10.1177/0091270004268128ARTICLE
`
`
`
`
`
`KIM ET ALDIFFERENCES IN DRUG PK BETWEEN EAST ASIANS AND CAUCASIANSPHARMACOGENETICS
`
`PHARMACOGENETICS
`
`Differences in Drug Pharmacokinetics
`Between East Asians and Caucasians and
`the Role of Genetic Polymorphisms
`
`Kiman Kim, Julie A. Johnson, and Hartmut Derendorf
`
`Interethnic variability in pharmacokinetics can cause unex-
`pected outcomes such as therapeutic failure, adverse effects,
`and toxicity in subjects of different ethnic origin undergoing
`medical treatment. It is important to realize that both genetic
`and environmental factors can lead to these differences
`among ethnic groups. The International Conference on Har-
`monization (ICH) published a guidance to facilitate the regis-
`tration of drugs among ICH regions (European Union, Japan,
`the United States) by recommending a framework for evaluat-
`ing the impact of ethnic factors on a drug’s effect, as well as its
`efficacy and safety at a particular dosage and dosage regi-
`men. This review focuses on the pharmacokinetic differences
`between East Asians and Caucasians. Differences in metabo-
`
`lism between East Asians and Caucasians are common, espe-
`cially in the activity of several phase I enzymes such as
`CYP2D6 and the CYP2C subfamily. Before drug therapy,
`identification of either the genotype and/or the phenotype for
`these enzymes may be of therapeutic value, particularly for
`drugs with a narrow therapeutic index. Furthermore, these
`differences are relevant for international drug approval when
`regulatory agencies must decide if they accept results from
`clinical trials performed in other parts of the world.
`
`Keywords: Interethnic variability; East Asians; Caucasians;
`pharmacokinetics; genetic polymorphisms
`Journal of Clinical Pharmacology, 2004;44:1083-1105
`©2004 the American College of Clinical Pharmacology
`
`It is generally recognized that most drugs show wide
`
`intersubject and interethnic variability in their dis-
`position and efficacy. This variability varies substan-
`tially among drugs and depends on a variety of factors.
`Potential factors for the variability in drug pharma-
`cokinetics and pharmacodynamics include food, drug
`interactions, body weight, gender, age, genetics, dis-
`ease states of patients, and lifestyle variables (smoking
`and alcohol consumption), which may correlate with
`one another. Among these, various environmental fac-
`tors (diet, nutrition, climate, lifestyle, and health) and
`cultural differences in attitudes toward the disorder are
`considered to be very substantial.1-11 Climate influ-
`ences food product and, therefore, is a factor determin-
`ing the ethnic features of nutrition. Factors such as diet
`and Chinese medicinal herbs can cause the different
`expression of drug-metabolizing enzymes, enzyme in-
`
`From the Department of Pharmaceutics, University of Florida, Gainesville,
`Florida. Submitted for publication February 27, 2004; revised version ac-
`cepted June 11, 2004. Address for reprints: Hartmut Derendorf, PhD, FCP,
`100494, Department of Pharmaceutics, University of Florida, Gainesville,
`FL 32610.
`DOI: 10.1177/0091270004268128
`
`duction, and enzyme inhibition.12,13 Thus, environ-
`mental factors might be considerably responsible for
`the population differences in the disposition and
`response of many drugs.
`The genetic factor is also one of the most potent de-
`terminants of drug response.14 Variation in genotype
`for drug-metabolizing enzymes, drug receptors, and
`drug transporters is associated with interindividual
`and interethnic variation in drug response.15-21 The
`most common situations causing these variabilities are
`differences in drug metabolism.22-27 Both genetic and
`environmental factors can lead to ethnic differences in
`drug metabolism to a varying extent, depending on the
`ethnic groups and substrates. These ethnic differences
`in the drug’s safety, efficacy, dosage, and dosage regi-
`men have given rise to a reluctance to rely on foreign
`clinical data for drug approval. Therefore, require-
`ments for the extensive duplication of a clinical evalu-
`ation for every drug can waste much time and expense
`during the new drug approval process in other regions.
`Recently, regulatory authorities and industry associa-
`tions have made efforts to promote international
`harmonization of regulatory requirements.
