Conference report: pharmacogenomics in special populations at WCP2018

Abstract

The 18th World Congress of Basic and Clinical Pharmacology (WCP2018), coordinated by IUPHAR and hosted by the Japanese Pharmacological Society and the Japanese Society of Clinical Pharmacology and Therapeutics, was held in July 2018 at the Kyoto International Conference Center, in Kyoto, Japan. Having as its main theme ‘Pharmacology for the Future: Science, Drug Development and Therapeutics’, WCP2018 was attended by over 4500 delegates, representing 78 countries. The present report is an overview of a symposium at WCP2018, entitled Pharmacogenomics in Special Populations, organized by IUPHAR´s Pharmacogenetics/Genomics (PGx) section. The PGx section congregates distinguished scientists from different continents, covering expertise from basic research, to clinical implementation and ethical aspects of PGx, and one of its major activities is the coordination of symposia and workshops to foster exchange of PGx knowledge (https://iuphar.org/sections‐subcoms/pharmacogenetics‐genomics/). The symposium attracted a large audience to listen to presentations covering various areas of research and clinical adoption of PGx in Oceania, Africa, Latin America and Asia.

Introduction

The 18th World Congress of Basic and Clinical Pharmacology (WCP2018), coordinated by IUPHAR and hosted by the Japanese Pharmacological Society and the Japanese Society of Clinical Pharmacology and Therapeutics, was held in July 2018 at the Kyoto International Conference Center, in Kyoto, Japan. Having as its main theme ʻPharmacology for the Future: Science, Drug Development and Therapeuticsʼ, WCP2018 was attended by over 4500 delegates, representing 78 countries.

The present report is an overview of a symposium at WCP2018, entitled Pharmacogenomics in Special Populations, organized by IUPHAR's Pharmacogenetics/Genomics (PGx) section. The PGx section congregates distinguished scientists from different continents, covering expertise from basic research, to clinical implementation and ethical aspects of PGx, and one of its major activities is the coordination of symposia and workshops to foster exchange of PGx knowledge (https://iuphar.org/sections‐subcoms/pharmacogenetics‐genomics/). The symposium attracted a large audience to listen to presentations covering various areas of research and clinical adoption of PGx in Oceania, Africa, Latin America and Asia.

Pharmacogenetics in the Oceania region: precision medicine challenges

Pharmacogenetics in the Oceania region: precision medicine challenges was presented by Professor Andrew Somogyi (University of Adelaide, Adelaide, Australia). Oceania, comprising the large islands of Australia and New Zealand (Australasia), and Papua New Guinea (PNG), and a very large number of relatively small islands of Melanesia, Micronesia and Polynesia, covers over 8.5 million square kilometres, but only contains about 40 million inhabitants. PGx studies of the inhabitants of PNG have been recently reviewed 1, whereas for Aboriginal Australians and Melanesian and Polynesian peoples, only the genes encoding cytochrome P450 (CYP) CYP2D6 and CYP2C19 have been investigated to a certain extent 2, 3. In general, results cannot be readily predicted from one region to another. For example, the frequency of CYP2C19*2 varies from 45% in PNG to 24% in Aboriginal Australian and Maori peoples, whereas CYP2C19*3, another nonfunctional allele, ranges from 2% in Maori individuals to about 20% in PNG and Australian Aborigines. The CYP2C19*17 allele frequency is much lower in the latter population, resulting in a predicted 50% lower frequency of increased enzyme function compared with Caucasians.

The CYP2D6 genotype and predicted phenotype are dependent on copy number and sequence variation detection platforms used; nevertheless, it appears that poor metabolizers (PMs) comprise only about 2% over all of Oceania. This may have implications for CYP2D6‐catalysed primaquine dosing for Plasmodium vivax malaria. Indeed, the impact of CYP2D6 polymorphism on the effectiveness of primaquine to prevent P. vivax malaria relapses was discussed in another presentation at the symposium (see below).

In PNG, most PGx studies have focused on infectious diseases, and results pertinent to the antiretroviral agent efavirenz in HIV‐infected patients were presented. Efavirenz is mainly metabolized by CYP2B6, and poor metabolizer status is associated with central nervous system (CNS)/psychiatric effects. The frequency of the major CYP2B6 variant *6 is about 60% in PNG, compared with less than 20% in Caucasian and South Asian populations. Data from 52 PNG subjects, most of whom had CNS/psychiatric adverse effects, revealed, however, that only drowsiness was related to CYP2B6*6 carrier status.

