Adverse drug reactions
`
`Pharmacogenetics and adverse drug reactions
`
`ADVERSE DRUG REACTIONS
`
`Urs A Meyer
`
`Polymorphisms in the genes that code for drug-metabolising enzymes, drug transporters, drug receptors, and ion
`channels can affect an individual’s risk of having an adverse drug reaction, or can alter the efficacy of drug
`treatment in that individual. Mutant alleles at a single gene locus are the best studied individual risk factors for
`adverse drug reactions, and include many genes coding for drug-metabolising enzymes. These genetic
`polymorphisms of drug metabolism produce the phenotypes of “poor metabolisers” or “ultrarapid metabolisers” of
`numerous drugs. Together, such phenotypes make up a substantial proportion of the population. Pharmacogenomic
`techniques allow efficient analysis of these risk factors, and genotyping tests have the potential to optimise drug
`mtmm
`
`The person—to—person variability of a drug response is a
`major problem in
`clinical practice
`and in drug
`development.
`It can lead to therapeutic failure or
`adverse effects of drugs in individuals or subpopulations
`of patients. The occurrence of serious or fatal adverse
`drug reactions has been extensively analysed in hospital
`inpatients. A meta—analysis of 39 prospective studies
`from US hospitals suggests that 6-7% of inpatients have
`seriousadvepsedrugreaetionsand0-3°°(aha—vefatal
`reactions, the latter causing about 100000 deaths per
`year
`in the USA. This figure makes adverse drug
`reactions between the fourth and sixth leading causes of
`death in hospital inpatients.‘
`What determines an individual’s risk of developing an
`adverse drug reaction or of gaining no benefit from the
`drug? How much of this pharmacological variability can
`be predicted? How many adverse drug reactions can
`thereby be prevented? These are some of the questions
`addressed in this review.
`
`History of pharmacogenetics
`Pharmacogenetics had its beginning in the 1950s when
`researchers realised that some adverse drug reactions
`could be caused by genetically determined variations in
`enzyme
`activity. For
`example,
`prolonged muscle
`relaxation after suxamethonium was explained by an
`inherited deficiency of a plasma cholinesterase, and
`haemolysis caused by antimalarials was recognised as
`I.
`11.1.!
`E16
`phosphate dehydrogenase. Similarly, inherited changes
`in a patient’s ability to acetylate isoniazid was found to
`be the cause of the peripheral neuropathy caused by this
`drug. Nlore recently, adverse drug reactions such as
`nausea,
`diplopia,
`and
`blurred
`vision
`after
`the
`antiarrhythmic
`and
`oxytocic
`drug
`sparteine,
`or
`incapacitating
`orthostatic
`hypotension
`after
`the
`antihypertensive
`agent
`debrisoquine
`have
`led
`to
`the discovery of
`the genetic polymorphism of
`the
`drug—n1etabolising
`enzyme Cytochrome P450
`2D6
`(C¥P%D6).7 Adverse drug reactions were
`also the
`clinical events that revealed genetic variants of other
`drug—metabolising enzymes or drug targets, a selection
`of which is
`listed in table
`l.“*’” Thus, genetic
`polymorphisms
`were
`discovered
`by
`incidental
`observations
`that
`some
`patients
`or
`volunteers
`experienced unpleasant and disturbing adverse drug
`reactions when given standard doses of drugs.
`
`Genotype and phenotype
`Molecular genetics and genomics (the study of the entire
`. '
`I
`L" .
`V"
`.
`" I
`' '
`'
`.5
`
`in the past decade. The two alleles carried by an
`individual at a given gene locus,
`referred to as the
`genotype, can now be characterised at the DNA level;
`their influence on the kinetics of the drug or a receptor
`function, the phenotype, can be measured by advanced
`analytical methods
`for metabolite detection or by
`sophisticated
`clinical
`investigations—eg,
`receptor-
`density studies by positron emission tomography.
`Molecular studies in pharmacogenetics started with the
`cloning and characterisation of CYP2D6,“’ and have
`now been extended to numerous other human genes,
`including those
`coding for more
`than 20 drug-
`metabolising enzymes and drug receptors, and several
`drug transport
`systems
`(www.sciencemag.org/feature/
`data/104'-l449.shl).
`More than 70 variant alleles of the CYPZD6 locus have
`been
`described
`(www.imm.ki.se/CYPalleles/cyp2d6.
`htm), of which at least 15 encode non—functional gene
`
`Lancet 2000; 356: 1667-71
`
`Division of Pharmacology/Neurobiology, Biozentrum of the
`University of Basel, CH—4056 Basel, Switzerland (U A Meyer MD)
`(e-mail: Urs—A_Meyer@unibas_ch)
`
`THE LANCET - Vol 356 - November ll, 2000
`
`1667
`
`Vanda Exhibit 2035 - Page 1
`
`VNDA 02700304
`
`
`
`include drug—drug interactions, the patient’s age, renal
`and liver function or other disease factors, and lifestyle
`variables such as smoking and alcohol consumption.