`
`J Clin Pharmacol 2004;44:1083-1105
`
`1083
`
`Novo Nordisk Exhibit 2066
`Mylan Pharms. Inc. v. Novo Nordisk A/S
`IPR2023-00724
`Page 00001
`
`

`

`KIM ET AL
`
`The International Conference on Harmonization
`(ICH) published the “Guidance on Ethnic Factors in the
`Acceptability of Foreign Clinical Data” in 1998.28 The
`purpose of this guidance is to facilitate the registration
`of drugs among ICH regions (European Union, Japan,
`the United States) by recommending a framework for
`evaluating the impact of ethnic factors on a drug’s effect
`(ie, its efficacy and safety at a particular dosage and
`dosage regimen). This guidance is intended to recom-
`mend regulatory strategies for accepting foreign clini-
`cal data as full or partial support for approval of an ap-
`plication in a new region and also the use of bridging
`studies, when necessary, to allow extrapolation of for-
`eign clinical data to a new region. One of the most im-
`portant determinants in extrapolating clinical data
`from one region to another involves ethnic differences
`in the pharmacokinetics of the drug.
`All drugs during the development process could be
`classified into 2 groups, ethnically sensitive or insensi-
`tive. Especially for drugs with narrow therapeutic indi-
`ces, the ethnic differences in pharmacokinetics should
`be carefully evaluated. The ICH guideline suggested
`certain factors that were likely to make a drug’s
`pharmacokinetics ethnically sensitive. Those include
`such factors as nonlinear pharmacokinetics, a steep
`dose-response curve in the range of the recommended
`dosage and dosage regimen, a narrow therapeutic dos-
`age range, rapid metabolism, metabolism by enzymes
`showing genetic polymorphism, administration as a
`prodrug, high intersubject variation in bioavailability,
`and low bioavailability.
`Thus far, there have been many papers and reviews
`on the effects of ethnicity on the pharmacokinetics and
`pharmacodynamics of a variety of drugs.4,29-40 The aim
`of this work is to review the differences between East
`Asians (Chinese, Koreans, and Japanese) and Cauca-
`sians with respect to pharmacokinetic profiles of a
`variety of drugs. Because there are limited data show-
`ing the differences between East Asians and Cauca-
`sians in the other pharmacokinetic phases such as ab-
`sorption, distribution, and excretion, these will only
`be covered briefly. This review focuses on cytochrome
`P450 enzyme-dependent metabolism influenced by
`genetic factors, which has been studied the most and is
`an area in which ethnic influence is characterized best.
`
`ABSORPTION/BIOAVAILABILITY
`
`Oral drug administration is the most common and pre-
`dominant route of drug administration. In most cases,
`drugs permeate across the gastrointestinal (GI) epithe-
`lium through passive diffusion. Alternatively, some
`drugs are absorbed through facilitated diffusion or ac-
`
`1084 • J Clin Pharmacol 2004;44:1083-1105
`
`tive transport. There are numerous factors affecting
`oral drug bioavailability, delivery to the intestine (gas-
`tric emptying, pH, food), absorption from the lumen
`(dissolution, lipophilicity, particle size, active uptake),
`intestinal metabolism (phase I and/or phase II en-
`zymes), active extrusion (drug efflux pumps), and first-
`pass hepatic extraction.
`P-glycoprotein (P-gp), the MDR1 (human multidrug
`resistance) gene product, is believed to decrease the
`bioavailability of many drugs and also affect the distri-
`bution of drugs in the body after they are absorbed.41-50
`P-gp, an adenosine triphosphate (ATP)–dependent
`drug efflux pump, decreases the intracellular concen-
`trations of drugs and their metabolites by pumping
`them out of the cells against a concentration gradient
`into the intestine. MDR1 is polymorphic, and numer-
`ous single-nucleotide polymorphisms (SNPs) have
`been identified.49 Allelic variants, especially 2 synony-
`mous SNPs (C1236T in exon 12 and C3435T in exon
`26) and a nonsynonymous SNP (G2677T, Ala893Ser)
`in exon 21, are likely to be associated with P-gp func-
`tion leading to the variabilities in pharmacokinetics
`and/or pharmacodynamics for the drugs.51-59 The
`cytochrome P450 (CYP) 3A subfamily (CYP3A4/5) ac-
`counts for most of the total CYPs found in the intestine
`and significant presystemic metabolism of some drugs
`in the intestine.60-66 Interindividual variability in its ex-
`pression is more than 20-fold high. Both P-gp and the
`CYP3A subfamily have a significant substrate overlap
`in their substrate specificities and share a location in
`the small intestinal enterocytes.67-69 Also, the absorp-
`tion of the CYP3A subfamily and/or P-gp substrates in-
`creases when they are orally coadministered with
`grapefruit juice.13,70-93 Grapefruit juice contains high
`concentrations of flavonoids and furanocoumarins,
`which inhibit CYP3A in the intestine and may affect
`P-gp-mediated transport.
