Ex) Article Title, Author, Keywords
Ex) Article Title, Author, Keywords
R Clin Pharm 2023; 1(2): 144-154
Published online December 31, 2023 https://doi.org/10.59931/rcp.23.0003
Copyright © Asian Conference On Clinical Pharmacy.
Hyun Kyung Lee1* 
, Ha Young Jang1,2*, Yu Hyun Lee1, Nayoung Han1,3, In-Wha Kim1 , Jung Mi Oh1
Correspondence to:Jung Mi Oh
E-mail jmoh@snu.ac.kr
ORCID
https://orcid.org/0000-0002-1836-1707
* These authors contributed equally to this study.
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/bync/4.0/), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background: Following the rapid growth in genetic information related to drug responses, the urgent need to organize this information for clinical use has arisen. The current databases used in Korea lack information about drug responses as they pertain to specific ethnic groups, including Koreans, and the frequency of genomic variants. This study aimed to develop a pharmacogenetics–alternative allele frequency (AAF) database (PharmGAF DB) in South Korea to address this issue.
Methods: Drugs were selected from the drug response database of the Korean Ministry of Food and Drug Safety by studying various pharmacogenetic guidelines such as Clinical Pharmacogenetics Implementation Consortium, Dutch Pharmacogenetics Working Group, and Canadian Pharmacogenomics Network for Drug Safety and drug labels accepted by various health authorities such as the United States Food and Drug Administration, European Medicines Agency, and the Pharmaceuticals and Medical Devices Agency. Information on drug responses was collected and updated for a set of selected drugs from pharmacogenetics literature over the past ten years. AAF information was gathered from the Korean Reference Genome Database, the single nucleotide polymorphism database, and the Allele Frequency Net Database.
Results: In total, 80 drugs were investigated, and the pharmacogenomic effects of 142 variants in relation to these drugs were updated. Pharmacokinetic effects were found for 51 variants and pharmacodynamic effects were found for 111 variants. AAF information was collected for these variants in Korean, East Asian, and Caucasian populations. Variants for which AAF significantly differed between Koreans and Caucasians were identified. Finally, PharmGAF DB was created by combining information on the pharmacogenomic effects and the AAF.
Conclusion: The newly developed database, PharmGAF DB, was created by merging information on how drug responses relate to genotypes and allele frequencies among different ethnic groups. PharmGAF DB should improve drug efficacy and reduce the occurrence of side effects by supporting the implementation of precision medicine in clinical practice.
KeywordsPharmacogenetics; Pharmacogenomic variants; Drug effects; Allele frequency
Genetic variations play a role in determining an individual’s drug response, such as the efficacy and toxicity of the drug [1]. Advances in genomic analysis technology have made it easier and faster to generate genome data. This has resulted in a growing amount of genetic information related to drug responses, leading to the need for a system to organize this information for clinical use. One example of such a system is the Pharmacogenomics Knowledgebase (PharmGKB), which compiles genetic information from various sources and provides refined data on variant annotations, guideline annotations, and drug label annotations [2]. However, PharmGKB only includes information on allele or genotype frequencies based on studies in the literature. Because there is no alternative allele frequency (AAF) information in other countries, it is challenging to directly infer the drug responses in Koreans and compare these with those of other ethnic groups. Such information is scattered across databases such as the National Center for Biotechnology Information National Library of Medicine dbSNP [3] and the Korean Reference Genome Database (KRGDB) [4]. At present, databases do not contain an overview of drug responses based on the genomes and the frequencies of genomic variants across ethnic groups, including Koreans. The significance of the frequency of alleles or genotypes in pharmacogenomics is paramount, as the patterns of distribution and the expression levels of biomarkers vary among ethnic groups [5,6] and this information is crucial for the effective implementation of pharmacogenomic information in clinical settings [7].
Korean-specific genetic information, including the pharmacogenetic effects of variants and their allele frequencies, is important for precision medicine in Korea. In recognition of this, the Ministry of Food and Drug Safety (MFDS) of South Korea has developed a drug response database based on race/ethnicity [8]. However, this database has become outdated and has not been updated to reflect the latest research, guidelines, and differences in drug responses due to genetic polymorphisms, as well as the allele frequencies of variants in Koreans. Thus, the goal of this study is to establish a pharmacogenetic-alternative allele frequency database (PharmGAF DB) in South Korea by investigating and analyzing the drug responses found in the pharmacogenomic literature from the last ten years (2010 to 2019) as well as the variant allele frequencies in Koreans.
The drug candidates for this study were selected from three different sources: 1) the drug response database in the Korean Ministry of Food and Drug Safety (MFDS); 2) clinical recommendations from pharmacogenetic guidelines (Clinical Pharmacogenetics Implementation Consortium [CPIC], Dutch Pharmacogenetics Working Group [DPWG], and Canadian Pharmacogenomics Network for Drug Safety [CPNDS]); and 3) genetic information listed on the drug labels from the U.S. Food and Drug Administration (FDA), European Medicines Agency (EMA), and Pharmaceuticals and Medical Devices Agency (PMDA). Information from these sources was collected using the PharmGKB database [9]. Duplicate drugs, drugs not used in South Korea, topical or inhalant drugs, and drugs with genetic information only related to diagnosis and indication were excluded, as well as drugs with low evidence levels (Level of Evidence [LOE] 3–4) unless they were included in clinical recommendations from the guidelines. LOE information was sourced from PharmGKB.
A literature review was conducted, surveying studies over a 10-year period to gather information on the selected drugs. The literature from 2010 to 2019 was reviewed, as the drug response database in the MFDS had been constructed based on research up to the year 2010. Two researchers collected the information, and if there were differences among them, a third researcher confirmed the information. Detail information about the studies of clinical annotations in PharmGKB was screened between February of 2020 and November of 2020. In vitro and animal studies, studies of drug-drug interactions, studies focusing on specific populations (e.g., infants or pregnant women), studies that did not provide additional information, and case reports with fewer than five subjects were excluded. Up to five studies for each drug were selected based on the following criteria: ethnicity (prioritization according to similarity with Koreans - Koreans as the first priority, followed by Han Chinese, Japanese, and Caucasians), the type of study (meta-analysis study being the first priority, followed by case/control study, and cohort study), and the publication year (with recent studies being assigned higher priority).
The purpose of this study was to gather information on how the pharmacokinetics (PK) and pharmacodynamics (PD) of drugs are affected by genotypes. The selected studies were reviewed and information such as genotypes, ethnic groups, the number of participants, doses, Cmax, T1/2, and clearance (CL) of drugs for each gene were documented in the sections on ‘PK differences by genotypes’ and ‘PD differences by genotypes’. The final racial/ethnic drug response database was verified and revised through an internal review by multiple researchers.
The rs numbers of genotypes that affect drug responses were investigated and the alternative allele frequencies for those genotypes were then collected. The rs numbers of the affecting genotypes were collected using a summary of very important pharmacogenes (VIPs) from PharmGKB. If the rs number was not found in the VIP summary, it was identified using SNPedia [10], a database of single nucleotide polymorphisms (SNPs). The alternative allele frequency (AAF) was collected by examining the variant alleles of Koreans, East Asians, and Caucasians using the rs numbers from the Korean Reference Genome Database (KRGDB) [4]. If the AAF of Koreans was not found in KRGDB, in order to infer the drug response in Koreans, the AAFs of other countries were investigated instead of Koreans in the order of Japanese, Chinese, and other Asians based on the degree of ethnic similarity to Koreans using the dbSNP from the National Center for Biotechnology Information (NCBI) [3]. The frequencies of Human Leukocyte Antigen (HLA) genotypes in Koreans were collected using the Allele Frequency Net Database (AFND) [11], and the data from the study with the largest sample size were selected when multiple data sources were available. Variant frequencies not collected from databases were obtained through manual literature searches. The final race/ethnic drug response database was double-checked and verified through a peer review by internal researchers, and the collected rs numbers and allele frequencies of Koreans, Asians, and Caucasians were summarized in the section ‘allele frequency by ethnicity’.
A total of 192 drugs in the existing racial/ethnic drug response database; 81 drugs from the CPIC, DPWG and CPNDS guidelines; and 111 drugs from FDA, EMA, and PMDA approvals for genetic testing recommendation information were collected (Fig. 1). After removing duplicates, 311 candidate drugs were selected. Among these, drugs not approved or marketed in South Korea (n=89), topical or inhaled medications (n=7), and drugs with only diagnosis or indication-related genetic information (n=44) were excluded, leaving 171 eligible drugs (with LOE: ≤2 [n=80], >2 [n=91]). The final list of 80 drugs was selected for analysis. Table 1 displays a list of clinical annotations with LOE 1, with the full list provided in the Supplementary Table 1.