Regarding N‐acetyltransferase 2 (NAT2) and acetylator status, no genomic studies have been conducted in PNG but almost all individuals are rapid acetylators, and therefore the incidence of isoniazid‐induced hepatotoxicity is rare, although patients might be being underdosed. In Australian Aborigines, about one‐third are slow acetylators and have a relatively high frequency of the NAT2*7 allele, at 40% compared with 1% in Europeans 4.

Minimal data are available on drug transporters in Oceania; however, the frequency of the gene encoding ATP‐binding cassette subfamily B member 1 (ABCB1) 3435C > T variant was 67% in PNG subjects, and even higher, at 71%, in Australian Aborigines. These are substantially above the 40–50% range found in Europeans and South and East Asians. Such high allele variant frequencies in peoples from Oceania may have major implications for the efficacy of, and adverse effects associated with, many medicines which are ABCB1 substrates.

The gene encoding human leukocyte antigen (HLA) B*13:01 is associated with severe hypersensitivity reactions [Stevens–Johnson syndrome (SJS); toxic epidermal necrolysis (TEN); and drug reaction with eosinophilia and systemic symptoms (DRESS) to phenytoin, the frequency of which is relatively high in several South Asian countries]. In PNG and in Aboriginal Australians from Northern Australia, the HLA‐B*13:01 frequency is almost 25%. Another variant, HLA‐B*56:02, has been associated with phenytoin‐induced DRESS and several case reports of phenytoin‐associated morbidity and mortality. The frequency of this allele can be over 5% in Aboriginal Australians, but is essentially absent in Europeans.

Although the frequencies of some important pharmacogenes are markedly different in Oceania (especially in PNG and in Aboriginal Australians) compared with Caucasian and some Asian populations, these frequencies can be fairly divergent across the region. Many important genes and genotype–phenotype correlations have not been assessed, with clinical relevance and translation assessment faced by limited local resources. Caution should be exercised when interpreting the genotype with respect to the phenotype, with the vexing issue that alleles rarely found in Europeans may be common in Oceania.

The challenges in conducting PGx studies are, firstly, ethical, in terms of demonstrating that PGx testing will help and not hinder the health of indigenous peoples in Oceania; and, secondly, presenting evidence that the efficacy and toxicity of some drugs can be different, as now shown with phenytoin in Aboriginal Australians. Having indigenous precision medicine champions with community support who can drive the research direction is critical for drug therapy optimization. In PNG, logistics are a major challenge, as biological sampling is often conducted in remote communities; thus, sample collection, processing and transport are problematic. The results to date and the above challenges result in research being needed to address cost‐effective and nondiscriminatory precision medicine for the understudied indigenous peoples of Oceania.

African Pharmacogenomics Research Consortium: Focus on HIV, tuberculosis and malaria treatment

African Pharmacogenomics Research Consortium: Focus on HIV, tuberculosis and malaria treatment was presented by Professor Eleni Aklillu (Karolinska Institutet, Stockholm, Sweden). Populations of sub‐Saharan Africa (SSA) are the most genetically and ethnically diverse in the world, displaying extensive population substructure and less linkage disequilibrium between loci compared with peoples of non‐African ancestry 5. This wide genetic heterogeneity in African populations offers the opportunity to identify rare alleles and haplotypes that play a role in determining susceptibility to diseases and adverse drug responses. The African Pharmacogenomics Consortium was established to prompt PGx research and clinical implementation in African populations 6. Through regional and international research collaborations, PGx research in Africa is progressing to genomics and environmental determinants of common diseases, with the goal of improving the health of African populations 7. Policy makers in Africa have started using PGx data to recommend population‐specific treatment guidelines. For instance, CYP2D6 metabolizes codeine into its active metabolite, morphine, and CYP2D6 ultrarapid metabolizers may experience the symptoms of morphine overdose, which include life‐threatening or fatal respiratory depression 8. The US Federal Drug Administration (FDA) issued a safety alert on the use of codeine for pain in children, particularly ultrarapid metabolizers 9, which is fairly common (29% functional CYP2D6 gene duplication frequency) in those with Ethiopian origin 10. Later, the Ethiopian Food, Medicine and Health Care Administration and Control Authority banned codeine use in the country.