`Drugs undergoing active transport, efflux by P-gp,
`and gut metabolism via CYP3A are most likely to show
`ethnic differences in drug bioavailability.39,94,95 For ex-
`ample, people of African origin can be expected to have
`higher plasma concentrations for P-gp substrates than
`other ethnic groups, such as East Asians, Indians, and
`Caucasians, due to their lower T allele frequency in
`exon 26 of P-gp.55,96,97 The frequency of homozygotes
`for the T allele in exon 12 in Caucasians (13.3%) is
`much lower than in Japanese (37.5%),98 and there is
`also a significant difference in the allele frequencies in
`exon 26 between Indians and German Caucasians.99
`Considering the differences in drug absorption of sev-
`eral drugs between African Americans and Cauca-
`sians100-104 and MDR1 allelic variants between Japa-
`nese and Caucasians, these may indicate differences
`
` 15524604, 2004, 10, Downloaded from https://accp1.onlinelibrary.wiley.com/doi/10.1177/0091270004268128 by Sayem Osman , Wiley Online Library on [04/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
`
`Novo Nordisk Exhibit 2066
`Mylan Pharms. Inc. v. Novo Nordisk A/S
`IPR2023-00724
`Page 00002
`
`

`

`DIFFERENCES IN DRUG PK BETWEEN EAST ASIANS AND CAUCASIANS
`
`between East Asians and Caucasians in drug absorp-
`tion. However, there are no clear studies showing vari-
`ations in drug absorption between East Asians and
`Caucasians. The interrelationship between P-gp and
`CYP3A, as well as a highly variable expression of them
`among individuals, along with potential food interac-
`tions, may make it difficult to elucidate the ethnic dif-
`ferences in drug absorption. Hence, further research
`about the ethnic differences of active transport, the P-
`gp efflux mechanism, and CYP3A metabolism in the GI
`tract is needed.
`
`DISTRIBUTION
`
`Plasma protein binding is one of the most crucial fac-
`tors affecting drug pharmacokinetics and efficacy. Al-
`bumin and α
`1-acid glycoprotein (AGP) are major drug-
`binding proteins.105 The determinants of protein bind-
`ing are protein concentration, the number of binding
`sites per unit of protein, and binding affinity.94,95,106
`The free plasma fractions of propranolol, diso-
`pyramide, and diphenhydramine that bind to AGP
`were approximately 17%, 26%, and 44% higher, re-
`spectively, in Chinese who resided in the United States
`from 1 month to 4 years than in American Cauca-
`sians.107 For these drugs with a basic character, binding
`to AGP was generally more extensive than that to albu-
`min because basic drugs had a higher affinity to AGP
`than acidic drugs.108,109 The AGP plasma concentration
`was about 25% higher in American Caucasians,
`whereas the albumin level was similar in the 2
`groups.107 Zhou et al110 reported that the unbound frac-
`tion of orally administered propranolol in plasma was
`significantly higher in Chinese who lived in the United
`States for an average of 1.6 years (16.7% ± 1.6%) com-
`pared to American Caucasians (11.5% ± 0.5%). An-
`other study showed that the unbound plasma fraction
`of diphenhydramine was significantly higher in East
`Asians (24.0% ± 1.9%) than in Caucasians (14.8% ±
`1.5%), which probably led to the increased volume of
`distribution in East Asians.111
`Reboxetine, a specific norepinephrine reuptake in-
`hibitor, is administered as a racemic compound. The
`(S,S)-(+)-enantiomer is more potent in the in vivo assay
`of antidepressant effects than the (R,R)-(–)-form.112 The
`drug undergoes extensive hepatic metabolism, and
`more than 97% is bound to plasma proteins, mainly to
`AGP.112,113 Hendershot et al114 noted that Chinese Amer-
`icans exhibited greater free concentrations for (S,S)-
`(+)-reboxetine than American Caucasians after oral ad-
`ministration. The free plasma fraction for (S,S)-(+)-
`reboxetine was about 53% greater, and the unbound
`
`clearance (CLu) was about 29% lower in Chinese Amer-
`icans than in American Caucasians. Also, the unbound
`area under the curve extrapolated to infinity (AUC0-∞)
`was significantly greater in Chinese Americans (20.2 ±
`7.1 ng(cid:127)h/mL) compared to American Caucasians (13.2
`± 3.2 ng(cid:127)h/mL).