Table 1 Summary table of PharmGAF DB (key excerpt [level of evidence: 1A])
| Drug | Gene | Variant | PK | PD |
|---|---|---|---|---|
| Abacavir | *57:01:01 | Toxicity | ||
| Allopurinol | *58:01 | Toxicity | ||
| Amitriptyline | *1, *2, *3, *17 | Metabolism | ||
| *1/*1xN, *2/*2xN | Metabolism | Toxicity | ||
| Atazanavir | rs887829 | Toxicity | ||
| Atomoxetine | *1, *2, *3, *4, *5, *10 | Metabolism | Toxicity | |
| Azathioprine | *1, *2, *3A, *3B, *3C | Dosage/toxicity | ||
| rs116855232 | Toxicity | |||
| Capecitabine | rs75017182 | Toxicity | ||
| rs55886062 | Toxicity | |||
| rs67376798 | Toxicity | |||
| rs3918290 | Toxicity | |||
| Carbamazepine | *31:01:02 | Toxicity | ||
| *15:02:01 | Toxicity | |||
| Celecoxib | *1, *2, *3, *13 | Metabolism | ||
| Citalopram | *1, *2, *3, *4, *17 | Metabolism | Toxicity | |
| Clomipramine | *1, *1xN, *2, *2xN, *3, *4, *5, *6 | Toxicity | ||
| *1, *2, *3 | Metabolism | |||
| Clopidogrel | *1, *2, *3, *4, *5, *6, *8 | Efficacy/toxicity | ||
| rs4986893 636G>A *3 | Efficacy/toxicity | |||
| rs4244285 681G>A *2 | Efficacy/toxicity | |||
| rs12248560 -806C>T *17 | Efficacy/toxicity | |||
| Codeine | *1, *1xN, *2, *2xN, *3, *4, *5, *6, *10, *17, *40, *41 | Metabolism | Efficacy/toxicity | |
| Dexlansoprazole | *1, *2, *3 | Metabolism | ||
| Doxepin | *1, *1xN, *2, *2xN, *3, *4, *5 | Metabolism | ||
| *1, *2 | Metabolism | |||
| Efavirenz | rs3745274 516G>T | Metabolism | ||
| *1, *6, *26 516G>T | Dosage/toxicity | |||
| *1, *4, *6, *9, *16, *18, *28 | Metabolism | |||
| Escitalopram | *1, *2, *3, *4, *17 | Metabolism | ||
| Fluorouracil | rs3918290 | Toxicity | ||
| rs75017182 | Toxicity | |||
| rs55886062 | Toxicity | |||
| rs67376798 | Toxicity | |||
| Fluvoxamine | *1, *3, *4, *5, *6, *10 | Metabolism | ||
| Ibuprofen | *1, *2, *3 | Metabolism | ||
| Imipramine | *1, *2, *3, *17 | Metabolism | ||
| *1, *3, *4, *5 | Metabolism | |||
| Lansoprazole | *1, *2, *3, *8, *9, *17 | Efficacy | ||
| Lornoxicam | *1, *3, *13 | Metabolism | ||
| Meloxicam | *1, *2, *3, *13 | Metabolism | ||
| Mercaptopurine | rs116855232 | Metabolism | Toxicity | |
| *1, *2, *3A, *3B, *3C, *4 | Dosage/toxicity | |||
| Nortriptyline | *1, *2, *2xN, *3, *4, *5, *6, *10 | Metabolism | ||
| Omeprazole | *1, *2, *3, *9, *10, *17, *24, *26 | Metabolism | Efficacy | |
| Ondansetron | *1, *1xN | Metabolism | ||
| Oxcarbazepine | *15:02:01 | Toxicity | ||
| Pantoprazole | *1, *2, *3, *17 | Metabolism | Efficacy | |
| Paroxetine | *1, *1xN, *2, *2xN, *3, *4, *5, *6, *10 | Efficacy | ||
| Peginterferon alfa-2a | rs12979860 | Efficacy | ||
| Phenytoin | *1, *2, *3 | Metabolism | Toxicity | |
| rs1057910 | Toxicity | |||
| 15:02:01 | Toxicity | |||
| Piroxicam | *1, *2, *3 | Metabolism | ||
| Rasburicase | A-202A_376 | Toxicity | ||
| Ribavirin | rs12979860 | Efficacy | ||
| rs8099917 | Efficacy | |||
| Sertraline | CYP2D6 | *1,*3,*4,*5,*10,*17,*41 | Efficacy | |
| Simvastatin | rs4149056 | Efficacy/toxicity | ||
| Tacrolimus | *1, *3, *6, *7 | Efficacy/dosage | ||
| Tamoxifen | *1, *3, *4, *5, *6, *10, *41 | Efficacy | ||
| Voriconazole | *1, *2, *3, *17 | Metabolism | Efficacy/toxicity | |
| Warfarin | *1, *2, *3, *5, *6, *11 | Toxicity | ||
| *1, *3 | Dosage | |||
| rs9923231 | Dosage |
CYP=cytochrome P450, DPYD=dihydropyrimidine dehydrogenase, G6PD=glucose-6-phosphate dehydrogenase, HLA=human leukocyte antigen, IFNL=interferon lambda, NUDT15=nudix hydrolase 15, SLCO1B1=solute carrier organic anion transporter family member 1B1, TPMT=thiopurine S methyltransferase, UGT1A1=UDP-glucuronosyltransferase 1A1, VKORC1=vitamin k epoxide reductase complex subunit 1.
Figure 1. Drug selection process for updating drug response database by race/ethnicity.A total of 258 papers were reviewed. The authors collected data on 142 variants and their pharmacogenomic effects on 80 drugs, with 51 variants affecting PK and 111 affecting PD. Of the 111 PD variants, 68 were related to drug efficacy, 51 to toxicity, and 15 to dosing. The top ten variants with the highest number of affecting drugs were as follows: cytochrome P450 enzymes (
Table 2 Alternative allele frequency for cytochrome P450 (CYP) enzyme
| Gene | Allele | rs number | Alternative allele frequency | Reference | ||
|---|---|---|---|---|---|---|
| Korean | East Asian | Caucasian | ||||
| *2 (2850C>T) | rs16947 | 0.17 | 0.15 | 0.34 | [4,20] | |
| *4 (1846 C>T) | rs3892097 | 0.0016 | 0.0017 | 0.19 | ||
| rs1800716 | ||||||
| *10 (100C>T) | rs1065852 | 0.51 | 0.52 | 0.2 | ||
| rs1081003 | ||||||
| *17 (1023C>T) | rs28371706 | 0.49 | 0.52 | 0.02 | ||
| rs1081003 | ||||||
| *41† | rs28371725 | 0.027 | 0.03 | 0.09 | ||
| A>C | rs4646 | 0.69 | 0.72 | 0.71 | [4] | |
| g.-48T>G | rs28399433 | 0.25 | 0.27 | 0.08 | [4,21] | |
| 6600G>T | rs28399468 | 0.025 | 0.04 | N/A | ||
| g.6558T>C | rs5031016 | 0.11 | 0.1 | 0.0013 | ||
| 1839G>T | rs8192726 | 0.14 | 0.19 | 0.08 | ||
| 516G>T | rs3745274 | 0.13 | 0.18 | 0.23 | [4,22] | |
| A>G | rs2279343 | 0.084 | N/A | N/A | ||
| *2 (430C>T) | rs1799853 | 0.0080 | 0.0028 | 0.0013 | [4] | |
| rs7900194 | ||||||
| *3 (1075A>C) | rs1057910 | 0.042 | 0.04 | 0.06 | ||
| *13 (269T>C) | rs72558187 | 0.0040 | 0.0017 | N/A | ||
| *2 (681G>A) | rs4244285 | 0.26 | 0.33 | 0.15 | [4,23] | |
| *3 (636G>A) | rs4986893 | 0.093 | 0.05 | N/A | ||
| *17 (-806C>T) | rs12248560 | 0.012 | 0.02 | 0.23 | ||
| *1B (-392A>G) | rs2740574 | 0.0016 | 0 | 0.03 | [4] | |
| *1G (20230G>A) | rs2242480 | 0.21 | 0.25 | 0.07 | ||
| *3 (6986A>G) | rs776746 | 0.22 | 0.28 | 0.05 | [4,24] | |
| *6 (14690G>A) | rs10264272 | 0.0016 | 0.0017 | N/A | ||
| rs56411402 | ||||||
| *3 (C>T) | rs2108622 | 0.32 | 0.21 | 0.27 | [4] | |
†2988G>A or 2989G>A on NG_008376.3.