`Of even greater importance in the determination of
`individual
`risk are inherited factors that affect
`the
`kinetics and dynamics of numerous drugs. Thus, genetic
`variation in genes
`for drug—metabolising enzymes,
`drug receptors,
`and drug transporters have been
`associated with individual variability in the efficacy
`and toxicity of drugs.“ It
`is of course difficult
`to
`
`in an individual patient. A major
`genetic factors
`difference between genetic and environmental variation
`is
`that
`an inherited mutation or
`trait
`is present
`throughout life and has to be tested for only once in a
`lifetime, whereas environmental effects are continually
`changing. If mutant or variant genes exist at a frequency
`of more than 1% in the normal population, they are
`called genetic polymorphisms? Genetic polymorphisms
`explain why a small proportion of the population may be
`at higher risk of drug inefficacy or toxicity; the study of
`such polymorphisms has given rise to the field of
`pharmacogenetics.”
`
`Vanda Exhibit 2035 - Page 1
`
`

`
`gene products involved in drug disposition and drug
`action. Polygenic inheritance is more difficult to detect
`and to distinguish from environmental
`factors. The
`analysis of polygenic inheritance is similar to the analysis
`of complex diseases such as cancer, mental diseases,
`arthritis, and asthma, in which primary genes, modifier
`genes and environmental factors interact.""”
`A simple method to distinguish between hereditary
`and environmental components of variability is
`the
`comparison of small series of monozygotic and dizygotic
`twins, or repeated drug administration and comparison
`of the variability of the responses within and between
`individuals as proposed by Kalow.“’ The techniques have
`revealed
`important
`genetic
`factors
`in
`the
`pharmacokinetics
`of
`drugs
`such
`as
`dicoumarol,
`halothane, phenytoin, tolbutamide, and midazolam.
`In the future, the discovery of pharmacogenetic traits
`
`Drug effect
`
`ADVERSE DRUG REACTIONS
`
`Enzyme
`
`CYP2C9
`
`CYPQDG
`
`Drug
`
`Frequency of
`polymorphism
`14-28% (heterozygotes) Wariarin
`O-2—:L%(homozygotes)
`Tolbutarride
`Phenytoln
`Glipizide
`Losartan
`
`/mtiarrlrythmics
`
`540% {poor
`metabolisers)
`1—10% {ultra-rapid
`metabolisers)
`
`Haemorrhagei
`Hypoglycaemiai
`Prienytoln toxicity”
`Hypoglycaeniia”
`Decreased
`arltihypertensive effect“
`Proarrhythmic and
`other toxic e"fects3
`Antidepressants Toxicity in poor
`rretabolisers, ireffrcaey
`in ultrarapid metabolisersi
`Tardive dysklnesiai
`Irefficacy of codeine as
`analgesic, narcotic side-
`effects. dependence‘
`Increased B—b|ocl<ade3
`
`Antipsychotios
`Opioids
`
`Badrenoceptor
`antagonists
`Omeprazole
`
`Diazepam
`Fluorouracrl
`
`Higher cure rates when
`given witr clarithrom;/cin1*
`Prolongec seclation3
`Neuroioxicity,
`rryelotoxicity“
`
`Succinylcholire
`
`Prolongec apnoea’
`
`Sulphonamides Hypersensitil/ity3
`Amonafide
`Mylelotoxrcrty (rapid
`acetylators)“
`Drcginduccc lupus
`erythematosus‘
`
`Prooalnamide,
`hydralazine,
`isoniazid
`Merccptopurinc,
`tniogtanine,
`azotnioprine
`Irinotecan
`
`Myelotoxicity“
`
`Diarrhoea,
`rryelosbppression“
`
`CYP2C19
`
`3-696 (whites)
`8—23% lAsrans)
`
`Dihydropyri-
`midine dehyclro-
`genase
`Plasma pseudo
`cholinesterase
`N-acety|trans-
`ferase
`
`0-1°/n
`
`1-5%
`
`40-70% (whites)
`10-20% (Asians)
`
`03%
`
`Thropurire
`methyltrans
`ferase
`UDP—gIucurorlosyI— 10—:L5%
`transierase
`
`Table 1: Clinically important genetic polymorphisms of drug
`metabolism that influence drug response
`
`products. These alleles are different from the normal
`(wild—type) gene by one or more point mutations, but
`
`
`
`
`
`genomics. Rapid sequencing and single—nucleotide
`polymorphisms (SNPS) will have a major role in the
`linking of sequence variations with heritable clinical
`phenotypes of drug response. SNPs occur in about one
`of every 100-1500 bp of the genomes if two unrelated
`individuals are compared. Any two individuals therefore
`differ by about 3 million bp—ie, in only about 0-1% of
`the 3 billion bp of the haploid genome. Common or
`informative SNPS are those that occur at frequencies of
`greater than 1%, or even better, of greater than 10%.