`Drugs that bind to AGP and/or albumin may show
`ethnic differences in plasma protein binding. The re-
`sult that Chinese have higher unbound fractions of
`drugs with a basic character compared to American
`Caucasians is consistent with the lower AGP concen-
`tration in Chinese.107 The ethnic difference in plasma
`protein binding of drugs to AGP appears to be common,
`indicating that Caucasians show higher binding than
`other ethnic groups.94,95,106 These differences are likely
`to be caused by the differences in protein concentra-
`tions, not by the differences in the number of binding
`sites on the protein or in affinity. Therefore, the lower
`AGP concentrations in East Asians would contribute to
`the greater free fractions of AGP-bound drugs com-
`pared to Caucasians. However, these differences do not
`seem to be clinically significant because usually, phar-
`macologically relevant unbound concentrations do not
`change with changes in protein binding.
`
`METABOLISM
`
`Among the 4 pharmacokinetic phases (absorption, dis-
`tribution, metabolism, and excretion), the variability in
`metabolism is most clearly associated with interindi-
`vidual and interethnic variation in drug pharma-
`cokinetics. These pharmacokinetic differences may or
`may not translate into the differences in efficacy or tox-
`icity. Those who display significantly slower rates of
`metabolism of a certain drug than the extensive
`metabolizers (EMs) can be classified as poor
`metabolizers (PMs). Polymorphisms in genes encoding
`particular enzymes may lead to absent or varied enzy-
`matic activity. Genetic polymorphism is defined as a
`monogenic trait in which allelic variation occurs at a
`single gene locus with a frequency of at least 1% in the
`population.115 The genetic variation among ethnic
`groups in enzyme activities is very important due to
`the expression of polymorphic enzymes, and it may ex-
`plain the variability in a trait. Therefore, pharma-
`cokinetics and drug response would be affected by eth-
`nicity if the enzyme activity was polymorphically dis-
`tributed and the frequency of the phenotype differed
`among ethnic groups. However, in about 50% of all
`drug therapies, the genetic influence on drug therapy
`might be insignificant because of gene products with-
`out any significant functional polymorphism and pre-
`dominant effect of environmental factors over it.14
`
`PHARMACOGENETICS
`
`1085
`
` 15524604, 2004, 10, Downloaded from https://accp1.onlinelibrary.wiley.com/doi/10.1177/0091270004268128 by Sayem Osman , Wiley Online Library on [04/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
`
`Novo Nordisk Exhibit 2066
`Mylan Pharms. Inc. v. Novo Nordisk A/S
`IPR2023-00724
`Page 00003
`
`

`

`KIM ET AL
`
`Most drugs are primarily metabolized by
`microsomal enzymes (especially CYPs) localized in
`the liver and, usually to a smaller extent, in the small
`intestine. Hepatic metabolism partly regulated by ge-
`netic factors is a major determinant for drug response.
`CYPs are responsible for oxidative, peroxidative, and
`reductive metabolic transformations of drugs.116 CYPs
`usually transform lipophilic drugs to more hydrophilic
`compounds, which are mostly prone to phase II metab-
`olism (conjugation reactions). Although many other
`enzymes, including phase II enzymes, also show eth-
`nic differences, most information is available on the
`differences in CYPs such as CYP2D6, CYP2C, CYP3A,
`CYP1A2, and CYP2A6, with sometimes substantial
`differences between East Asians and Caucasians.
`
`CYP2D6
`
`CYP2D6 catalyzes the metabolism of clinically impor-
`tant drugs that are generally lipophilic, such as cardio-
`vascular substances, neuroactive substances, and en-
`dogenous compounds.117-124 It is expressed at about 3%
`of the total CYPs in the liver.116 Debrisoquine,
`sparteine, metoprolol, and dextromethorphan are used
`as probe drugs in population phenotyping studies.