Table 3 Alternative allele frequency for gene types other than CYP450 genes
| Gene | Allele | rs number | Alternative allele frequency | Reference | ||
|---|---|---|---|---|---|---|
| Korean | East Asian | Caucasian | ||||
| *57:01 (T>G) (HCP5) | rs2395029 | 0.0032 | 0.01 | 0.05 | [4,25,26] | |
| *15:02:01 | N/A | 0.0033 | N/A | N/A | [27] | |
| *13:01:01 | N/A | 0.021 | N/A | N/A | [27] | |
| *58:01 | rs9263726 | 0.0096 | 0.05 | 0.13 | [4,28] | |
| *31:01:02 | rs1061235 | 0.054 | N/A | N/A | [27] | |
| *33:03 | N/A | 0.16 | N/A | N/A | [27] | |
| *03:02 | rs2523447 | 0.11 | N/A | N/A | [27] | |
| *04:01:01:01 | N/A | 0 | 0.01 | N/A | [27] | |
| *02:01 | N/A | 0.073 | N/A | N/A | [27] | |
| *01:01:01 | N/A | 0.065 | N/A | N/A | [27] | |
| 3435T>C | rs1045642 | 0.65 | 0.6 | 0.47 | [4,29] | |
| 2677T>G/A | rs2032582 | 0.62 | 0.55 | 0.57 | ||
| 1236C>T | rs1128503 | 0.38 | 0.34 | 0.57 | ||
| 421C>A | rs2231142 | 0.27 | 0.29 | 0.1 | [4,30] | |
| G>A | rs1872328 | 0.013 | 0.01 | 0.04 | [4] | |
| C>T | rs7412 | 0.051 | 0.08 | 0.07 | [4] | |
| T>C | rs4673993 | 0.19 | 0.21 | 0.31 | [4] | |
| A>G | rs4680 | 0.28 | 0.29 | 0.52 | [4] | |
| *2A (1905+1G>A) | rs3918290 | 0.10 | 0.12 | 0.04 | [4,23] | |
| rs17376848 | ||||||
| 85T>C | rs3918290 | 0 | N/A | N/A | ||
| A>G | rs78060119 | 0.016 | 0.0035 | 0.13 | ||
| rs56293913 | ||||||
| T>A, G | rs121434568 | N/A | N/A | N/A | N/A | |
| A>G | rs11615 | 0.73 | 0.72 | 0.4 | [4] | |
| C>A | rs3212986 | 0.24 | 0.28 | 0.27 | ||
| 313A>G | rs1695 | 0.18 | 0.17 | 0.32 | [4] | |
| G>A | rs1050828 | N/A | N/A | N/A | N/A | |
| C>G | rs6295 | 0.75 | 0.78 | 0.44 | [4] | |
| C>G | rs1414334 | 0.98 | 0.99 | 0.85 | [4] | |
| T>G | rs8099917 | 0.047 | 0.07 | 0.17 | [4] | |
| A>G | rs11881222 | 0.052 | 0.07 | 0.3 | ||
| C>T | rs12979860 | 0.051 | 0.08 | 0.32 | ||
| C>T | rs12979860 | 0.051 | 0.08 | 0.32 | [4] | |
| C>G | rs368234815 | 0.038 | 0.06 | 0.17 | ||
| rs4803221 | ||||||
| T>C | rs5219 | 0.60 | 0.62 | 0.66 | [4] | |
| C>A | rs489693 | 0.25 | 0.24 | 0.31 | [4] | |
| G>A | rs1801133 | 0.42 | 0.37 | 0.35 | [4] | |
| *5 (341T>C) | rs1801280 | 0.026 | 0.03 | 0.45 | [4,31] | |
| *6A (590G>A) | rs1799930 | 0.18 | 0.23 | 0.28 | ||
| *7 (857G>A) | rs1799931 | 0.13 | 0.16 | 0.02 | ||
| *7B (282C>T) | rs1041983 | 0.32 | 0.39 | 0.3 | ||
| *14 (191G>A) | rs1801279 | 0.0024 | 0.0061 | 0.0031 | ||
| rs1805158 | ||||||
| p.R139C (C>T) | rs116855232 | 0.12 | 0.12 | 0.004 | [4] | |
| G>A | rs10306114 | 0.0016 | 0.01 | 0.0041 | [4] | |
| rs142017527 | ||||||
| T>C | rs1051266 | 0.43 | 0.47 | 0.56 | [4] | |
| *1B (492A>G or 388A>G) | rs2306283 | 0.72 | 0.75 | 0.4 | [4,32] | |
| *5 (625T>C or 521T>C) | rs4149056 | 0.14 | 0.13 | 0.17 | ||
| *3B (460G>A) | rs1800460 | 0.73 | 0.75 | 0.78 | [4,33] | |
| rs2842934 | ||||||
| *3C (719A>G) | rs1142345 | 0.016 | 0.02 | 0.03 | ||
| *80 (364C>T) | rs887829 | 0.13 | 0.13 | 0.3 | [4,34] | |
| *6 (71G>A) | rs4148323 | 0.18 | 0.17 | 0.01 | ||
| 3673C>T or -1639C>T | rs9923231 | 0.92 | 0.92 | 0.4 | [4,35] | |
| 9041C>T or 3730C>T | rs7294 | 0.074 | 0.08 | 0.35 | ||
| 6484C>T or 1173C>T | rs9934438 | 0.92 | 0.92 | 0.4 | ||
ABCB1=ATP binding cassette subfamily B, ABCG2=ATP-binding cassette super-family G member 2, ACYP2=acylphosphatase 2, APOE=apolipoprotein E, ATIC=5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase/IMP cyclohydrolase, COMT=catechol-O-methyltransferase, DPYD=dihydropyrimidine dehydrogenase, EGFR=epidermal growth factor receptor, ERCC1=excision repair cross-complementing group, G6PD=glucose-6-phosphate dehydrogenase, GSTP1=glutathione S-transferase Pi, HLA=human leukocyte antigen, HTR1A=5-hydroxytryptamine receptor 1A, HTR2C=5-hydroxytryptamine receptor 2C, IFNL=interferon lambda 3, KCNJ11=potassium inwardly rectifying channel subfamily J, MC4R=melanocortin-4 receptor, MTHFR=methylenetetrahydrofolate reductase, NAT2=N-acetyltransferase 2, NUDT15=nudix hydrolase 15, PTGS1=prostaglandin-endoperoxide synthase 1, SLC19A1=solute carrier family 19, SLCO1B1=solute carrier organic anion transporter family member 1B1, TPMT=thiopurine S methyltransferase, UGT1A1=UDP-glucuronosyltransferase 1A1, VKORC1=vitamin K epoxide reductase complex subunit 1.
The PharmGAF DB was constructed by combining the results of the pharmacogenomic literature analysis and the AAF information. The PharmGAF DB consists of 80 drugs, each of which displays drug responses and AAFs based on genetic variations. The PharmGAF DB was developed as some parts of the MFDS drug response DB by race/ethnicity and the full DB is shown in MFDS website [8]. The full DB includes information about related genes, allele frequencies by ethnicity, PK differences by genotype, PD differences by genotype, clinical applications, and drug labels.
In this study, a reliable database regarding drug responses related to genetic variants was compiled. Drugs included in regulatory affairs (MFDS, FDA, EMDA and PMDA) and pharmacogenetic guidelines were listed, and the corresponding PK/PD responses related to genetic variants were collected by painstaking manual curation. Furthermore, it is well known that the frequency of PK/PD responses can vary depending on the gene variant frequency. Therefore, we also collected information on AAFs by ethnicity for all gene variants in the literature surveyed here and thereby link these data to the genome-drug response DB for Koreans. It is expected that our PharmGAF DB will be utilized as a reference by those conducting studies related to drug effectiveness and safety. Given that the PharmGAF DB contains both the frequency of variability and the risk (e.g., odds ratio), it can also be used to design clinical trials to test the clinical and economic effects of genetic testing while reducing both the time and cost.
Many studies have constructed genetic variance-drug response DBs. For instance, Ubiquitous Pharmacogenomics (U-PGx) aims to increase the quality of life and reduce the costs borne by patients by constructing a DB containing 50 variants of 13 genes and the corresponding drug responses [12]. The ClinGen PGx Working Group curated 1,750 gene-related data instances based on 11,413 experts [13]. The PharmacoGenomic Mutation Database (PGMD) released a version including over 117,000 unique pharmacogenomic observations, covering all 24 disease superclasses and nearly 1,400 drugs [14]. PharmVar has established a repository of pharmacogenomics variations for
The PharmGAF DB can have clinical implications. For example, azathioprine induced leukopenia is associated with rs116855232 [18]. The allele frequency of this variant is higher in Koreans and East Asians than in Caucasians, with frequencies of 0.12, 0.12, and 0.004, respectively [4]. This suggests that there should be greater attention to azathioprine induced leukopenia in East Asians compared to Caucasians. In fact, a higher incidence of leukopenia has been reported in Asians compared to the Caucasian population, aligning with the frequency trend of rs116855232 [18]. The differences in allele frequencies among ethnicities presented in this database imply that the frequency of related PK or PD phenotypes may also vary. Therefore, this information can provide insights into which phenotypes need more attention in each ethnic group.
The advantage of this study is that AAF information is combined with genome-drug response information. A researcher would be able to know at a glance how frequently a genomic variant will affect the drug response in a specific race. Risk management would also be possible by multiplying the risk (e.g., odds ratio) and frequency (e.g. AAF) (modified from the risk priority number formula suggested by Lee et al. [19]), thereby preventing serious adverse events from happening.