`Qneea—largenu~ml9eroftheseSN-Psaandtheir
`frequencies in different populations are known, they can
`be used to correlate a patient’s genetic “fingerprint” with
`the probable individual drug response.°~'*5 SNPs in coding
`regions of genes (about 30 000—l00 000 per genome)
`can cause aminoacid changes and changes in protein
`function, or they can be neutral. SNPS inside genes or in
`regulatory regions can cause differences
`in protein
`expression. The proteomes of two individuals can differ
`by only 30 000 proteins, and some of these differences
`might affect drug response. The ability to predict
`
`‘
`“ ‘
`‘
`‘
`““‘ “
`“‘ 6r 0fl
`the basis of genetic factors will
`thus be a realistic
`scenario for future drug treatments.
`
`multiduplications. The mutations can have no effect on
`enzyme activity, or can code for enzymes with decreased
`or absent activity; duplications lead to increased enzyme
`activity.
`Extensive metabolisers
`are
`defined
`as
`individuals homozygous or heterozygous for the wild-
`type or normal
`activity enzymes
`(75—85% of
`the
`population);
`intermediate metabolisers (10—l5%)” or
`poor metabolisers
`(5~l0%)
`are
`carriers
`of
`two
`decreased—activity alleles or loss—of—functi0n alleles; and
`ultrarapid metabolisers
`(l—10%)
`are
`carriers
`of
`
`II
`i
`‘I
`'
`I II i
`"'i
`V‘ 1‘ __
`-
`’Ii_
`the number and complexity of the mutations may be
`large overall, only a few mutant genes, usually three to
`five alleles, are common and account for most (usually
`>95%) of mutant alleles. These alleles can be detected
`by modern DNA methods,
`including DI\'A chip
`microarrays, allowing most patients to be assigned to a
`particular phenotype group.
`CYPZD6 remains one of the best studied polymorphic
`genes of pharmacological interest; numerous other genes
`have been or are presently being studied by identical
`approaches. The poor~metaboliser phenotype for most
`of the polymorphic genes is inherited as an autosomal
`recessive trait, requiring the presence of two mutant
`alleles. These considerations apply to monogenic traits,
`where one gene has a major effect on the phenotype and
`divides the population into two or three distinct groups.
`However,
`the responses to most medications are not
`monogenic but involve the interaction between multiple
`
`
`
`Population differences in pharmacogenetic
`traits
`All pharmacogenetic variations studied to date occur at
`different frequencies among subpopulations of different
`ethnic or racial origin. For instance, important cross-
`ethnic differences
`exist
`in the
`frequency of
`slow
`acetylators of
`isoniazid due
`to mutations of N-
`
`—
`.
`,
`-
`i
`3
`-
`.
`mutations of CYPZC9 and CYPZCI9, and
`metabolisers due to mutations of CYP2D6 (table 2).“
`Some of the mutations of these genes occur uniquely in
`certain ethnic subpopulations. This ethnic diversity, also
`called “gene geography”, implies that ethnic origin has
`to be considered in pharmacogenetic studies and in
`pharmacotherapy.
`
`Therapeutic issues
`In principle,
`three pharmacogenetic mechanisms can
`influence pharmacotherapy. The first and‘best studied to
`date are genetic polymorphisms of genes
`that are
`associated with altered metabolism of drugs
`(eg,
`metabolism of tricyclic antidepressants). Increased or
`decreased metabolism of
`a drug can change
`the
`concentrations of that drug, as well as those of its active,
`inactive, or toxic metabolites. Second, genetic variants
`can produce an unexpected drug effect (eg, haemolysis
`
`1668
`
`THE LANCET - Vol 356 - November 11, 2000
`
`Vanda Exhibit 2035 - Page 2
`
`VNDA 02700305
`
`Vanda Exhibit 2035 - Page 2
`
`

`
`ADVERSE DRUG REACTIONS
`
`Genotype
`
`Frequency
`Whites
`
`Asians
`
`Black Africans
`
`Ethiopians & Saudi
`Arabians
`
`Homozygous or compound heterozygous for
`CYP2C9 *2 or *3
`Heterozygous for C‘YP2C9*2 or *3 (decreasec activity) 14-O—37-0%
`Homo ygois otcompo nd nereroazgo s for
`5-1-13-5%
`numerous |oss—of—function allclos
`
`0 2-1 0%
`
`ND
`
`2-O—3-0°/c
`ill —:I -0%
`
`l\D
`
`l\D
`
`O-5%
`Q-0-8-:I.°A 1 B-7°u)*
`
`l\D
`l'_8:2'O°u
`
`Gene duplication or rrulticluplication (dominant)
`
`1-0-10-0°/u
`
`O-0-2-O
`
`2-0%
`
`10-0-29-0%
`
`Homozygous or compound heterozygous for
`numerous decreased-function alleles
`
`40-0-70-0°/u
`
`10-0-20-0°/u
`
`50-0-60-0%
`
`l\D
`
`Poor metabolisers
`CYPZC9
`
`C_Y_l22D6
`
`Ultrarapid metabolisers
`CYPQDB
`Slow acetylators
`N-acyltransierase 2
`
`ND=rot determined. *In San bushmen.
