for the drug response to 32. Angiolillo, D.J. et al. Randomized comparison of a high clopidogrel
proton pump inhibitors and antidepressants. Clin. Pharmacol. maintenance dose in patients with diabetes mellitus and coronary
Ther. 79, 103–113 (2006). artery disease: results of the Optimizing Antiplatelet Therapy in
10. Ingelman-Sundberg, M., Daly, A.K., Oscarson, M. & Nebert, D.W. Diabetes Mellitus (OPTIMUS) study. Circulation 115, 708–716
Human cytochrome P450 (CYP) genes: recommendations for the (2007).
nomenclature of alleles. Pharmacogenetics 10, 91–93 (2000). 33. Zanger, U.M. & Schwab, M. Cytochrome P450 enzymes in drug
11. Daly, A.K. et al. Nomenclature for human CYP2D6 alleles. metabolism: regulation of gene expression, enzyme activities,
Pharmacogenetics 6, 193–201 (1996). and impact of genetic variation. Pharmacol. Ther. 138, 103–141
12. Ingelman-Sundberg, M., Oscarson, M., Daly, A.K., Garte, S. & (2013).
Nebert, D.W. Human cytochrome P-450 (CYP) genes: a web page 34. Laine, K., Tybring, G. & Bertilsson, L. No sex-related differences
for the nomenclature of alleles. Cancer Epidemiol. Biomarkers but significant inhibition by oral contraceptives of CYP2C19
Prev. 10, 1307–1308 (2001). activity as measured by the probe drugs mephenytoin and
13. Gaedigk, A. et al. The pharmacogene variation (PharmVar) omeprazole in healthy Swedish white subjects. Clin. Pharmacol.
consortium: incorporation of the human cytochrome P450 Ther. 68, 151–159 (2000).
(CYP) allele nomenclature database. Clin. Pharmacol. Ther. 103, 35. Hagg, S., Spigset, O. & Dahlqvist, R. Influence of gender and
399–401 (2018). oral contraceptives on CYP2D6 and CYP2C19 activity in healthy
14. Gaedigk, A., Whirl-Carrillo, M., Pratt, V.M., Miller, N.A. & Klein, volunteers. Br. J. Clin. Pharmacol. 51, 169–173 (2001).
T.E. PharmVar and the landscape of pharmacogenetic resources. 36. Xie, H.G., Huang, S.L., Xu, Z.H., Xiao, Z.S., He, N. & Zhou, H.H.
Clin. Pharmacol. Ther. 107, 43–46 (2020). Evidence for the effect of gender on activity of (S)-mephenytoin

364 VOLUME 109 NUMBER 2 | February 2021 | www.cpt-journal.com

 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

4’-hydroxylase (CYP2C19) in a Chinese population. 57. Dai, D.P. et al. In vitro functional analysis of 24 novel CYP2C19
Pharmacogenetics 7, 115–119 (1997). variants recently found in the Chinese Han population.
37. Sviri, S., Shpizen, S., Leitersdorf, E., Levy, M. & Caraco, Y. Xenobiotica 45, 1030–1035 (2015).
Phenotypic-genotypic analysis of CYP2C19 in the Jewish Israeli 58. Lau, P.S., Leong, K.V., Ong, C.E., Dong, A.N. & Pan, Y. In vitro
population. Clin. Pharmacol. Ther. 65, 275–282 (1999). functional characterisation of cytochrome P450 (CYP) 2C19 allelic
38. Yanni, S.B., Annaert, P.P., Augustijns, P., Ibrahim, J.G., Benjamin, variants CYP2C19*23 and CYP2C19*24. Biochem. Genet. 55,
D.K. Jr & Thakker, D.R. In vitro hepatic metabolism explains 48–62 (2017).
higher clearance of voriconazole in children versus adults: role of 59. Zhang, H., De, T., Zhong, Y. & Perera, M.A. The advantages
CYP2C19 and flavin-containing monooxygenase 3. Drug Metab. and challenges of diversity in pharmacogenomics: can minority
Dispos. 38, 25–31 (2010). populations bring us closer to implementation? Clin. Pharmacol.