SSA carries a high burden of HIV/AIDS, tuberculosis (TB) and malaria. Coinfection is common, and cotreatment is challenging owing to drug interactions and overlapping toxicities. Genetic and environmental factors, nutrition, coinfection, comorbidity and use of herbal medicines may result in different drug–response profiles. Specific variant alleles such as CYP2D6*17, CYP3A5*6, CYP3A*7 or the occurrence of decreased function alleles such as CYP2B6*6 at a higher frequency may account for population‐specific phenotype variations in peoples with black African ancestry 11, 12. Key PGx studies in different African populations, with a focus on drug interactions between first‐line antiretroviral, anti‐TB and antimalarial drugs, and their impact on treatment outcomes, were presented at the symposium. The need for studies in different African populations was highlighted as the extensive genetic diversity of SSA restricts extrapolation of PGx data among countries 13, 14.

Efavirenz, a potent non‐nucleoside reverse transcriptase inhibitor and a cornerstone first‐line antiretroviral drug in Africa, is metabolized mainly by CYP2B6. CYP2B6*6, a decreased‐function variant allele causing higher plasma efavirenz exposure, is implicated in drug‐induced hepatotoxicity 15, 16, 17 and neuropsychiatric toxicities 18, 19. CYP2B6*6 occurs at a higher frequency in people of African origin, reaching up to 50% 20, 21. CYP2B6*18, another decreased‐function variant allele associated with high efavirenz plasma concentrations, occurs with an allele frequency of up to 17% in Black Africans but is absent in Asians and Caucasians.

PGx studies in genes coding for HLA revealed that HLA‐B*57 alleles (B*57:03 and B*57:02) confer susceptibility to anti‐TB and antiretroviral drug‐induced liver toxicity 22. HLA‐B*57:03 is the most prevalent HLA‐B*57 subtype in black African populations, whereas HLA‐B*57:01, which is strongly associated with abacavir‐induced hypersensitivity reaction, is absent in black Africans 22. HLA‐B*57:01 occurs with frequencies up to 5% in Asians and Caucasians, and prior screening for this allele is highly recommended in these populations.

The importance of pharmacogenetic variations in modulating interactions between antiretroviral and anti‐TB drugs was noted. A distinct population‐specific efavirenz–rifampicin drug interaction profile required varying efavirenz dose recommendations for TB‐HIV patients during rifampicin‐based anti‐TB cotreatment. Rifampicin, a potent inducer of efavirenz‐metabolizing enzymes (CYP2B6, CYP2A6 and CYP3A4), significantly reduces plasma efavirenz concentration in white individuals 23, 24. Accordingly, the FDA and the British HIV Association recommend that the dose of efavirenz is increased (to 800 mg day^–1) during rifampicin cotherapy 25, 26, 27. However, studies in African patients reported no significant effect of rifampicin‐based anti‐TB therapy on efavirenz plasma concentration 28, 29, 30, 31, and the World Health organization recommends the standard 600 mg day^–1 efavirenz dose during rifampicin–efavirenz cotreatment 32. Recent studies reported that even the standard 600 mg daily efavirenz dose is unnecessarily high for SSA populations, and suggested lowering the efavirenz dose based on CYP2B6 genotype 33, 34. Future randomized, clinical trials are required to confirm that the proposed CYP2B6 genotype‐based efavirenz dosing strategy is safe and effective for clinical application in Africa.

Few recent studies have investigated the pharmacogenetics of antimalarial drugs 35 and its relevance for antiretroviral and antimalarial drug interaction in HIV–malaria coinfected patients receiving dual therapy in Africa 36, 37. Lumefantrine, a long‐acting antimalarial drug, is metabolized by CYP3A4, an enzyme inducible by efavirenz 38. Efavirenz‐based antiretroviral cotreatment significantly reduces lumefantrine plasma exposure, leading to a poor malaria treatment response, particularly in CYP2B6 slow metabolizers 36.