`
`Table 2: Distribution of polymorphic genes encoding CYP2C9, CYP2D6, and N-acyltransferase 2 in different populations
`
`deficiency).
`dehydrogenase
`glucose—6—phosphate
`in
`Third,
`variation in a drug target can alter the
`clinical
`response and frequency of
`side—effects
`(eg,
`variants of the [5—adrenergic receptor alter response to 3-
`agonists in asthma patients).
`Whether a genetic polymorphism has relevance for
`drug therapy mainly depends on the characteristics of
`the drug in question. The quantitative role of a drug-
`metabolising enzyme, or a drug—uptal<e mechanisms in
`the overall kinetics of a drug and the agent’s therapeutic
`range will determine how much the dose has to be
`adjusted in poor metabolisers or ultrarapid metabolisers.
`"l1heexai=npleoftheC¥P2D6pol—ymorphismaga4r1
`provides incontroversible clinical evidence for
`these
`ideas.
`Nlost
`patients
`(about
`90%)
`require
`75—l50 mg/day of nortriptyline to reach a “therapeutic”
`plasma steady—state concentration of 200-600 nmol/L,
`but poor metabolisers need only 10—20 mg/day to reach
`the same concentrations. Ultrarapid metabolisers, on the
`other hand, may require 300—500 mg/day or even more
`to reach the same plasma concentration (figure).”‘”
`Obviously, if the genotype or phenotype of the patient is
`not known, poor metabolisers will be overdosed and be
`
`metabolisers will be underdosed.
`Another situation is presented if the therapeutic effect
`depends on the formation of an active metabolite (eg,
`morphine from codeine). Poor metabolisers will have no
`drug effect and ultrarapid metabolisers may have
`exaggerated drug responses.” The drug—related criteria
`that make a genetic polymorphism clinically relevant are
`similar to those for drug—concentration monitoring—ie,
`narrow therapeutic
`range
`or
`large
`intcrindividual
`
`In
`suspicion of overdose.
`variation in kinetics or
`pharn1aeogene1ies,ho_wever,asin.gleDIS.Atestdo.n.e
`once in a lifetime can identify the predisposition of
`patients, at least of those with extreme phenotypes.
`
`Examples of genetic polymorphisms affecting
`drug kinetics and drug toxicity
`Drug metabolism
`Many or most of the 50-100 drug—metabolising enzymes
`known are subject to common genetic polymorphisms.
`Table
`1
`lists
`some
`in which clinically important
`differences in drug response have been documented. As
`an example, there is substantial intcrindividual variation
`in plasma concentrations of antidepressants.“"
`
`anti-
`tricyclic
`the
`CYP2D6—The metabolism of
`depressants amitriptyline, clomipramine, desipramine,
`imipramine, and nortriptyline, and of the tetracyclic
`compounds maprotiline and mianserin is influenced by
`the CYPZD6 polymorphism to various degrees. For these
`agents, there are therefore two groups of patients that
`may pose clinical problems.
`The poor metabolisers (and to a lesser degree the
`
`plasma concentrations of tricyclic antidepressants when
`given recommended doses of
`the drugs. The other
`group are the ultrarapid metabolisers who are prone to
`therapeutic failure because the drug concentrations at
`normal doses are far too low (figure). 5—20% of patients
`can belong to one of these risk groups, depending on the
`population studied. Adverse effects clearly occur more
`frequently
`in
`poor metabolisers
`and may
`be
`misinterpreted as symptoms of depression and lead to
`erroneous further increases in the dose.
`
`in psychiatry are affected by the CYPZD6 polymorphism.