39. Zane, N.R., Chen, Y., Wang, M.Z. & Thakker, D.R. Cytochrome Ther. 106, 338–349 (2019).
P450 and flavin-containing monooxygenase families: age- 60. Sirugo, G., Williams, S.M. & Tishkoff, S.A. The missing diversity in
dependent differences in expression and functional activity. human genetic studies. Cell 177, 1080 (2019).
Pediatr. Res. 83, 527–535 (2018). 61. Nofziger, C. et al. PharmVar genefocus: CYP2D6. Clin. Pharmacol.
40. Koukouritaki, S.B. et al. Developmental expression of human Ther. 107, 154–170 (2020).
hepatic CYP2C9 and CYP2C19. J. Pharmacol. Exp. Ther. 308, 62. Cavallari, L.H. et al. Multi-site investigation of strategies for the
965–974 (2004). clinical implementation of CYP2D6 genotyping to guide drug
41. Burgess, K.S. et al. Age-related changes in microRNA expression prescribing. Genet. Med. 21, 2255–2263 (2019).
and pharmacogenes in human liver. Clin. Pharmacol. Ther. 98, 63. Caraballo, P.J., Bielinski, S.J., St. Sauver, J.l. & Weinshilboum,
205–215 (2015). R.M. Electronic medical record-integrated pharmacogenomics and
42. Whirl-Carrillo, M. et al. Pharmacogenomics knowledge for related clinical decision support concepts. Clin. Pharm. Ther. 102,
personalized medicine. Clin. Pharmacol. Ther. 92, 414–417 (2012). 254–264 (2017).
43. PharmGKB CYP2C19 gene page <https://www.pharm​gkb.org/ 64. Hicks, J.K., Dunnenberger, H.M., Gumpper, K.F., Haidar, C.E. &
gene/PA124>. Accessed April 6, 2020. Hoffman, J.M. Integrating pharmacogenomics into electronic
44. PharmGKB Gene-specific Information Tables for CYP2C19 health records with clinical decision support. Am. J. Health Syst.
<https://www.pharm​gkb.org/page/cyp2c​19Ref​Mater​ials>. Pharm. 73, 1967–1976 (2016).
Accessed April 6, 2020. 65. Gray, K.A., Yates, B., Seal, R.L., Wright, M.W. & Bruford, E.A.
45. Sangkuhl, K. et al. Pharmacogenomics clinical annotation tool Genenames.org: the HGNC resources in 2015. Nucleic Acids Res.
(PharmCAT). Clin. Pharmacol. Ther. 107, 203–210 (2020). 43, D1079–D1085 (2015).
46. Caudle, K.E. et al. Incorporation of pharmacogenomics into 66. Halavi, M., Maglott, D., Gorelenkov, V. & Rubinstein, W.The NCBI
routine clinical practice: the Clinical Pharmacogenetics Handbook [Internet]. 2nd edition. MedGen <https://www.ncbi.
Implementation Consortium (CPIC) guideline development nlm.nih.gov/books/​NBK15​9970/> (2018).
process. Curr. Drug Metab. 15, 209–217 (2014). 67. Pratt, V.M. et al. Characterization of 137 genomic DNA reference
47. Relling, M.V., Klein, T.E., Gammal, R.S., Whirl-Carrillo, M., materials for 28 pharmacogenetic genes: a GeT-RM collaborative
Hoffman, J.M. & Caudle, K.E. The clinical pharmacogenetics project. J. Mol. Diagn. 18, 109–123 (2016).
implementation consortium: 10 years later. Clin. Pharmacol. Ther. 68. Moyer, A.M., Rohrer Vitek, C.R., Giri, J. & Caraballo, P.J.
107, 171–175 (2020). Challenges in ordering and interpreting pharmacogenomic tests in
48. Hicks, J.K. et al. Clinical pharmacogenetics implementation clinical practice. Am. J. Med. 130, 1342–1344 (2017).
consortium guideline (CPIC) for CYP2D6 and CYP2C19 genotypes 69. Santos, M. et al. Novel copy-number variations in pharmacogenes
and dosing of tricyclic antidepressants: 2016 update. Clin. contribute to interindividual differences in drug pharmacokinetics.