`This enzyme is highly polymorphic.125 So far, approxi-
`mately 80 CYP2D6 variants have been identified.126
`CYP2D6*2 has the same activity as the reference but
`a tendency to duplicate or amplify.127-130 Ultra-rapid
`metabolizers (URs) have been found to have
`haplotypes with 2 or more functional CYP2D6 genes of
`CYP2D6*2.131 Individuals carrying CYP2D6*2 may
`have very high enzymatic activity and therefore have to
`take unusually high doses of drugs metabolized by this
`enzyme to maintain therapeutic concentrations. Their
`metabolic ratio (MR: parent drug/metabolite) values
`are very low compared to the EMs.132
`Although functional, CYP2D6*10 is an unstable
`variant with decreased catalytic activity.133-143 There are
`2 CYP2D6*10 haplotypes in East Asians, CYP2D6*10A
`and CYP2D6*10B. 1 2 7 , 1 3 3 , 1 3 4 , 1 4 4 Homozygous
`CYP2D6*10A Chinese had higher AUC values, maxi-
`mum plasma concentration (Cmax) values, and lower
`clearances of metoprolol enantiomers than either the
`heterozygous CYP2D6*10A or the reference.145 Korean
`CYP2D6 EMs with either homozygous or heterozygous
`CYP2D6*10B had a significantly lower clearance of
`paroxetine, an antidepressant, than the EMs.146
`A significantly higher AUC and Cmax of orally admin-
`istered propranolol and a substantially lower AUC of 4-
`hydroxypropranolol were revealed in homozygous
`CYP2D6*10 Chinese than in either the reference or the
`heterozygous CYP2D6*10 because 4-hydroxylation of
`
`1086 • J Clin Pharmacol 2004;44:1083-1105
`
`the drug was catalyzed by CYP2D6.135 The oral clear-
`ance of venlafaxine, an antidepressant, was about 4
`times higher in CYP2D6 EMs compared to the PMs.147
`Its disposition in Japanese was also affected by the
`CYP2D6*10 genotype after oral administration.138,140
`The Cmax and AUC0-24 of venlafaxine were about 4 and
`5.5 times higher, respectively, in homozygous
`CYP2D6*10 Japanese than in the EMs. Also, those were
`about 1.9 and 2.2 times higher in heterozygous
`CYP2D6*10 Japanese than in the EMs.140
`The pharmacokinetics of both nortriptyline and its
`metabolite (10-hydroxynortriptyline) were dependent
`on CYP2D6 genotypes. 1 3 2 , 1 4 8 - 1 5 0 Homozygous
`CYP2D6*10 Chinese showed doubled half-life (t1/2)
`values, doubled AUC0-∞ values, and halved oral plasma
`clearances compared to the homozygous reference.149
`Also, homozygous CYP2D6*10 Chinese had a 60%
`higher AUC0-∞ of nortriptyline than the heterozygous
`CYP2D6*10. Both nortriptyline clearance and the AUC
`ratio (nortriptyline to its metabolite) correlated signifi-
`cantly with the CYP2D6 MR,132 like debrisoquine.151
`The plasma t1/2 and the AUC0-∞ of nortriptyline de-
`creased with the increases in the number of the
`CYP2D6 functional variant.132 Also, nortriptyline dis-
`positions in Caucasian heterozygous EMs were very
`similar to those in the PMs, with the MR values greater
`than 1. Kishimoto et al152 reported that nortriptyline
`Cmax was about 23% higher and the AUC was about
`58% greater in the Japanese subjects, even though the
`Japanese received only half the dose that American
`Caucasians had received (50 mg, 100 mg, respectively).
`Table I provides the differences in various pharma-
`cokinetic parameters of CYP2D6 substrates between
`East Asians and Caucasians.