However, there are several limitations in our study. First, although drug reactions and AAF information for various races were collected, drugs used in other countries were not included in the DB because the collected drugs were limited to drugs marketed in Korea. Second, only high-grade genomic information suggested by PharmGKB was collected. Even if the grade is low, there may be drug-variant annotations with high variant frequencies. However, as combination drugs were not included in our study design, caution should be exercised when using the PharmGAF DB. Future studies are needed in which the AAFs for all drug-variant pairs against all ethnicities are presented. This would provide a fuller set of integrated information.
The PharmGAF DB was developed by merging information about drug responses related to genotypes and allele frequencies among different ethnic groups. The PharmGAF DB as presented here could contribute to maximizing drug effects and minimizing side effects by defining customized treatments applicable to clinical practice. We hope this freely available DB will be widely used for a wide range of clinical purposes.
Not applicable.
This research was supported by a grant (20182MFDS444) from the Ministry of Food and Drug Safety in 2020.
We would like to thank Yoonjae Hwang, Woo Bin Lee, Won-gu Kang, Geon-yeon Kim, Minseo Kang, Dohye Kim, Jonghyun Sung, Sung-eun Jo, Yejin Shin for their contributions to data collection.
Supplementary materials can be found via https://doi.org/10.59931/rcp.23.0003.
rcp-1-2-144-supple.pdf
R Clin Pharm 2023; 1(2): 144-154
Published online December 31, 2023 https://doi.org/10.59931/rcp.23.0003
Copyright © Asian Conference On Clinical Pharmacy.
Hyun Kyung Lee1* , Ha Young Jang1,2*, Yu Hyun Lee1, Nayoung Han1,3, In-Wha Kim1 , Jung Mi Oh1
1College of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Seoul, Korea
2College of Pharmacy, Gachon University, Incheon, Korea
3College of Pharmacy, Jeju National University, Jeju, Korea
Correspondence to:Jung Mi Oh
E-mail jmoh@snu.ac.kr
ORCID
https://orcid.org/0000-0002-1836-1707
* These authors contributed equally to this study.
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/bync/4.0/), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background: Following the rapid growth in genetic information related to drug responses, the urgent need to organize this information for clinical use has arisen. The current databases used in Korea lack information about drug responses as they pertain to specific ethnic groups, including Koreans, and the frequency of genomic variants. This study aimed to develop a pharmacogenetics–alternative allele frequency (AAF) database (PharmGAF DB) in South Korea to address this issue.
Methods: Drugs were selected from the drug response database of the Korean Ministry of Food and Drug Safety by studying various pharmacogenetic guidelines such as Clinical Pharmacogenetics Implementation Consortium, Dutch Pharmacogenetics Working Group, and Canadian Pharmacogenomics Network for Drug Safety and drug labels accepted by various health authorities such as the United States Food and Drug Administration, European Medicines Agency, and the Pharmaceuticals and Medical Devices Agency. Information on drug responses was collected and updated for a set of selected drugs from pharmacogenetics literature over the past ten years. AAF information was gathered from the Korean Reference Genome Database, the single nucleotide polymorphism database, and the Allele Frequency Net Database.
Results: In total, 80 drugs were investigated, and the pharmacogenomic effects of 142 variants in relation to these drugs were updated. Pharmacokinetic effects were found for 51 variants and pharmacodynamic effects were found for 111 variants. AAF information was collected for these variants in Korean, East Asian, and Caucasian populations. Variants for which AAF significantly differed between Koreans and Caucasians were identified. Finally, PharmGAF DB was created by combining information on the pharmacogenomic effects and the AAF.
Conclusion: The newly developed database, PharmGAF DB, was created by merging information on how drug responses relate to genotypes and allele frequencies among different ethnic groups. PharmGAF DB should improve drug efficacy and reduce the occurrence of side effects by supporting the implementation of precision medicine in clinical practice.
Keywords: Pharmacogenetics, Pharmacogenomic variants, Drug effects, Allele frequency
Genetic variations play a role in determining an individual’s drug response, such as the efficacy and toxicity of the drug [1]. Advances in genomic analysis technology have made it easier and faster to generate genome data. This has resulted in a growing amount of genetic information related to drug responses, leading to the need for a system to organize this information for clinical use. One example of such a system is the Pharmacogenomics Knowledgebase (PharmGKB), which compiles genetic information from various sources and provides refined data on variant annotations, guideline annotations, and drug label annotations [2]. However, PharmGKB only includes information on allele or genotype frequencies based on studies in the literature. Because there is no alternative allele frequency (AAF) information in other countries, it is challenging to directly infer the drug responses in Koreans and compare these with those of other ethnic groups. Such information is scattered across databases such as the National Center for Biotechnology Information National Library of Medicine dbSNP [3] and the Korean Reference Genome Database (KRGDB) [4]. At present, databases do not contain an overview of drug responses based on the genomes and the frequencies of genomic variants across ethnic groups, including Koreans. The significance of the frequency of alleles or genotypes in pharmacogenomics is paramount, as the patterns of distribution and the expression levels of biomarkers vary among ethnic groups [5,6] and this information is crucial for the effective implementation of pharmacogenomic information in clinical settings [7].
Korean-specific genetic information, including the pharmacogenetic effects of variants and their allele frequencies, is important for precision medicine in Korea. In recognition of this, the Ministry of Food and Drug Safety (MFDS) of South Korea has developed a drug response database based on race/ethnicity [8]. However, this database has become outdated and has not been updated to reflect the latest research, guidelines, and differences in drug responses due to genetic polymorphisms, as well as the allele frequencies of variants in Koreans. Thus, the goal of this study is to establish a pharmacogenetic-alternative allele frequency database (PharmGAF DB) in South Korea by investigating and analyzing the drug responses found in the pharmacogenomic literature from the last ten years (2010 to 2019) as well as the variant allele frequencies in Koreans.
The drug candidates for this study were selected from three different sources: 1) the drug response database in the Korean Ministry of Food and Drug Safety (MFDS); 2) clinical recommendations from pharmacogenetic guidelines (Clinical Pharmacogenetics Implementation Consortium [CPIC], Dutch Pharmacogenetics Working Group [DPWG], and Canadian Pharmacogenomics Network for Drug Safety [CPNDS]); and 3) genetic information listed on the drug labels from the U.S. Food and Drug Administration (FDA), European Medicines Agency (EMA), and Pharmaceuticals and Medical Devices Agency (PMDA). Information from these sources was collected using the PharmGKB database [9]. Duplicate drugs, drugs not used in South Korea, topical or inhalant drugs, and drugs with genetic information only related to diagnosis and indication were excluded, as well as drugs with low evidence levels (Level of Evidence [LOE] 3–4) unless they were included in clinical recommendations from the guidelines. LOE information was sourced from PharmGKB.
A literature review was conducted, surveying studies over a 10-year period to gather information on the selected drugs. The literature from 2010 to 2019 was reviewed, as the drug response database in the MFDS had been constructed based on research up to the year 2010. Two researchers collected the information, and if there were differences among them, a third researcher confirmed the information. Detail information about the studies of clinical annotations in PharmGKB was screened between February of 2020 and November of 2020. In vitro and animal studies, studies of drug-drug interactions, studies focusing on specific populations (e.g., infants or pregnant women), studies that did not provide additional information, and case reports with fewer than five subjects were excluded. Up to five studies for each drug were selected based on the following criteria: ethnicity (prioritization according to similarity with Koreans - Koreans as the first priority, followed by Han Chinese, Japanese, and Caucasians), the type of study (meta-analysis study being the first priority, followed by case/control study, and cohort study), and the publication year (with recent studies being assigned higher priority).
The purpose of this study was to gather information on how the pharmacokinetics (PK) and pharmacodynamics (PD) of drugs are affected by genotypes. The selected studies were reviewed and information such as genotypes, ethnic groups, the number of participants, doses, Cmax, T1/2, and clearance (CL) of drugs for each gene were documented in the sections on ‘PK differences by genotypes’ and ‘PD differences by genotypes’. The final racial/ethnic drug response database was verified and revised through an internal review by multiple researchers.
The rs numbers of genotypes that affect drug responses were investigated and the alternative allele frequencies for those genotypes were then collected. The rs numbers of the affecting genotypes were collected using a summary of very important pharmacogenes (VIPs) from PharmGKB. If the rs number was not found in the VIP summary, it was identified using SNPedia [10], a database of single nucleotide polymorphisms (SNPs). The alternative allele frequency (AAF) was collected by examining the variant alleles of Koreans, East Asians, and Caucasians using the rs numbers from the Korean Reference Genome Database (KRGDB) [4]. If the AAF of Koreans was not found in KRGDB, in order to infer the drug response in Koreans, the AAFs of other countries were investigated instead of Koreans in the order of Japanese, Chinese, and other Asians based on the degree of ethnic similarity to Koreans using the dbSNP from the National Center for Biotechnology Information (NCBI) [3]. The frequencies of Human Leukocyte Antigen (HLA) genotypes in Koreans were collected using the Allele Frequency Net Database (AFND) [11], and the data from the study with the largest sample size were selected when multiple data sources were available. Variant frequencies not collected from databases were obtained through manual literature searches. The final race/ethnic drug response database was double-checked and verified through a peer review by internal researchers, and the collected rs numbers and allele frequencies of Koreans, Asians, and Caucasians were summarized in the section ‘allele frequency by ethnicity’.