Pharmacol. Ther. 102, 37–44 (2017). Genet. Med. 20, 622–629 (2018).
49. Hicks, J.K. et al. Clinical pharmacogenetics implementation 70. Botton, M.R. et al. Structural variation at the CYP2C locus:
consortium (CPIC) guideline for CYP2D6 and CYP2C19 genotypes characterization of deletion and duplication alleles. Hum. Mutat.
and dosing of selective serotonin reuptake inhibitors. Clin. 40, e37–e51 (2019).
Pharmacol. Ther. 98, 127–134 (2015). 71. Huddart, R. et al. Standardized biogeographic grouping system
50. Scott, S.A. et al. Clinical pharmacogenetics implementation for annotating populations in pharmacogenetic research. Clin.
consortium guidelines for CYP2C19 genotype and clopidogrel Pharmacol. Ther. 105, 1256–1262 (2019).
therapy: 2013 update. Clin. Pharmacol. Ther. 94, 317–323 72. Cespedes-Garro, C. et al. Relevance of the ancestry for the
(2013). variability of the drug-metabolizing enzymes CYP2C9, CYP2C19
51. Moriyama, B. et al. Clinical pharmacogenetics implementation and CYP2D6 polymorphisms in a multiethnic Costa Rican
consortium (CPIC) guidelines for CYP2C19 and voriconazole population. Rev. Biol. Trop. 64, 1067–1076 (2016).
therapy. Clin. Pharmacol. Ther. 102, 45–51 (2017). 73. Zhou, Y., Ingelman-Sundberg, M. & Lauschke, V.M. Worldwide
52. Caudle, K.E. et al. Standardizing terms for clinical distribution of cytochrome P450 alleles: a meta-analysis of
pharmacogenetic test results: consensus terms from the Clinical population-scale sequencing projects. Clin. Pharmacol. Ther. 102,
Pharmacogenetics Implementation Consortium (CPIC). Genet. 688–700 (2017).
Med. 19, 215–223 (2017). 74. Roco, A. et al. Frequencies of 23 functionally significant variant
53. Caudle, K.E. et al. Standardizing CYP2D6 genotype to phenotype alleles related with metabolism of antineoplastic drugs in the
translation: consensus recommendations from the clinical Chilean population: comparison with Caucasian and Asian
pharmacogenetics implementation consortium and Dutch populations. Front. Genet. 3, 229 (2012).
Pharmacogenetics Working Group. Clin. Transl. Sci. 13, 116–124 75. Henderson, L.M. et al. P450 pharmacogenetics in indigenous
(2020). North American populations. J. Pers. Med. 8, 9
54. Xu, R.A., Gu, E.M., Liu, T.H., Ou-Yang, Q.G., Hu, G.X. & Cai, J.P. (2018).
The effects of cytochrome P450 2C19 polymorphism on the 76. Salazar-Flores, J. et al. Distribution of CYP2D6 and CYP2C19
metabolism of voriconazole in vitro. Infect. Drug Resist. 11, polymorphisms associated with poor metabolizer phenotype in
2129–2135 (2018). five Amerindian groups and western Mestizos from Mexico. Genet.
55. Wang, H. et al. Evaluation of the effects of 20 nonsynonymous Test Mol. Biomarkers 16, 1098–1104 (2012).
single nucleotide polymorphisms of CYP2C19 on S-mephenytoin 77. Zhang, F. & Finkelstein, J. Inconsistency in race and ethnic
4’-hydroxylation and omeprazole 5’-hydroxylation. Drug Metab. classification in pharmacogenetics studies and its potential
Dispos. 39, 830–837 (2011). clinical implications. Pharmgenomics Pers. Med. 12, 107–123
56. Takahashi, M. et al. Functional characterization of 21 CYP2C19 (2019).
allelic variants for clopidogrel 2-oxidation. Pharmacogenomics J. 78. PharmVar. Allele designation criteria and evidence levels <https://
15, 26–32 (2015). www.pharm​var.org/criteria>. Accessed April 6, 2020.