Overall, many pharmacogenetic studies in African patients are done using a candidate gene approach, and genome‐wide association studies (GWAS) in well‐characterized matched case–control cohorts are scarce. The available few GWAS revealed potential novel genetic biomarkers for antiretroviral and anti‐TB drug‐induced liver toxicities 39.The HLA‐C*04:01 allele is associated with nevirapine‐induced SJS/TEN in sub‐Saharan Africans 40. Currently, most genome‐wide single nucleotide polymorphism (SNP) arrays are designed for European ancestry, and the tag SNPs are less efficient for non‐Europeans, particularly for SSA populations 41, 42. The lack of cost‐effective, high‐throughput genotyping technologies and microarrays with good SNP density coverage specific for African ancestry remains a challenge for the progress of PGx in Africa 42, 43. Over the past two decades, PGx research in Africa has generated ample knowledge on variant alleles that determine patient variability in the plasma exposure and treatment outcome of drugs used to treat HIV, TB and malaria. There is a serious need to reduce the burden of HIV, TB and malaria on the continent, and wide‐scale pharmacogenetic research in different African countries will significantly contribute to improving patient care and optimize treatment outcomes. The development of a point‐of‐care genotype facility to guide drug dosing is imperative to promote the translation of PGx research into clinical practice in Africa.

Pharmacogenomic implications of population admixture in the Americas

The Pharmacogenomic implications of population admixture in the Americas, especially Latin America, were examined by Professor Guilherme Suarez‐Kurtz (Brazilian Pharmacogenetics Network and Brazilian National Cancer Institute, Rio de Janeiro, Brazil). The first part of the talk expanded on the notion that PGx research and implementation in Latin America must take into account the extensive genetic heterogeneity of the population. Although Native American, European and sub‐Saharan African biogeographical ancestries are common to all Latin American countries, the relative proportions of each ancestral root vary across nations, and, most importantly, within each nation. This was illustrated with data from Brazil and Mexico, the two most populous Latin American countries, with a remarkably distinct population structure: European and sub‐Saharan ancestry predominate largely over Native American ancestry in Brazilians 44, 45, whereas in Mexico the main ancestral contributions are European and Native American, with a relatively small African contribution 46, 47. However, the individual proportions of biogeographical ancestry vary widely and, most importantly, in a continuous pattern, irrespective of self‐reported race/colour in Brazilians or geographical origin in Mexicans 48. Corollaries to this pattern are: (i) there are no ʻcutoff pointsʼ for race/colour discrimination in Brazilians or associated with geographical origin in Mexicans; (ii) average proportions of biogeographical ancestry are not predictive of the corresponding proportions at the individual level in Brazilians and Mexicans, and most likely other Latin American peoples; and (iii) combining Latin Americans into a single population entity (e.g. Latinos or Hispanics in the US) is inappropriate in view of their diversity and heterogeneity.

Professor Suarez‐Kurtz then moved on to the more formal PGx arena, and discussed the clinical implications of population admixture, using data from his studies in Brazilians. The first example related to the association of CYP3A5 polymorphisms and tacrolimus dose requirement in renal transplant patients. Santoro et al. 49 confirmed the impact of CYP3A5*3 on tacrolimus dose requirement, previously described in other populations, but disclosed two novel and important findings: first, the association of CYP3A5*6 and CYP3A5*7 with tacrolimus dosing. Second, the inclusion of Brazilian patients with the CYP3A5*1/*1 genotype, which is rare or absent in populations of European or Asian descent, allowed demonstration of a gene–dose effect of CYP3A5 polymorphisms on tacrolimus pharmacokinetics. This study provided the basis for inclusion of CYP3A5*6 and *7, in addition to the *3 allele, in the Clinical Pharmacogenetics Implementation Consortium (CPIC) guideline for tacrolimus 50.

The second example presented at the symposium pertained to warfarin. Prior to the original CPIC guideline for warfarin, published in 2011 51, Brazilian researchers 52, 53 developed and validated PGx algorithms for warfarin dosing, which explained 51–60% of the variability in stable dose requirement. Importantly, these algorithms performed equally well in self‐reported White and Black individuals, which is in marked contrast to the considerably poorer performance of several warfarin algorithms in Black Africans and African Americans vs. Europeans and White Americans 54, 55. This discrepancy was explained by the considerably higher allele frequency of the major PGx variables used in most algorithms – namely, the gene encoding vitamin K epoxide reductase complex subunit‐1 (VKORC1), rs9923231 and decreased‐function CYP2C9*2 and *3 alleles – in the extensively admixed Black Brazilians compared with Black Africans or African‐Americans 52, 56, 57. Of note, the 2017 updated CPIC guideline for warfarin 58 incorporated dosing recommendations for warfarin ʻthat are specific for continental ancestryʼ. Nevertheless, continental ancestry in admixed populations is not a categorical, but rather a continuous variable, even within self‐reported race/colour categories 56, 57. This is a caveat to the implementation of guidelines based on continental ancestry – as well as ʻraceʼ or ethnicity – in Brazil and most likely other Latin American nations.