A total of 192 drugs in the existing racial/ethnic drug response database; 81 drugs from the CPIC, DPWG and CPNDS guidelines; and 111 drugs from FDA, EMA, and PMDA approvals for genetic testing recommendation information were collected (Fig. 1). After removing duplicates, 311 candidate drugs were selected. Among these, drugs not approved or marketed in South Korea (n=89), topical or inhaled medications (n=7), and drugs with only diagnosis or indication-related genetic information (n=44) were excluded, leaving 171 eligible drugs (with LOE: ≤2 [n=80], >2 [n=91]). The final list of 80 drugs was selected for analysis. Table 1 displays a list of clinical annotations with LOE 1, with the full list provided in the Supplementary Table 1.
Table 1 . Summary table of PharmGAF DB (key excerpt [level of evidence: 1A]).
| Drug | Gene | Variant | PK | PD |
|---|---|---|---|---|
| Abacavir | *57:01:01 | Toxicity | ||
| Allopurinol | *58:01 | Toxicity | ||
| Amitriptyline | *1, *2, *3, *17 | Metabolism | ||
| *1/*1xN, *2/*2xN | Metabolism | Toxicity | ||
| Atazanavir | rs887829 | Toxicity | ||
| Atomoxetine | *1, *2, *3, *4, *5, *10 | Metabolism | Toxicity | |
| Azathioprine | *1, *2, *3A, *3B, *3C | Dosage/toxicity | ||
| rs116855232 | Toxicity | |||
| Capecitabine | rs75017182 | Toxicity | ||
| rs55886062 | Toxicity | |||
| rs67376798 | Toxicity | |||
| rs3918290 | Toxicity | |||
| Carbamazepine | *31:01:02 | Toxicity | ||
| *15:02:01 | Toxicity | |||
| Celecoxib | *1, *2, *3, *13 | Metabolism | ||
| Citalopram | *1, *2, *3, *4, *17 | Metabolism | Toxicity | |
| Clomipramine | *1, *1xN, *2, *2xN, *3, *4, *5, *6 | Toxicity | ||
| *1, *2, *3 | Metabolism | |||
| Clopidogrel | *1, *2, *3, *4, *5, *6, *8 | Efficacy/toxicity | ||
| rs4986893 636G>A *3 | Efficacy/toxicity | |||
| rs4244285 681G>A *2 | Efficacy/toxicity | |||
| rs12248560 -806C>T *17 | Efficacy/toxicity | |||
| Codeine | *1, *1xN, *2, *2xN, *3, *4, *5, *6, *10, *17, *40, *41 | Metabolism | Efficacy/toxicity | |
| Dexlansoprazole | *1, *2, *3 | Metabolism | ||
| Doxepin | *1, *1xN, *2, *2xN, *3, *4, *5 | Metabolism | ||
| *1, *2 | Metabolism | |||
| Efavirenz | rs3745274 516G>T | Metabolism | ||
| *1, *6, *26 516G>T | Dosage/toxicity | |||
| *1, *4, *6, *9, *16, *18, *28 | Metabolism | |||
| Escitalopram | *1, *2, *3, *4, *17 | Metabolism | ||
| Fluorouracil | rs3918290 | Toxicity | ||
| rs75017182 | Toxicity | |||
| rs55886062 | Toxicity | |||
| rs67376798 | Toxicity | |||
| Fluvoxamine | *1, *3, *4, *5, *6, *10 | Metabolism | ||
| Ibuprofen | *1, *2, *3 | Metabolism | ||
| Imipramine | *1, *2, *3, *17 | Metabolism | ||
| *1, *3, *4, *5 | Metabolism | |||
| Lansoprazole | *1, *2, *3, *8, *9, *17 | Efficacy | ||
| Lornoxicam | *1, *3, *13 | Metabolism | ||
| Meloxicam | *1, *2, *3, *13 | Metabolism | ||
| Mercaptopurine | rs116855232 | Metabolism | Toxicity | |
| *1, *2, *3A, *3B, *3C, *4 | Dosage/toxicity | |||
| Nortriptyline | *1, *2, *2xN, *3, *4, *5, *6, *10 | Metabolism | ||
| Omeprazole | *1, *2, *3, *9, *10, *17, *24, *26 | Metabolism | Efficacy | |
| Ondansetron | *1, *1xN | Metabolism | ||
| Oxcarbazepine | *15:02:01 | Toxicity | ||
| Pantoprazole | *1, *2, *3, *17 | Metabolism | Efficacy | |
| Paroxetine | *1, *1xN, *2, *2xN, *3, *4, *5, *6, *10 | Efficacy | ||
| Peginterferon alfa-2a | rs12979860 | Efficacy | ||
| Phenytoin | *1, *2, *3 | Metabolism | Toxicity | |
| rs1057910 | Toxicity | |||
| 15:02:01 | Toxicity | |||
| Piroxicam | *1, *2, *3 | Metabolism | ||
| Rasburicase | A-202A_376 | Toxicity | ||
| Ribavirin | rs12979860 | Efficacy | ||
| rs8099917 | Efficacy | |||
| Sertraline | CYP2D6 | *1,*3,*4,*5,*10,*17,*41 | Efficacy | |
| Simvastatin | rs4149056 | Efficacy/toxicity | ||
| Tacrolimus | *1, *3, *6, *7 | Efficacy/dosage | ||
| Tamoxifen | *1, *3, *4, *5, *6, *10, *41 | Efficacy | ||
| Voriconazole | *1, *2, *3, *17 | Metabolism | Efficacy/toxicity | |
| Warfarin | *1, *2, *3, *5, *6, *11 | Toxicity | ||
| *1, *3 | Dosage | |||
| rs9923231 | Dosage |
CYP=cytochrome P450, DPYD=dihydropyrimidine dehydrogenase, G6PD=glucose-6-phosphate dehydrogenase, HLA=human leukocyte antigen, IFNL=interferon lambda, NUDT15=nudix hydrolase 15, SLCO1B1=solute carrier organic anion transporter family member 1B1, TPMT=thiopurine S methyltransferase, UGT1A1=UDP-glucuronosyltransferase 1A1, VKORC1=vitamin k epoxide reductase complex subunit 1..
Figure 1. Drug selection process for updating drug response database by race/ethnicity.A total of 258 papers were reviewed. The authors collected data on 142 variants and their pharmacogenomic effects on 80 drugs, with 51 variants affecting PK and 111 affecting PD. Of the 111 PD variants, 68 were related to drug efficacy, 51 to toxicity, and 15 to dosing. The top ten variants with the highest number of affecting drugs were as follows: cytochrome P450 enzymes (
Table 2 . Alternative allele frequency for cytochrome P450 (CYP) enzyme.
| Gene | Allele | rs number | Alternative allele frequency | Reference | ||
|---|---|---|---|---|---|---|
| Korean | East Asian | Caucasian | ||||
| *2 (2850C>T) | rs16947 | 0.17 | 0.15 | 0.34 | [4,20] | |
| *4 (1846 C>T) | rs3892097 | 0.0016 | 0.0017 | 0.19 | ||
| rs1800716 | ||||||
| *10 (100C>T) | rs1065852 | 0.51 | 0.52 | 0.2 | ||
| rs1081003 | ||||||
| *17 (1023C>T) | rs28371706 | 0.49 | 0.52 | 0.02 | ||
| rs1081003 | ||||||
| *41† | rs28371725 | 0.027 | 0.03 | 0.09 | ||
| A>C | rs4646 | 0.69 | 0.72 | 0.71 | [4] | |
| g.-48T>G | rs28399433 | 0.25 | 0.27 | 0.08 | [4,21] | |
| 6600G>T | rs28399468 | 0.025 | 0.04 | N/A | ||
| g.6558T>C | rs5031016 | 0.11 | 0.1 | 0.0013 | ||
| 1839G>T | rs8192726 | 0.14 | 0.19 | 0.08 | ||
| 516G>T | rs3745274 | 0.13 | 0.18 | 0.23 | [4,22] | |
| A>G | rs2279343 | 0.084 | N/A | N/A | ||
| *2 (430C>T) | rs1799853 | 0.0080 | 0.0028 | 0.0013 | [4] | |
| rs7900194 | ||||||
| *3 (1075A>C) | rs1057910 | 0.042 | 0.04 | 0.06 | ||
| *13 (269T>C) | rs72558187 | 0.0040 | 0.0017 | N/A | ||
| *2 (681G>A) | rs4244285 | 0.26 | 0.33 | 0.15 | [4,23] | |
| *3 (636G>A) | rs4986893 | 0.093 | 0.05 | N/A | ||
| *17 (-806C>T) | rs12248560 | 0.012 | 0.02 | 0.23 | ||
| *1B (-392A>G) | rs2740574 | 0.0016 | 0 | 0.03 | [4] | |
| *1G (20230G>A) | rs2242480 | 0.21 | 0.25 | 0.07 | ||
| *3 (6986A>G) | rs776746 | 0.22 | 0.28 | 0.05 | [4,24] | |
| *6 (14690G>A) | rs10264272 | 0.0016 | 0.0017 | N/A | ||
| rs56411402 | ||||||
| *3 (C>T) | rs2108622 | 0.32 | 0.21 | 0.27 | [4] | |
†2988G>A or 2989G>A on NG_008376.3..