CLINICAL PHARMACOLOGY & THERAPEUTICS | VOLUME 109 NUMBER 2 | February 2021 365
 15326535, 2021, 2, Downloaded from https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.1973 by Fernanda Rodrigues Soares - UFTM - Universidade Federal do Triangulo Mineiro , Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
REVIEW

79. PharmVar CYP2C19 Gene Expert Panel roster <https://www. 89. Astrolabe Tool for allele calling <https://www.child​rensm​ercy.org/
pharm​var.org/exper​t-panels>. Accessed April 6, 2020. genom​esoft​warep​ortal/>. Accessed April 6, 2020.
80. NCBI Reference Sequences database <https://www.ncbi.nlm.nih. 90. Twist, G.P. et al. Constellation: a tool for rapid, automated
gov/refse​q />. Accessed April 6, 2020. phenotype assignment of a highly polymorphic pharmacogene,
81. Locus Reference Genomic (LRG) project <https://www.lrg-seque​ CYP2D6, from whole-genome sequences. NPJ Genom. Med. 1,
nce.org/>. Accessed April 6, 2020. 15007 (2016).
82. PharmVar Standards <https://www.pharm​var.org/genes>. 91. Regan, J.F. et al. A rapid molecular approach for chromosomal
Accessed April 6, 2020. phasing. PLoS One 10, e0118270 (2015).
83. PharmVar CYP2C19 Change Log <https://a.story​blok. 92. Boone, E.C. et al. Long-distance phasing of a tentative ‘enhancer’
com/f/70677/​x /01a6f​89a17/​chang​e -log_cyp2c​19_v1-9.pdf>. SNP with CYP2D6 star allele definitions. Front. Pharm. 11, 486
Accessed April 6, 2020. (2020).
84. Drogemoller, B.I. et al. Characterization of the genetic profile of 93. PharmVar Read Me for CYP2C19 <https://a.story​blok.
CYP2C19 in two South African populations. Pharmacogenomics com/f/70677/​x /74372​b2381/​gene-info_cyp2c​19_v2-2.pdf>.
11, 1095–1103 (2010). Accessed April 6, 2020.
85. Novkovic, M., Matic, D., Kusic-Tisma, J., Antonijevic, N., 94. PharmVar Structural Variation for CYP2C19 <https://a.story​blok.
Radojkovic, D. & Rakicevic, L. Analysis of the CYP2C19 genotype com/f/70677/​x /9225f​a2a03/​varia​tion_cyp2c​19.pdf>. Accessed
associated with bleeding in Serbian STEMI patients who have April 6, 2020.
undergone primary PCI and treatment with clopidogrel. Eur. J. Clin. 95. PharmVar P450 Nomenclature site – Archive <https://www.pharm​
Pharmacol. 74, 443–451 (2018). var.org/htdoc​s/archi​ve/index_origi​nal.htm>. Accessed April 6,
86. Gaedigk, A. et al. The evolution of PharmVar. Clin. Pharmacol. Ther. 2020.
105, 29–32 (2019). 96. CPIC Guidelines <https://cpicp​g x.org/guide​lines/>. Accessed
87. Numanagic, I. et al. Allelic decomposition and exact genotyping of April 6, 2020.
highly polymorphic and structurally variant genes. Nat. Commun. 97. 10xGenomics Linked-reads Genomics <https://www.10xge​
9, 828 (2018). nomics.com/linke​d -reads/>. Accessed April 6,
88. Lee, S.B. et al. Stargazer: a software tool for calling star alleles 2020.
from next-generation sequencing data using CYP2D6 as a model. 98. HGVS Sequence Variant Nomenclature <https://varno​men.hgvs.
Genet. Med. 21, 361–372 (2019). org/>. Accessed April 6, 2020.

366 VOLUME 109 NUMBER 2 | February 2021 | www.cpt-journal.com