The next topic in the presentation was the distribution of the nudix hydrolase 15 gene (NUDT15) rs116855232 (g.415C > T) SNP among Latin American populations. The variant T allele, associated with intolerance to thiopurines, is reported to be most common in East Asians and ʻHispanicsʼ, but appears to be rare or absent in Europeans and Africans 59, 60, with average frequencies of 10%, 4%, 0% and 0%, respectively, in the corresponding East Asian, admixed American, European and African groups of the 1000 Genomes Project. Suarez‐Kurtz and colleagues used the 1000 Genome Project markers to estimate the individual ancestry of the admixed American group and observed that the frequency of the rs116855232 T allele increases in parallel with the increase in the average proportion of Native American ancestry in the four subpopulations (Puerto Ricans, Colombians, Mexicans and Peruvians) comprising the admixed American group. This pattern is consistent with the notion that Asia is the likely origin of the first migrants into the American continent, 25 000–15 000 years ago.

In the ensuing discussion, the association of CYP2D6 polymorphism with the antimalarial effect of primaquine mentioned earlier by Professor Somogyi prompted reference to the recent demonstration of an increased risk of relapse of P. vivax malaria after chloroquine–primaquine combined therapy in patients from the Brazilian Amazon who were carriers of reduced‐function CYP2D6 alleles 61. The relapses were ascribed to a failure of primaquine hypnozoiticidal activity, which requires CYP2D6‐catalysed conversion of primaquine, a prodrug, into redox‐active metabolite(s).

Pharmacogenomics research and its implementation in Asia

Pharmacogenomics research and its implementation in Asia was reviewed by Dr Yoshiro Saito (National Institute of Health Sciences, Kawasaki, Japan). PGx research has been performed extensively in the Asian region, as summarized by Ang et al. 62. Publications focused on PGx are increasing in number in Asian countries/regions, particularly in relation to the CNS, cancer and blood. The majority of studies are national (local), with only a small number of collaborations across Asian countries/regions. Although most Asian publications were replication studies, several studies, primarily from Japanese and Korean groups, have reported novel findings. At the symposium, two examples were presented in which Asian researchers contributed significantly.

The first example relates to drug‐induced severe cutaneous adverse reactions (SCARs), such as SJS and TEN, important issues for drug safety. Although SJS/TEN is very rare, it requires hospitalization and, even after recovery, patients may suffer from severe sequelae such as vision loss 63. Chung and collaborators in Taiwan, Thailand, mainland China, Hong Kong, Malaysia, Singapore, Philippines and Japan (including Saito's group) assessed SJS/TEN cases using the algorithm of drug causality for epidermal necrolysis (ALDEN) score (≥4), resulting in the identification of 1028 verified cases 64. The top suspected drug classes were antiepileptic and antipsychotic agents, followed by antibiotics/antiviral drugs, and allopurinol. Lamotrigine was the leading suspected drug in Japan, whereas carbamazepine and allopurinol were the primary culprits in mainland China, Thailand, Malaysia, Singapore, Philippines and Taiwan.

In Chinese, Thai, Korean, Japanese, as well as Caucasian populations, allopurinol‐related SJS/TEN/hypersensitivity syndrome is associated with HLA‐B*58:01, which displayed relatively high frequencies in East, South‐East and South Asian populations (except for Japanese) compared with Caucasians 63, 65. The first report, from Taiwan, revealed a high odds ratio (580) for an association of SJS/TEN with HLA‐B*58:01 in the Han Chinese 66. Screening for HLA‐B*58:01 prior to starting allopurinol administration prevented allopurinol‐induced SCARs in Taiwan 67. Accordingly, pre‐emptive testing is highly recommended in the allopurinol label in Taiwan, whereas the labels in Singapore (and the UK) state that prescreening should be considered for high‐risk (high‐frequency) populations, and the Japanese label refers to studies of the causal association of HLA‐B*58:01 with the SJS/TEN/hypersensitivity syndrome.