`Haloperidol is biotransformed to reduced
`haloperidol through reduction, which undergoes oxi-
`dation back to haloperidol.153 CYP2D6 seems to be in-
`volved in their metabolism. The plasma concentrations
`of both haloperidol and reduced haloperidol were sig-
`nificantly higher in Caucasian CYP2D6 PMs than in the
`EMs after a single dose of haloperidol.154,155 Moreover,
`after multiple dosing, the mean steady-state concentra-
`tions (Css) of them were significantly higher in Japanese
`with either 1 or 2 CYP2D6*10 than in those without
`this variant.156,157 The Css of haloperidol in Koreans was
`significantly different among CYP2D6 genotype
`groups when doses lower than 20 mg were given, but
`no differences were observed at higher doses.158
`Asians require less neuroleptic medication to
`achieve a therapeutic response than Caucasians.8,159-161
`Asians are known to have significantly lower target
`concentrations and neuroleptic threshold haloperidol
`concentrations when the doses are determined based
`
` 15524604, 2004, 10, Downloaded from https://accp1.onlinelibrary.wiley.com/doi/10.1177/0091270004268128 by Sayem Osman , Wiley Online Library on [04/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
`
`Novo Nordisk Exhibit 2066
`Mylan Pharms. Inc. v. Novo Nordisk A/S
`IPR2023-00724
`Page 00004
`
`

`

`DIFFERENCES IN DRUG PK BETWEEN EAST ASIANS AND CAUCASIANS
`
`Table I Differences in Various Pharmacokinetic
`Parameters of CYP2D6 Substrates Between East Asians and Caucasians
`
`Consequencesa
`
`23%, 58% higher in Japanese having received only half the dose than in
`American Caucasians
`More than 2-fold higher in American-born and foreign-born Asians than
`in American Caucasians after oral administration
`87% higher in Asian males than in Canadian Caucasian males
`26%, 64%, 52%, and 73% higher in Chinese than in Swedish Caucasians
`
`Reference
`
`152
`
`165
`
`166
`170
`
`171
`171
`167
`174
`175
`178
`
`Substrate
`
`Parameters
`
`Nortriptyline
`
`Cmax, AUC
`
`Haloperidol
`
`
`
`bCmaxb, Cp
`
`
`
`Codeine
`
`Debrisoquine
`Desipramine
`Clomipramine
`
`Cp
`t1/2, AUC0-∞,
`Cmax, UR
`b
`40% lower in Chinese than in Swedish Caucasians
`CLo
`63% higher in Chinese than in Caucasians 33%
`AUC0-∞
`b
`33%, 61% lower in Chinese than in Caucasians
`CLo, CLm
`67% higher in Chinese than in Swedish Caucasians
`UR
`75% higher in Chinese than in Canadian Caucasians
`UR
`40% lower in Chinese than in American Caucasians
`CLt
`More than 70% higher in Japanese than in Swedish Caucasians
`Cp
`b
`More than 70% higher in Japanese than in Swedish Caucasians
`CLo
`AUC, area under the plasma concentration-time curve; AUC0-∞, AUC from 0 to infinity; Cmax, maximum plasma concentration; Cp, plasma concentration; CLo,
`oral clearance; CLt, total clearance; t1/2, half-life; UR, urinary recovery; CLm, partial metabolic clearance.
`a. Statistically significant.
`b. Normalized for body weight or body surface area.
`
`on clinical response.162-164 Lin et al165 noted that both
`American-born and foreign-born Asians (East Asian
`and Philippine origin) had significantly higher Cmax
`values and serum concentrations than American Cau-
`casians following oral and intramuscular haloperidol
`administration. Even after controlling the body surface
`area, significant ethnic differences in plasma con-
`centrations following oral administration existed.