`The metabolism of
`the
`recently
`introduced
`antidepressant venlafazine is controlled by CYP2D6,
`and poor metabolisers have a substantially decreased
`oral clearance and increased cardiovascular toxicity.“
`This effect has been documented, however, in only a few
`patients so far. Another group of antidepressants are the
`selective serotonin reuptake inhibitors which interact
`with CYPZD6 in three different ways. Paroxetine,
`fluvoxamine, and fluoxetine are in part metabolised by
`CYPZEO. However,
`the phenotype
`differences
`in
`clearance or plasma concentrations are small in relation
`to the relatively large therapeutic index of these drugs.
`Of substantial importance is the ability of these agents to
`act
`as potent
`competitive
`inhibitors of CYI’2D6
`(paroxetine, fluoxetine). Such inhibition means that the
`elimination of other CYP2D6 substrates—eg, of tricyclic
`antidepressants—is impaired, and that phenotyping with
`
`I
`100
`
`20~3O
`
`80
`70
`60
`50
`40
`30
`20
`10
`O '
`Metabolic ratlo 0-01
`
`
`Numberofpatients
`
`n
`
`/
`
`,
`0-1
`
`1-0
`
`10
`
`Nortriptyline >500
`requirement
`(mg/day)
`
`300
`
`2L50»1OO
`
`Dose requirement for nortriptyline in patients with different
`cYP2D6 phenotypes"‘”
`Poor metabolisers are defined as individuals with a debrisoquine urinary
`metabolic ratio of >12-6 (dashed line).
`
`THE LANCET - Vol 356 - November 1], 2000
`
`1669
`
`Vanda Exhibit 2035 - Page 3
`
`VNDA 02700306
`
`Vanda Exhibit 2035 - Page 3
`
`

`
`ADVERSE DRUG REACTIONS
`
`Gene
`Multidrug resistance
`gene LIVIDR1)”
`B, adrenergic receptor
`gene (,l32ARl”‘
`Sulphonylurea receptor
`gene (SURll
`LQT 1-5 mutations on
`five genes coding for
`cardiac ion crannelsl"
`
`Frequency
`24%
`
`Drug
`Digoxin
`
`37%
`
`Albuterol
`
`2—3%
`
`1—2%
`
`Tolbutamide
`
`Antiarrliytlimics,
`terfenadirie,
`many other drugs
`
`Drug effect
`Increased concentrations of
`digoxin in plasma
`Decreased response to B,
`aclrenergic agonists
`Decreased insulin response
`
`Sudden cardiac death due to
`long-QT syndrome
`
`Table 3: Clinically important genetic polymorphisms of drug
`targets and drug transporters
`
`debrisoquine, sparteine, or dextromethorphan results in
`false
`positive
`results
`or
`“phenocopies”
`of poor
`metabolisers,
`if serotonin—selective reuptake inhibitors
`are coadministered. These interactions are phenotype-
`dependent——ie,
`restricted to extensive metabolisers.
`Citalopram, fiuvoxamine, and sertraline do not share
`this inhibitory property and do not cause CYP2D6—
`specific interactions. Fluvoxamine is also a substrate and
`potent inhibitor of CYP1A2, and thus causes important
`interactions with drugs that are partly metabolised by
`this cytochrome P450 enzyme, such as amitriptyline,
`clomipramine, imipramine, clozapine, and theophylline.
`Thus, polymorphic drug metabolism affects
`a large
`number of drugs used in psychiatric patients, raising the
`question of how this
`information will be used by
`physicians
`in the future. Retrospective
`analysis of
`psychiatric patients treated with substrates of CYTPZD6
`strongly indicates that genotyping can improve efficacy,
`prevent adverse drug reactions, and decrease costs of
`therapy with these agents.“ Obviously, prospective trials
`are needed to prove the value of phenotyping or
`genotyping patients with depression in selecting the
`proper starting dose to increase therapeutic efficacy and
`prevent toxicity. Striking differences in the adverse drug
`reactions of opioids are associated with the CYPZD6
`polymorphism. Dextromethorphan,
`codeine, hydro-
`codone, oxycodone, ethylmorphine, and dihydrocodeine
`are
`dealkylated
`by
`polymorphic CYP2D6. The
`polymorphic O—demetliylation of codeine is of clinical
`importance when this drug is given as an analgesic.
`About 10% of codeine is O—demethylated to morphine,
`and this pathway is deficient in poor metabolisers. Poor
`metabolisers therefore experience no analgestic effects of
`codeine.“" Similarly,
`respiratory, psychomotor,
`and
`pupillary effects of codeine
`are decrease in poor
`metabolisers compared with extensive metabolisers.
`Codeine is frequently recommended as a drug of first
`choice for treatment of chronic severe pain. Physicians
`must appreciate that no analgesic effect is to be expected
`in the 5—l0% of whites who are of the poor metaboliser
`phenotype, or who are extensive metabolisers and
`receive concomitant treatment with a potent inhibitor of
`CYPZD6 such as quinidine. The inability to form active
`metabolites
`from opioids
`could
`protect
`poor
`metabolisers against oral opiate dependence.