Carbamazepine‐related SJS/TEN is associated with HLA‐A*31:01 and HLA‐B75 (such as HLA‐B*15:02 and B‐15:11). In addition, milder skin reactions are also linked to HLA‐A*31:01, which occurs in relatively high frequency in Japanese and Korean populations 65. HLA‐B75 (especially HLA‐B*15:02) is prevalent in Chinese and South‐East Asians 68. HLA‐B*15:02 was most strongly associated with carbamazepine‐induced SJS/TEN in Han Chinese in Taiwan, with an odds ratio of 1357 69. Testing for HLA‐B15:02 prior to carbamazepine administration completely prevented the onset of SJS/TEN in Taiwanese patients 70, whereas pre‐emptive testing for HLA‐A*31:01 significantly reduced the rate of carbamazepine‐related adverse skin reactions in Japanese patients 71. Table 1 compares the information on prescreening tests for HLA‐B*15:02 and HLA‐A*31:01 in the carbamazepine package inserts in different Asian countries, and, for comparison, Australia, Canada and the US. In Taiwan and Singapore, the labels state that the pre‐emptive testing is highly recommended for HLA‐B*15:02 and should be considered for HLA‐A*31:01 in patients of Asian ancestry or ancestry in a genetically at‐risk population. The Japanese label refers to studies of the association between these HLA variants and carbamazepine‐induced SCARs, but does not include recommendations for PGx testing 72. Of note, testing for HLA‐B*15:02 is covered by the national health insurance in Taiwan.

Table 1.

Pharmacogenetic testing for carbamazepine‐related SCARs in selected countries

HLA type	Country	Description of prescreening HLA test in the package inserts (section)
B*15:02	Taiwan	(Warning and precautions) Highly recommended in patients of Asian ancestrya
	Singapore	(Warning and precautions) Highly recommended in patients of Asian ancestry
	Japan	(Side effects) HLA–SCAR association mentioned by referenced papers
	USA	(Box warning) HLA test should be performed in patients with ancestry in genetically at‐risk populations
	Canada	(Special warning and precautions) HLA test to be considered in patients with ancestry in genetically at‐risk populations
	Australia	(Precautions) HLA test should be considered in patients with ancestry in genetically at‐risk populations
A*31:01	Taiwan	(Warning and precautions) HLA test should be considered in patients with ancestry in genetically at‐risk populations
	Singapore	(Warning and precautions) HLA test should be considered in patients with ancestry in genetically at‐risk populations
	Japan	(Side effects) HLA–SCAR association mentioned by referenced papers
	USA	(Warning) The risks and benefits of carbamazepine therapy should be weighed
	Canada	(Special warning and precautions) HLA test to be considered in patients with ancestry in genetically at‐risk populations
	Australia	(Precautions) HLA test should be considered in patients with ancestry in genetically at‐risk populations

HLA, human leukocyte antigen; SCAR, severe cutaneous adverse reaction

^aPharmacogenetics/genomics test reimbursed by national health insurance

Regarding the PGx of other drug‐induced SCARs, an association of CYP2C9*3 with phenytoin‐related SCARs was first reported in Han Chinese, Japanese and Malay populations 73 and then replicated in Thailand 74. CYP2C9 is the major metabolizing enzyme of phenytoin, and CYP2C9*3 is a reduced‐function allele. High plasma concentrations after phenytoin withdrawal were observed in SCAR patients 73. The CYP2C9*3 frequency is higher in South Asian and European populations than in East and South‐East Asians 65. A dapsone‐related SCAR has been associated with HLA‐B*13:01 in Han Chinese and Tai populations 75, 76.