`The average and the maximal prolactin concentra-
`tions following intramuscular administration were
`significantly lower in American Caucasians. Another
`study indicated that Asian males (East Asian and Viet-
`namese origin) had significantly higher plasma
`haloperidol concentrations than Canadian Caucasian
`males at 2 mg/day.166
`Codeine is mainly metabolized via glucuronidation
`to codeine-6-glucuronide, O-demethylation to mor-
`phine, and N-demethylation to nor-codeine.167 There
`were pronounced differences in the formation of its
`metabolites between Swedish Caucasian CYP2D6 EMs
`and PMs, despite the fact that both codeine AUC0-∞ and
`plasma clearance were similar.168 The PMs had very
`low or undetectable concentrations of morphine. The
`partial clearance of morphine formation also showed a
`significant decreasing trend among homozygous
`CYP2D6*10, heterozygous CYP2D6*10, and reference
`Chinese.169
`Remarkable ethnic differences were shown in co-
`deine metabolism.167,169-171 Chinese produced less mor-
`
`phine from codeine and exhibited reduced sensitivity
`to the pharmacologic effects of morphine because the
`production of morphine from codeine was mediated
`through CYP2D6.169 The respiratory depressant effect
`of codeine was lower in Chinese living in the United
`States for an average of 3.1 years compared to Cauca-
`sians.171 This lower effect was mainly due to the re-
`duced production of morphine and morphine-6-
`glucuronide, an active metabolite. Apparent codeine
`clearance and its partial metabolic clearance by O-
`demethylation were significantly greater in Caucasians
`than in Chinese. Even after weight correction, the par-
`tial metabolic clearance through O-demethylation was
`still about 2.5 times greater in Caucasians. After a sin-
`gle oral 50-mg dose of codeine phosphate, Chinese
`staying in Sweden for less than 5 years had about 52%
`higher Cmax and about 64% larger AUC0-∞ compared to
`Swedish Caucasians; thus, plasma clearance was lower
`in Chinese (1.26 ± 0.22 vs 2.10 ± 0.84 L/h(cid:127)kg).170 The
`frequency distribution of the MR values for metabolic
`pathways was shifted toward higher values in Chi-
`nese.167 Chinese had longer plasma t1/2 values of co-
`deine and its metabolite, as well as significantly higher
`urinary recovery of unchanged codeine, compared to
`Swedish Caucasians.167,170
`Slight differences were observed in codeine metabo-
`lism among East Asians living in Sweden.172 Chinese
`metabolized codeine less than both Japanese and Kore-
`ans, whereas the latter 2 groups did not differ in the
`
`PHARMACOGENETICS
`
`1087
`
` 15524604, 2004, 10, Downloaded from https://accp1.onlinelibrary.wiley.com/doi/10.1177/0091270004268128 by Sayem Osman , Wiley Online Library on [04/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
`
`Novo Nordisk Exhibit 2066
`Mylan Pharms. Inc. v. Novo Nordisk A/S
`IPR2023-00724
`Page 00005
`
`

`

`KIM ET AL
`
`overall metabolic pattern of codeine.172 In addition,
`Korean/Japanese CYP2D6 EMs had a higher median
`enzyme activity than Chinese EMs assessed using
`metoprolol.173 Korean/Japanese EMs excreted more
`metabolite (α-hydroxymetoprolol) than metoprolol,
`whereas Chinese EMs excreted more metoprolol than
`the metabolite.
`Chinese students in Canada excreted significantly
`more orally administered debrisoquine than Canadian
`Caucasians.174 Rudorfer et al175 found that oral
`desipramine clearance was about 67% higher in Amer-
`ican Caucasians than in Chinese studying in the United
`States for average 4.2 years, whereas the plasma con-
`centrations of both desipramine and its metabolite
`were significantly higher in Chinese than in American
`Caucasians at all times. The difference in clearances
`was unchanged even after weight correction.
`The 2- and 8-hydroxylation of clomipramine are cat-
`alyzed by CYP2D6, and the N-demethylation is cata-
`lyzed by several enzymes, including CYP2C19.176,177
`Shimoda et al178 found that the plasma concentrations
`of clomipramine were more than 70% higher in Japa-
`nese than in Swedish patients, although the average
`daily dose for Japanese was lower than that for Swed-
`ish Caucasians by about 6%. The lower oral clearance
`of clomipramine in Japanese was not influenced by the
`lower body weight of Japanese or the concomitant
`treatment with benzodiazepines.