Table 3 . Alternative allele frequency for gene types other than CYP450 genes.
| Gene | Allele | rs number | Alternative allele frequency | Reference | ||
|---|---|---|---|---|---|---|
| Korean | East Asian | Caucasian | ||||
| *57:01 (T>G) (HCP5) | rs2395029 | 0.0032 | 0.01 | 0.05 | [4,25,26] | |
| *15:02:01 | N/A | 0.0033 | N/A | N/A | [27] | |
| *13:01:01 | N/A | 0.021 | N/A | N/A | [27] | |
| *58:01 | rs9263726 | 0.0096 | 0.05 | 0.13 | [4,28] | |
| *31:01:02 | rs1061235 | 0.054 | N/A | N/A | [27] | |
| *33:03 | N/A | 0.16 | N/A | N/A | [27] | |
| *03:02 | rs2523447 | 0.11 | N/A | N/A | [27] | |
| *04:01:01:01 | N/A | 0 | 0.01 | N/A | [27] | |
| *02:01 | N/A | 0.073 | N/A | N/A | [27] | |
| *01:01:01 | N/A | 0.065 | N/A | N/A | [27] | |
| 3435T>C | rs1045642 | 0.65 | 0.6 | 0.47 | [4,29] | |
| 2677T>G/A | rs2032582 | 0.62 | 0.55 | 0.57 | ||
| 1236C>T | rs1128503 | 0.38 | 0.34 | 0.57 | ||
| 421C>A | rs2231142 | 0.27 | 0.29 | 0.1 | [4,30] | |
| G>A | rs1872328 | 0.013 | 0.01 | 0.04 | [4] | |
| C>T | rs7412 | 0.051 | 0.08 | 0.07 | [4] | |
| T>C | rs4673993 | 0.19 | 0.21 | 0.31 | [4] | |
| A>G | rs4680 | 0.28 | 0.29 | 0.52 | [4] | |
| *2A (1905+1G>A) | rs3918290 | 0.10 | 0.12 | 0.04 | [4,23] | |
| rs17376848 | ||||||
| 85T>C | rs3918290 | 0 | N/A | N/A | ||
| A>G | rs78060119 | 0.016 | 0.0035 | 0.13 | ||
| rs56293913 | ||||||
| T>A, G | rs121434568 | N/A | N/A | N/A | N/A | |
| A>G | rs11615 | 0.73 | 0.72 | 0.4 | [4] | |
| C>A | rs3212986 | 0.24 | 0.28 | 0.27 | ||
| 313A>G | rs1695 | 0.18 | 0.17 | 0.32 | [4] | |
| G>A | rs1050828 | N/A | N/A | N/A | N/A | |
| C>G | rs6295 | 0.75 | 0.78 | 0.44 | [4] | |
| C>G | rs1414334 | 0.98 | 0.99 | 0.85 | [4] | |
| T>G | rs8099917 | 0.047 | 0.07 | 0.17 | [4] | |
| A>G | rs11881222 | 0.052 | 0.07 | 0.3 | ||
| C>T | rs12979860 | 0.051 | 0.08 | 0.32 | ||
| C>T | rs12979860 | 0.051 | 0.08 | 0.32 | [4] | |
| C>G | rs368234815 | 0.038 | 0.06 | 0.17 | ||
| rs4803221 | ||||||
| T>C | rs5219 | 0.60 | 0.62 | 0.66 | [4] | |
| C>A | rs489693 | 0.25 | 0.24 | 0.31 | [4] | |
| G>A | rs1801133 | 0.42 | 0.37 | 0.35 | [4] | |
| *5 (341T>C) | rs1801280 | 0.026 | 0.03 | 0.45 | [4,31] | |
| *6A (590G>A) | rs1799930 | 0.18 | 0.23 | 0.28 | ||
| *7 (857G>A) | rs1799931 | 0.13 | 0.16 | 0.02 | ||
| *7B (282C>T) | rs1041983 | 0.32 | 0.39 | 0.3 | ||
| *14 (191G>A) | rs1801279 | 0.0024 | 0.0061 | 0.0031 | ||
| rs1805158 | ||||||
| p.R139C (C>T) | rs116855232 | 0.12 | 0.12 | 0.004 | [4] | |
| G>A | rs10306114 | 0.0016 | 0.01 | 0.0041 | [4] | |
| rs142017527 | ||||||
| T>C | rs1051266 | 0.43 | 0.47 | 0.56 | [4] | |
| *1B (492A>G or 388A>G) | rs2306283 | 0.72 | 0.75 | 0.4 | [4,32] | |
| *5 (625T>C or 521T>C) | rs4149056 | 0.14 | 0.13 | 0.17 | ||
| *3B (460G>A) | rs1800460 | 0.73 | 0.75 | 0.78 | [4,33] | |
| rs2842934 | ||||||
| *3C (719A>G) | rs1142345 | 0.016 | 0.02 | 0.03 | ||
| *80 (364C>T) | rs887829 | 0.13 | 0.13 | 0.3 | [4,34] | |
| *6 (71G>A) | rs4148323 | 0.18 | 0.17 | 0.01 | ||
| 3673C>T or -1639C>T | rs9923231 | 0.92 | 0.92 | 0.4 | [4,35] | |
| 9041C>T or 3730C>T | rs7294 | 0.074 | 0.08 | 0.35 | ||
| 6484C>T or 1173C>T | rs9934438 | 0.92 | 0.92 | 0.4 | ||
ABCB1=ATP binding cassette subfamily B, ABCG2=ATP-binding cassette super-family G member 2, ACYP2=acylphosphatase 2, APOE=apolipoprotein E, ATIC=5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase/IMP cyclohydrolase, COMT=catechol-O-methyltransferase, DPYD=dihydropyrimidine dehydrogenase, EGFR=epidermal growth factor receptor, ERCC1=excision repair cross-complementing group, G6PD=glucose-6-phosphate dehydrogenase, GSTP1=glutathione S-transferase Pi, HLA=human leukocyte antigen, HTR1A=5-hydroxytryptamine receptor 1A, HTR2C=5-hydroxytryptamine receptor 2C, IFNL=interferon lambda 3, KCNJ11=potassium inwardly rectifying channel subfamily J, MC4R=melanocortin-4 receptor, MTHFR=methylenetetrahydrofolate reductase, NAT2=N-acetyltransferase 2, NUDT15=nudix hydrolase 15, PTGS1=prostaglandin-endoperoxide synthase 1, SLC19A1=solute carrier family 19, SLCO1B1=solute carrier organic anion transporter family member 1B1, TPMT=thiopurine S methyltransferase, UGT1A1=UDP-glucuronosyltransferase 1A1, VKORC1=vitamin K epoxide reductase complex subunit 1..
The PharmGAF DB was constructed by combining the results of the pharmacogenomic literature analysis and the AAF information. The PharmGAF DB consists of 80 drugs, each of which displays drug responses and AAFs based on genetic variations. The PharmGAF DB was developed as some parts of the MFDS drug response DB by race/ethnicity and the full DB is shown in MFDS website [8]. The full DB includes information about related genes, allele frequencies by ethnicity, PK differences by genotype, PD differences by genotype, clinical applications, and drug labels.
In this study, a reliable database regarding drug responses related to genetic variants was compiled. Drugs included in regulatory affairs (MFDS, FDA, EMDA and PMDA) and pharmacogenetic guidelines were listed, and the corresponding PK/PD responses related to genetic variants were collected by painstaking manual curation. Furthermore, it is well known that the frequency of PK/PD responses can vary depending on the gene variant frequency. Therefore, we also collected information on AAFs by ethnicity for all gene variants in the literature surveyed here and thereby link these data to the genome-drug response DB for Koreans. It is expected that our PharmGAF DB will be utilized as a reference by those conducting studies related to drug effectiveness and safety. Given that the PharmGAF DB contains both the frequency of variability and the risk (e.g., odds ratio), it can also be used to design clinical trials to test the clinical and economic effects of genetic testing while reducing both the time and cost.