The second example of significant contributions of Asian researchers to PGx knowledge presented at the symposium referred to the anticancer agent, irinotecan, a prodrug which requires in vivo conversion into the active metabolite, SN‐38. SN‐38 is a substrate of UDP glucuronosyltransferase family 1 member A1 (UGT1A1), encoded by the polymorphic UGT1A1 gene. Patients with the UGT1A1*6 and UGT1A1*28 alleles have reduced glucuronidation of SN‐38 and therefore may be at increased risk of irinotecan toxicity. UGT1A1*6 is prevalent in East and South‐East Asian populations, whereas UGT1A1*28 occurs at higher frequencies in South Asian and Caucasian populations than in East and South‐East Asians 65. Both alleles were shown to associate with irinotecan‐related severe neutropenia in Japanese people 77, and a meta‐analysis of 20 irinotecan studies in East Asians confirmed this association, with an odds ratio of 3.3 78. Table 2 contrasts the UGT1A1 PGx information in the irinotecan package inserts in Japan, Singapore and Taiwan, and selected non‐Asian countries. Although pre‐emptive testing is not recommended in any of the package inserts, most provide information regarding the association of UGT1A1 genotypes with intolerance to irinotecan.

Table 2.

Pharmacogenetics information on the UDP glucuronosyltransferase family 1 member A1 gene (UGT1A1) for irinotecan‐related adverse reactions in selected countries

Country	UGT1A1 pharmacogenomic information in package insert (section)
Taiwan	(Warning and precautions) UGT1A1*28 homozygotes should be administered the normally indicated starting dose. The haematological toxicity of the patients should be monitored
Singapore	(Warning and precautions) Patients with UGT1A1*6 or *28 homozygotes or compound heterozygotes may be at increased risk of serious adverse reactions, especially neutropenia, caused by reduced glucuronidation of SN‐38. Caution should be exercised when administering irinotecan to such patients. The haematological toxicity of the patients should be monitored
Japan	(Pharmacokinetics) UGT1A1*6/*28 – neutropenia association is mentioned in referenced papersa
USA	(Warning and precautions) UGT1A1*28 homozygotes are at increased risk of neutropenia
	(Dosage and administration) A reduction in the starting dose by at least one level should be considered for UGT1A1*28 homozygous patients
Canada	(Warning and precautions) A reduced starting dose should be considered for UGT1A1*28 homozygous patients
UK	(Pharmacodynamic properties) UGT1A1*28 homozygotes should be administered the normally indicated starting dose. A reduced starting dose should be considered for patients who have experienced haematological toxicity with previous treatment
Australia	(Precautions) Same as in the UK

^aPharmacogenetics/genomics test reimbursed by national health insurance

In summary, this presentation showed selected examples of genetic polymorphisms related to drug efficacy and adverse reactions in Asian populations. Dr Saito highlighted the increasing importance of collaborative pharmacogenomic research among Asian countries/regions, and suggested that further efforts aimed toward the construction of an Asian PGx research network should be a priority.

Collectively, the presentations at this symposium confirmed that PGx information is an important factor for population differences in the pharmacokinetics, efficacy and safety of drugs. In addition to the International Conference on Harmonization of Technical Requirements for the International Council for Harmonisation (ICH) of Technical Requirements for Pharmaceuticals for Human Use E5 guideline on ethnic factors 79, two related ICH guidelines were adopted and issued in 2017 on multiregional clinical trials (E17) 80 and on genomic sampling and management of genomic data (E18) 81. A harmonized PGx‐specific guideline describing ethnic specificity would be a necessary next step.

Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY 82, and are permanently archived in the Concise Guide to PHARMACOLOGY 2017/18 83.

Competing Interests

All authors have completed the Unified Competing Interest Form at www.icmje.org/coi_disclosure.pdf, and declare: G.S.K. receives grant support from the Brazilian agencies Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ) and Departamento de Ciência e Tecnologia, Ministério da Saúde (DECIT); E.A. receives a grant from the European and Developing Countries Clinical Trial Partnership (EDCTP), Swedish Research Council and Swedish International Development Agency (SIDA); Y.S. receives grants from the Japan Agency for Medical Research and Development and JSPS KAKENHI; A.S. receives grant support from the National Health and Medical Research Council of Australia. None of the authors had any financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years, and no other relationships or activities that could appear to have influenced the submitted work.

Suarez‐Kurtz, G. , Aklillu, E. , Saito, Y. , and Somogyi, A. A. (2019) Conference report: pharmacogenomics in special populations at WCP2018. Br J Clin Pharmacol, 85: 467–475. 10.1111/bcp.13828.