`
`of a clinically important
`CYP2C9—Another example
`genetic polymorphism of drug metabolism is
`the
`association of variant alleles of CYPZC9 with lower
`r‘
`Wallallll dUSE IEQCIIIEIIIEXIES. ll! 3 IEU USPECUVE Slllldy 01 :1
`population from an anticoagulant clinic,
`the CYPZC9
`alleles associated with decreased enzyme activity (*2 and
`*3) were found to be over~represented in patients
`stabilised on low doses of warfarin. These patients had
`an increase incidence of major and minor haemorrhage.“
`Up to 37% of a British population were carriers of one
`mutant allele (*2 or *3) and were therefore at higher risk
`
`of haemorrhage (table 2). Homozygous carriers of
`mutant alleles of CYPZC9 are rare (O-2 l-0% of the
`population) and this genotype is associated with an even
`higher risk of warfarin adverse drug reactions and with
`severe impairment of the metabolism of tolbutamide,
`glipizide, and phenytoin.°'“’
`
`CamwT
`bone—marrow toxicity (acute leucopenia, anaemia, and
`pancytopenia)
`in patients with thiopurine methyl-
`transferase deficiency treated with standard doses of
`mercaptopurine, thioguanine, and azathioprine is a rare
`(about one in 300) event
`in the treatment of acute
`lymophoblastic leukaemia in children. Patients with this
`purine methyltransferase deficiency can require up to a
`15-fold reduction in mercaptopurine to prevent fatal
`haematotoxicity.5'*2“ Other pharmacogenetic examples
`of adverse drug reactions in cancer chemotherapy are the
`myelosuppression and neurotoxicity of fluorouracil
`in
`patients with
`a
`deficiency
`of dihydropyraiiiidine
`dehydrogenase;
`the myelosuppression and diarrhoea
`after the topoisomerasc I inhibitor irinotecan in patients
`with an inherited deficiency in glucuronidation by a
`promotor
`polymorphism of
`UGT—glucuronosyl—
`transferase UGTlAl; and the greater bone—marrow
`toxicity of the topoisomerase II inhibitor amonafide in
`N—acetyltransferase
`2
`rapid acetylators
`(30—60% of
`whites, 80-90% of Asians)?”
`
`Drug transport
`Several membrane transporters are involved in the
`absorption of drugs into the intestinal
`tract,
`in the
`uptake into the brain and other tissues, or
`in the
`transport into specific sites of action——eg, the synaptic
`cleft. However, little is known about transporter variants
`in relation to drug response. One such variant was
`recently discovered for the multidrug resistance gene
`/VIDRI, which
`codes
`for
`an
`ATP—dependent
`transmembrane efflux pump (l’—glycoprotein), whose
`function is the export of numerous substances including
`drugs from the inside of cells to the outside, protecting
`cells
`from accumulation
`of
`toxic
`substances
`or
`metabolites. A mutation in exon 26 of the ZVIDRI gene
`(C343‘3'l‘) correlated with the expression levels and the
`function
`of
`intestinal
`P—glycoprotein. Thus,
`the
`concentrations of digoxin in plasma were up to four—fold
`higher in individuals homozygous for this mutation after
`a
`single
`oral
`dose
`of digoxin. The maximum
`concentration in plasma (Cm) of digoxin was also
`increased after chronic administration.“ Homozygosity
`forthi variantwa obseLv_edin24"nofaGx:i:n1an
`population.
`Substrates
`of
`P—glycoprotein
`include
`numerous important drugs with narrow therapeutic
`ranges including chemotherapeutic agents, ciclosporin
`A, verapamil, terfenadine, fexofenadin, and most HIV-1
`protease inhibitors. Therefore, this polymorphism could
`have a major impact on the requirement for individual
`dose
`adjustments
`for
`carriers
`of
`this mutation.
`Mutations of other
`transporters, particularly those
`involved in reuptake of serotonin, dopamine, and
`y—aminobutyric acid (GABA) are presently being studied
`with regard to "‘l—l-Iii-Ga-ll-l—}
`relevant changes
`in dru"
`response. Transporter pharmacogenetics is
`a rapidly
`developing field.
`
`Drug targets
`The effects of most drug are exerted via interaction of
`the compound with membrane receptors (about 50% of
`drugs), enzymes (about 30%), or ion channels (about
`
`1670
`
`THE LANCET - Vol 356 - November 11, 2000
`
`Vanda Exhibit 2035 - Page 4
`
`VNDA 02700307
`
`Vanda Exhibit 2035 - Page 4
`
`

`
`ADVERSE DRUG REACTIONS
`
`7
`
`9
`
`5%).“ Many of the genes encoding these proteins exhibit
`polymorphisms which may
`alter
`drug
`response.