Many studies have constructed genetic variance-drug response DBs. For instance, Ubiquitous Pharmacogenomics (U-PGx) aims to increase the quality of life and reduce the costs borne by patients by constructing a DB containing 50 variants of 13 genes and the corresponding drug responses [12]. The ClinGen PGx Working Group curated 1,750 gene-related data instances based on 11,413 experts [13]. The PharmacoGenomic Mutation Database (PGMD) released a version including over 117,000 unique pharmacogenomic observations, covering all 24 disease superclasses and nearly 1,400 drugs [14]. PharmVar has established a repository of pharmacogenomics variations for
The PharmGAF DB can have clinical implications. For example, azathioprine induced leukopenia is associated with rs116855232 [18]. The allele frequency of this variant is higher in Koreans and East Asians than in Caucasians, with frequencies of 0.12, 0.12, and 0.004, respectively [4]. This suggests that there should be greater attention to azathioprine induced leukopenia in East Asians compared to Caucasians. In fact, a higher incidence of leukopenia has been reported in Asians compared to the Caucasian population, aligning with the frequency trend of rs116855232 [18]. The differences in allele frequencies among ethnicities presented in this database imply that the frequency of related PK or PD phenotypes may also vary. Therefore, this information can provide insights into which phenotypes need more attention in each ethnic group.
The advantage of this study is that AAF information is combined with genome-drug response information. A researcher would be able to know at a glance how frequently a genomic variant will affect the drug response in a specific race. Risk management would also be possible by multiplying the risk (e.g., odds ratio) and frequency (e.g. AAF) (modified from the risk priority number formula suggested by Lee et al. [19]), thereby preventing serious adverse events from happening.
However, there are several limitations in our study. First, although drug reactions and AAF information for various races were collected, drugs used in other countries were not included in the DB because the collected drugs were limited to drugs marketed in Korea. Second, only high-grade genomic information suggested by PharmGKB was collected. Even if the grade is low, there may be drug-variant annotations with high variant frequencies. However, as combination drugs were not included in our study design, caution should be exercised when using the PharmGAF DB. Future studies are needed in which the AAFs for all drug-variant pairs against all ethnicities are presented. This would provide a fuller set of integrated information.
The PharmGAF DB was developed by merging information about drug responses related to genotypes and allele frequencies among different ethnic groups. The PharmGAF DB as presented here could contribute to maximizing drug effects and minimizing side effects by defining customized treatments applicable to clinical practice. We hope this freely available DB will be widely used for a wide range of clinical purposes.
Not applicable.
This research was supported by a grant (20182MFDS444) from the Ministry of Food and Drug Safety in 2020.
We would like to thank Yoonjae Hwang, Woo Bin Lee, Won-gu Kang, Geon-yeon Kim, Minseo Kang, Dohye Kim, Jonghyun Sung, Sung-eun Jo, Yejin Shin for their contributions to data collection.
Supplementary materials can be found via https://doi.org/10.59931/rcp.23.0003.
rcp-1-2-144-supple.pdf
Table 1 Summary table of PharmGAF DB (key excerpt [level of evidence: 1A])
| Drug | Gene | Variant | PK | PD |
|---|---|---|---|---|
| Abacavir | *57:01:01 | Toxicity | ||
| Allopurinol | *58:01 | Toxicity | ||
| Amitriptyline | *1, *2, *3, *17 | Metabolism | ||
| *1/*1xN, *2/*2xN | Metabolism | Toxicity | ||
| Atazanavir | rs887829 | Toxicity | ||
| Atomoxetine | *1, *2, *3, *4, *5, *10 | Metabolism | Toxicity | |
| Azathioprine | *1, *2, *3A, *3B, *3C | Dosage/toxicity | ||
| rs116855232 | Toxicity | |||
| Capecitabine | rs75017182 | Toxicity | ||
| rs55886062 | Toxicity | |||
| rs67376798 | Toxicity | |||
| rs3918290 | Toxicity | |||
| Carbamazepine | *31:01:02 | Toxicity | ||
| *15:02:01 | Toxicity | |||
| Celecoxib | *1, *2, *3, *13 | Metabolism | ||
| Citalopram | *1, *2, *3, *4, *17 | Metabolism | Toxicity | |
| Clomipramine | *1, *1xN, *2, *2xN, *3, *4, *5, *6 | Toxicity | ||
| *1, *2, *3 | Metabolism | |||
| Clopidogrel | *1, *2, *3, *4, *5, *6, *8 | Efficacy/toxicity | ||
| rs4986893 636G>A *3 | Efficacy/toxicity | |||
| rs4244285 681G>A *2 | Efficacy/toxicity | |||
| rs12248560 -806C>T *17 | Efficacy/toxicity | |||
| Codeine | *1, *1xN, *2, *2xN, *3, *4, *5, *6, *10, *17, *40, *41 | Metabolism | Efficacy/toxicity | |
| Dexlansoprazole | *1, *2, *3 | Metabolism | ||
| Doxepin | *1, *1xN, *2, *2xN, *3, *4, *5 | Metabolism | ||
| *1, *2 | Metabolism | |||
| Efavirenz | rs3745274 516G>T | Metabolism | ||
| *1, *6, *26 516G>T | Dosage/toxicity | |||
| *1, *4, *6, *9, *16, *18, *28 | Metabolism | |||
| Escitalopram | *1, *2, *3, *4, *17 | Metabolism | ||
| Fluorouracil | rs3918290 | Toxicity | ||
| rs75017182 | Toxicity | |||
| rs55886062 | Toxicity | |||
| rs67376798 | Toxicity | |||
| Fluvoxamine | *1, *3, *4, *5, *6, *10 | Metabolism | ||
| Ibuprofen | *1, *2, *3 | Metabolism | ||
| Imipramine | *1, *2, *3, *17 | Metabolism | ||
| *1, *3, *4, *5 | Metabolism | |||
| Lansoprazole | *1, *2, *3, *8, *9, *17 | Efficacy | ||
| Lornoxicam | *1, *3, *13 | Metabolism | ||
| Meloxicam | *1, *2, *3, *13 | Metabolism | ||
| Mercaptopurine | rs116855232 | Metabolism | Toxicity | |
| *1, *2, *3A, *3B, *3C, *4 | Dosage/toxicity | |||
| Nortriptyline | *1, *2, *2xN, *3, *4, *5, *6, *10 | Metabolism | ||
| Omeprazole | *1, *2, *3, *9, *10, *17, *24, *26 | Metabolism | Efficacy | |
| Ondansetron | *1, *1xN | Metabolism | ||
| Oxcarbazepine | *15:02:01 | Toxicity | ||
| Pantoprazole | *1, *2, *3, *17 | Metabolism | Efficacy | |
| Paroxetine | *1, *1xN, *2, *2xN, *3, *4, *5, *6, *10 | Efficacy | ||
| Peginterferon alfa-2a | rs12979860 | Efficacy | ||
| Phenytoin | *1, *2, *3 | Metabolism | Toxicity | |
| rs1057910 | Toxicity | |||
| 15:02:01 | Toxicity | |||
| Piroxicam | *1, *2, *3 | Metabolism | ||
| Rasburicase | A-202A_376 | Toxicity | ||
| Ribavirin | rs12979860 | Efficacy | ||
| rs8099917 | Efficacy | |||
| Sertraline | CYP2D6 | *1,*3,*4,*5,*10,*17,*41 | Efficacy | |
| Simvastatin | rs4149056 | Efficacy/toxicity | ||
| Tacrolimus | *1, *3, *6, *7 | Efficacy/dosage | ||
| Tamoxifen | *1, *3, *4, *5, *6, *10, *41 | Efficacy | ||
| Voriconazole | *1, *2, *3, *17 | Metabolism | Efficacy/toxicity | |
| Warfarin | *1, *2, *3, *5, *6, *11 | Toxicity | ||
| *1, *3 | Dosage | |||
| rs9923231 | Dosage |
CYP=cytochrome P450, DPYD=dihydropyrimidine dehydrogenase, G6PD=glucose-6-phosphate dehydrogenase, HLA=human leukocyte antigen, IFNL=interferon lambda, NUDT15=nudix hydrolase 15, SLCO1B1=solute carrier organic anion transporter family member 1B1, TPMT=thiopurine S methyltransferase, UGT1A1=UDP-glucuronosyltransferase 1A1, VKORC1=vitamin k epoxide reductase complex subunit 1.