`The occurrence of CYP2D6 PMs is higher in Cauca-
`sians (5%-10%) than in East Asians (about 1%) (Table
`II).127,179-185 CYP2D6*3, CYP2D6*4, CYP2D6*5, and
`CYP2D6*6 are known to be the most common non-
`functional variants in Caucasians, accounting for more
`than 95% of the PMs.137,185 Although the frequency of
`CYP2D6*5 is similar in East Asians and Caucasians,
`CYP2D6*4 is almost absent in East Asians. However,
`CYP2D6 activity is lower in Chinese EMs compared to
`Caucasian EMs. The MR distribution of Chinese
`CYP2D6 EMs was shifted to the right compared to
`Swedish Caucasian CYP2D6 EMs, using debrisoquine
`as a probe drug.183 Most Swedish Caucasians had an
`MR < 1, whereas more than half of the Chinese had
`an MR > 1. Despite the low frequency of the PMs,
`this right shift in the MR in East Asians is due to
`CYP2D6*10.144,186-188 CYP2D6*10 exists at a higher fre-
`quency in East Asians (30%-50%) compared to Cauca-
`sians (about 5%), leading to the higher median of MR
`values in East Asians.141,189-191 Three ethnic-specific
`variants are known: CYP2D6*4 in Caucasians,
`CYP2D6*10 in East Asians, and CYP2D6*17 in Afri-
`cans.124 About 1% to 10% of Caucasians have 1 or more
`extra active CYP2D6 genes causing ultra-rapid metabo-
`lism of CYP2D6 substrates.130,137,192-194 In contrast to
`
`1088 • J Clin Pharmacol 2004;44:1083-1105
`
`CYP2D6*10, no duplicated or amplified haplotypes of
`the CYP2D6*2 gene were found in East Asians.187
`Therefore, despite the lower incidence of CYP2D6
`PMs, the higher occurrence of CYP2D6*10 and the al-
`most absent frequency of CYP2D6*2 cause an overall
`lower CYP2D6 activity in East Asians compared to
`Caucasians. This fact probably accounts for why East
`Asians are more sensitive to the drugs mainly metabo-
`lized by CYP2D6 and require lower doses than Cauca-
`sians for optimal treatment.
`The significant difference in the prevalence of
`CYP2D6*1 and CYP2D6*10 might lead to the differ-
`ence in CYP2D6 activity among East Asians.173,195
`CYP2D6*1 frequency in both Koreans and Japanese
`was significantly higher than in Chinese, whereas
`CYP2D6*10 was significantly lower in Koreans/
`Japanese. Koreans appear to be similar to Japanese but
`quite distinct from Chinese.173 Therefore, significant
`genetic differences may exist among East Asians, and
`the consequences of 1 ethnic group cannot be directly
`applied to another ethnic group, even though they are
`geographically near.195
`
`CYP2C Subfamily
`
`In human liver microsomes, the CYP2C subfamily is
`second in quantity only to the CYP3A subfamily and
`consists of 4 members: CYP2C8, CYP2C9, CYP2C18,
`and CYP2C19.116,196-198 It accounts for about 20% of the
`total CYPs in the liver. Among these, CYP2C19 and
`CYP2C9 are the predominant enzymes. All members of
`this subfamily exhibit genetic polymorphism.
`
`CYP2C19
`CYP2C19 is mainly present in the liver, but significant
`activity has also been identified in the gut wall.199
`Anticonvulsant mephenytoin is used as a model drug
`for CYP2C19.200 CYP2C19 plays an important role in
`the metabolism of many drugs such as benzo-
`diazepines, phenytoin, some barbiturates, tricyclic an-
`tidepressants, and proton pump inhibitors, which may
`be neutral, weak bases or weak acids.201-210
`At least 19 variants of CYP2C19 have been identi-
`fied. 1 2 6 Two null variants, CYP2C19*2 and
`CYP2C19*3, are responsible for the majority of
`CYP2C19 PMs.211,212 CYP2C19 EMs consist of both ho-
`mozygous and heterozygous genotypes, and their fre-
`quency in the population depends on ethnicity.206 This
`genetic effect is likely to be one of the reasons for the
`large interindividual and interethnic variation in
`CYP2C19 activity within the EMs.
`Plasma concentrations of omeprazole after applica-
`tion of a standard dose significantly depend on
`
` 15524604, 2004, 10, Downloaded from https://accp1.onlinelibrary.wiley.com/doi/10.1177/0091270004268128 by Sayem Osman , Wiley Online Library on [04/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
`
`Novo Nordisk Exhibit 2066
`Mylan Pharms. Inc. v. Novo Nordisk A/S
`IPR2023-00724
`Page 00006
`
`

`

`DIFFERENCES IN DRUG PK BETWEEN EAST ASIANS AND CAUCASIANS
`
`Table II Differences Between Poor Metabolizers (PMs) and
`Potent Variant Frequencies in Various CYP Enzymes in East Asians and Caucasians
`
`PM Frequencies (%)
`
`Variant Frequencies (%)
`
`East A