`Clinically
`relevant
`examples
`are
`summarised
`in
`table 3.3““7 One of the best studied drug receptors is the
`B, adrenergic receptor, and some of its mutations (eg,
`the common mutation Arg—~>Gly at aminoacid 16) are
`major determinants of the |3,—agonist bronchodilator
`response.“ Similarly mutations
`in the
`angiotensin
`converting enzyme (ACE) gene have been proposed to
`account
`for differences
`in the
`response
`to ACE
`inhibitors, but the data from different studies remain
`controversial.“ A combination of two mutations of the
`gene for a high~affinity sulphonylurea receptor lead to a
`40% reduction in the insulin response to tolbutamide”°
`and genetic polymorphisms of the 5—hydroxytryptamine
`(serotonin) receptor HTRZA could be associated with
`the
`response
`to
`clozapine
`in
`patients with
`S:]mZC]D]D]:Em]'S 19
`lflutations in five genes, each encoding structural
`subunits of cardiac ion channels, affect the risk of drug-
`induced long—QT syndrome—a potential
`cause of
`sudden cardiac death in young individuals without
`structural heart disease. The prevalence of long—QT
`syndrome is about one in 10 000. All five genes code for
`membrane ion channels affecting sodium or potassium
`transport and are influenced by antiarrhythmics and
`other drugs."
`
`drug treatment. Tram]: Phaw11aro/ Sci 1999; 20: 3/lr2—/l9.
`5 Evans WE, Relling MV. Pharmacogenomics: translating
`functional genomics into rational therapeutics. .S'cim/my 1999; 286:
`48 7—9 1 .
`6 Roses AD. Pharmacogenetics and future drug development and
`delivery. Lamcet 2000; 355: 1358—61.
`i\/leyer UA, Zanger UM. Molecular mechanisms of genetic
`polymorphisms ofdruv metabolism. Armu Rev Pharmacol Toxicol
`1997; 37: 269—96.
`8 Aithal GP, Day CP, Kesteven P], Daly AK. Association of
`polymorphisms in the cytochrome P450 CVPZC9 with warfarin dose
`requirement and risk ofbleeding complications. Lance! 1999; 353:
`71 7—l 9.
`l\/liners ]P, Birkett D]. Cytochrome P4502C9; an enzyme of major
`importance in human drug metabolism. Br} C/in Pharmacal 1998;
`45: 525 38.
`10 Kidd RS, Straughn AB, Meyer NSC, Blaisdell ], Goldstein ]A,
`Dalton ]T. Pharmacokinctics of chlorpheniramine, phenytoin,
`glipizide and nifedipine in an individual homozygous for the
`CYI’2C9*3 allele. Pharmawugenetics 1999; 9: 71—80.
`11 Furuta T, Ohashi K, Kobayashi K, et al. Effects of clarithromycin
`on the metabolism of omeprazole in relation to CYP2C19 genotype
`status in humans. Clin Pharmacul Ther 1999; 66: 265—74.
`12 Iyer L, Ratain N1]. Pharmacogenetics and cancer chemotherapy.
`Em‘ j’ Cancz2r'1998, 34: 1493—99.
`13 Gonzalez F], Skoda RC, I\'imura S, et al. Characterization of the
`common genetic defect in humans deficient in debrisoquine
`metabolism. I\z'amre 1988; 331: 442—46.
`14 Raimundo S, Fischer], Eichelbaum ZVI, Griese E—U, Schwab N1,
`Zanger UM. Elucidation of the genetic basis of the common
`“intermediate metabolizer” phenotype for drug oxidation by
`CYPZD6. P/iawriacogeneticc 2000; 10: l—"3.
`15 Brookes A]. The essence of SNPs. Gene 1999; 234: 177—86.
`16 Kalow W/', Tang B—K, Endrenyi L. Hypothesis: contparison of inter-
`Mo;-nook
`and intra-individual variations can substitute for twin studies in drug
`research. Pharmacagerietics 1998; 8: 283—89.
`Tlie systematic identification and functional analysis of
`17 Bertilsson L, Dahl M—L, Tybring G. Pharmacogenetics of
`human genes is
`revolutionising the study of disease
`antidepressants: clinical aspects. /lcza Psycliiatr Scand 1997; 96:
`processes and the development and rational use of
`14—2 1 .
`drugs. It enables physicians to make reliable assessments
`18 Dalen P, Dahl M—L, Bernal Ruiz ML, Nordin ], Bertilsson L.