Table 2 Alternative allele frequency for cytochrome P450 (CYP) enzyme
| Gene | Allele | rs number | Alternative allele frequency | Reference | ||
|---|---|---|---|---|---|---|
| Korean | East Asian | Caucasian | ||||
| *2 (2850C>T) | rs16947 | 0.17 | 0.15 | 0.34 | [4,20] | |
| *4 (1846 C>T) | rs3892097 | 0.0016 | 0.0017 | 0.19 | ||
| rs1800716 | ||||||
| *10 (100C>T) | rs1065852 | 0.51 | 0.52 | 0.2 | ||
| rs1081003 | ||||||
| *17 (1023C>T) | rs28371706 | 0.49 | 0.52 | 0.02 | ||
| rs1081003 | ||||||
| *41† | rs28371725 | 0.027 | 0.03 | 0.09 | ||
| A>C | rs4646 | 0.69 | 0.72 | 0.71 | [4] | |
| g.-48T>G | rs28399433 | 0.25 | 0.27 | 0.08 | [4,21] | |
| 6600G>T | rs28399468 | 0.025 | 0.04 | N/A | ||
| g.6558T>C | rs5031016 | 0.11 | 0.1 | 0.0013 | ||
| 1839G>T | rs8192726 | 0.14 | 0.19 | 0.08 | ||
| 516G>T | rs3745274 | 0.13 | 0.18 | 0.23 | [4,22] | |
| A>G | rs2279343 | 0.084 | N/A | N/A | ||
| *2 (430C>T) | rs1799853 | 0.0080 | 0.0028 | 0.0013 | [4] | |
| rs7900194 | ||||||
| *3 (1075A>C) | rs1057910 | 0.042 | 0.04 | 0.06 | ||
| *13 (269T>C) | rs72558187 | 0.0040 | 0.0017 | N/A | ||
| *2 (681G>A) | rs4244285 | 0.26 | 0.33 | 0.15 | [4,23] | |
| *3 (636G>A) | rs4986893 | 0.093 | 0.05 | N/A | ||
| *17 (-806C>T) | rs12248560 | 0.012 | 0.02 | 0.23 | ||
| *1B (-392A>G) | rs2740574 | 0.0016 | 0 | 0.03 | [4] | |
| *1G (20230G>A) | rs2242480 | 0.21 | 0.25 | 0.07 | ||
| *3 (6986A>G) | rs776746 | 0.22 | 0.28 | 0.05 | [4,24] | |
| *6 (14690G>A) | rs10264272 | 0.0016 | 0.0017 | N/A | ||
| rs56411402 | ||||||
| *3 (C>T) | rs2108622 | 0.32 | 0.21 | 0.27 | [4] | |
†2988G>A or 2989G>A on NG_008376.3.
Table 3 Alternative allele frequency for gene types other than CYP450 genes
| Gene | Allele | rs number | Alternative allele frequency | Reference | ||
|---|---|---|---|---|---|---|
| Korean | East Asian | Caucasian | ||||
| *57:01 (T>G) (HCP5) | rs2395029 | 0.0032 | 0.01 | 0.05 | [4,25,26] | |
| *15:02:01 | N/A | 0.0033 | N/A | N/A | [27] | |
| *13:01:01 | N/A | 0.021 | N/A | N/A | [27] | |
| *58:01 | rs9263726 | 0.0096 | 0.05 | 0.13 | [4,28] | |
| *31:01:02 | rs1061235 | 0.054 | N/A | N/A | [27] | |
| *33:03 | N/A | 0.16 | N/A | N/A | [27] | |
| *03:02 | rs2523447 | 0.11 | N/A | N/A | [27] | |
| *04:01:01:01 | N/A | 0 | 0.01 | N/A | [27] | |
| *02:01 | N/A | 0.073 | N/A | N/A | [27] | |
| *01:01:01 | N/A | 0.065 | N/A | N/A | [27] | |
| 3435T>C | rs1045642 | 0.65 | 0.6 | 0.47 | [4,29] | |
| 2677T>G/A | rs2032582 | 0.62 | 0.55 | 0.57 | ||
| 1236C>T | rs1128503 | 0.38 | 0.34 | 0.57 | ||
| 421C>A | rs2231142 | 0.27 | 0.29 | 0.1 | [4,30] | |
| G>A | rs1872328 | 0.013 | 0.01 | 0.04 | [4] | |
| C>T | rs7412 | 0.051 | 0.08 | 0.07 | [4] | |
| T>C | rs4673993 | 0.19 | 0.21 | 0.31 | [4] | |
| A>G | rs4680 | 0.28 | 0.29 | 0.52 | [4] | |
| *2A (1905+1G>A) | rs3918290 | 0.10 | 0.12 | 0.04 | [4,23] | |
| rs17376848 | ||||||
| 85T>C | rs3918290 | 0 | N/A | N/A | ||
| A>G | rs78060119 | 0.016 | 0.0035 | 0.13 | ||
| rs56293913 | ||||||
| T>A, G | rs121434568 | N/A | N/A | N/A | N/A | |
| A>G | rs11615 | 0.73 | 0.72 | 0.4 | [4] | |
| C>A | rs3212986 | 0.24 | 0.28 | 0.27 | ||
| 313A>G | rs1695 | 0.18 | 0.17 | 0.32 | [4] | |
| G>A | rs1050828 | N/A | N/A | N/A | N/A | |
| C>G | rs6295 | 0.75 | 0.78 | 0.44 | [4] | |
| C>G | rs1414334 | 0.98 | 0.99 | 0.85 | [4] | |
| T>G | rs8099917 | 0.047 | 0.07 | 0.17 | [4] | |
| A>G | rs11881222 | 0.052 | 0.07 | 0.3 | ||
| C>T | rs12979860 | 0.051 | 0.08 | 0.32 | ||
| C>T | rs12979860 | 0.051 | 0.08 | 0.32 | [4] | |
| C>G | rs368234815 | 0.038 | 0.06 | 0.17 | ||
| rs4803221 | ||||||
| T>C | rs5219 | 0.60 | 0.62 | 0.66 | [4] | |
| C>A | rs489693 | 0.25 | 0.24 | 0.31 | [4] | |
| G>A | rs1801133 | 0.42 | 0.37 | 0.35 | [4] | |
| *5 (341T>C) | rs1801280 | 0.026 | 0.03 | 0.45 | [4,31] | |
| *6A (590G>A) | rs1799930 | 0.18 | 0.23 | 0.28 | ||
| *7 (857G>A) | rs1799931 | 0.13 | 0.16 | 0.02 | ||
| *7B (282C>T) | rs1041983 | 0.32 | 0.39 | 0.3 | ||
| *14 (191G>A) | rs1801279 | 0.0024 | 0.0061 | 0.0031 | ||
| rs1805158 | ||||||
| p.R139C (C>T) | rs116855232 | 0.12 | 0.12 | 0.004 | [4] | |
| G>A | rs10306114 | 0.0016 | 0.01 | 0.0041 | [4] | |
| rs142017527 | ||||||
| T>C | rs1051266 | 0.43 | 0.47 | 0.56 | [4] | |
| *1B (492A>G or 388A>G) | rs2306283 | 0.72 | 0.75 | 0.4 | [4,32] | |
| *5 (625T>C or 521T>C) | rs4149056 | 0.14 | 0.13 | 0.17 | ||
| *3B (460G>A) | rs1800460 | 0.73 | 0.75 | 0.78 | [4,33] | |
| rs2842934 | ||||||
| *3C (719A>G) | rs1142345 | 0.016 | 0.02 | 0.03 | ||
| *80 (364C>T) | rs887829 | 0.13 | 0.13 | 0.3 | [4,34] | |
| *6 (71G>A) | rs4148323 | 0.18 | 0.17 | 0.01 | ||
| 3673C>T or -1639C>T | rs9923231 | 0.92 | 0.92 | 0.4 | [4,35] | |
| 9041C>T or 3730C>T | rs7294 | 0.074 | 0.08 | 0.35 | ||
| 6484C>T or 1173C>T | rs9934438 | 0.92 | 0.92 | 0.4 | ||
ABCB1=ATP binding cassette subfamily B, ABCG2=ATP-binding cassette super-family G member 2, ACYP2=acylphosphatase 2, APOE=apolipoprotein E, ATIC=5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase/IMP cyclohydrolase, COMT=catechol-O-methyltransferase, DPYD=dihydropyrimidine dehydrogenase, EGFR=epidermal growth factor receptor, ERCC1=excision repair cross-complementing group, G6PD=glucose-6-phosphate dehydrogenase, GSTP1=glutathione S-transferase Pi, HLA=human leukocyte antigen, HTR1A=5-hydroxytryptamine receptor 1A, HTR2C=5-hydroxytryptamine receptor 2C, IFNL=interferon lambda 3, KCNJ11=potassium inwardly rectifying channel subfamily J, MC4R=melanocortin-4 receptor, MTHFR=methylenetetrahydrofolate reductase, NAT2=N-acetyltransferase 2, NUDT15=nudix hydrolase 15, PTGS1=prostaglandin-endoperoxide synthase 1, SLC19A1=solute carrier family 19, SLCO1B1=solute carrier organic anion transporter family member 1B1, TPMT=thiopurine S methyltransferase, UGT1A1=UDP-glucuronosyltransferase 1A1, VKORC1=vitamin K epoxide reductase complex subunit 1.
PDF
Standard view
Export citation
Share 
Download
Supplementary file