`of an individuals’ risk of acquiring a particular disease,
`10—liydroxylation of nortiptyline in white persons with 0, 1, 2, 3,
`and 13 functional CYP2D6 genes. Clin Phmmacol Ther 1998; 63:
`raises the number and specificity of drug targets, and
`444—52.
`explains
`interindividual variation of the therapeutic
`19 Sindrup SH, Rrosen K. The pharmacogenetics of codeine
`effectiveness and toxicity of drugs. Mutant alleles at a
`hypoalgesia. P/Iarniacogerzetics 1995; 5: 335—46.
`single gene locus are the best studied individual risk
`20 Lessard E, Yessine M-A, Hamelin BA, O’Hara G, LeBl-anc ],
`factors for adverse drug reactions, including the genes
`Turgeon
`Influence of CYP2D6 activity on the disposition and
`cardiovascular toxicity of the antidepressant agent venlafaxine in
` ,
`humans. P/1arm.zcogc11czi¢:s 1999; 9: 435 43.
`dihydropyrimidine dehydrogenase, and the cytochrome
`21 Chou WVH, Yan F-X, De Leon ], et al. Extension of a pilot study:
`P450 enzymes. Genotyping can predict the extremes of
`impact of the cytochrome P450 ZD6 polymorphism on outcome and
`phenotypes in these situations. However, less definable
`costs associated with severe mental illness. ] czm P53/clioplwrnzauol
`pharmacogenetic factors produce a phenotype together
`2000; 20: 246—5l.
`22 Weinshilboum RlVl, Otterness DM, Szumlanski CL. Metliylation
`with other variant genes and with environmental factors
`pharmacogenetics: catecliol O-methyltransferase, thiopurine
`(eg, smoking, diet, &c). Genomics is providing the
`methyltransferase, and histamine N—methytransferase. Annu Rev
`information and technology to analyse these complex
`Plmrmacal Toxicol 1999; 39: 19—52.
`multifaetorial
`situations
`and to
`obtain individual
`23 Hoffmeyer S, Burk 0, Von Richter O, et al. Functional
`genotypic information. Awareness of inherited variations
`pOlyD10Ipl]lSlTlS ofthe huntan multidrug-resistance gene: multiple
`sequence variations and correlation of one allele with P—glyL;oprolein
`of drug responsiveness, which are constant throughout
`life,—can lead tl“)—&(T§E 8fifu. ,‘§4%%,97.
`3473—78.
`patient’s genetic makeup and are likely to prevent
`24 Drews ], Ryser S. The role ofinnovation in drug development. Nat
`adverse drug reactions.
`Biozec1moZl997;15: 1318—l9.
`25 Ligget SE. The pharmacogenetics ofbeta2—adrenergic receptors:
`relevance to asthma. _7'AZlergj/ Clirz Immmzol 2000; 105: 487—92.
`26 Hansen T, Echwald SM, Hansen L, et al. Decreased tolbutamidc—
`stimulated insulin secretion in healthy subjects with sequence
`variants in the high-affinity sulfonyurea receptor gene. Diabctes 1998;
`47: 598 605.
`27 Priori SG, Barhanin ], Hauer RN, et al. Genetic and molecular basis
`of cardiac arrhythmias: impact on clinical management parts I and
`I1‘. Czrculazzrm I999; 99: 5 ] 8—2H.
`28—Na is , dc , dt ]-on-g-PE. i-
`converting enzyme gene l/1) polymorphism and renal disease. ff [Mal
`.Med 1999; 77: 781—91.
`29 Arranz M], Munro], Sham P, et al. Meta-analysis of studies on
`genetic variation in 5—HT2A receptors and clozapine response.
`Schz’z0pIzrRe5 1998; 32: 93—99.
`
`'lhis study was supported by the Swiss National Science Foundation.
`
`References
`l Lazarou ], Pomeranz BH, Corey PN. Incidence of adverse drug
`reactions in hospitalized patients: a meta-analysis of prospective
`studies.,7A}\/IA 1998; 279: 1200—O5.
`2 Meyer UA. Genotype or phenotype: the definition of a
`pharmacogenetic polymorphism. Pharmacogenetics 1991; 1: 66—67.
`3 Meyer U731. Drugs in special patient groups: clinical importance ot
`genomics in drug effects. In: Carruthers GS, Hoffmann BR,
`Melmon KL, Nierenberg DW/, eds. New York: McGrawI Iill, 2000:
`1 179—205.
`Ingelman—Sundberg N1, Oscarson M, McLellan RA. Polymorphic
`human cytochrome P450 enzymes; an opportunity for individualized
`
`4
`
`THE LANCET - Vol 356 - November 11, 2000
`
`1671
`
`Vanda Exhibit 2035 - Page 5
`
`VNDA 02700308
`
`Vanda Exhibit 2035 - Page 5