PHARMACY LIBRARY
`UNIVERSITY OF WISCONSIN
`MAR 0 8 2004
`
`Madison , WI 53705
`
`Roxane Labs., Inc.
`Exhibit 1016
`Page 001
`
`

`
`Vol. 27, No.3, 2004
`
`Contents
`
`Leading Article
`
`Review Articles
`
`Pharmacogenetic Aspects of Drug-Induced Torsade de
`Pointes: Potential Tool for Improving Clinical Drug
`Development and Prescribing
`RR Shah
`
`Safety and Tolerability of Lamotrigine for Bipolar
`Disorder
`CL Bowden, GM Asnis, LD Ginsberg, B Bentley, R Leadbetter,
`R White
`Benefit-Risk Assessment of Rofecoxib in the Treatment of
`Osteoarthritis
`H Schmidt, BG Woodcock, G Geiss/inger
`
`Original Research
`Articles
`
`Malformation Rates· in Children of Women with
`Untreated Epilepsy: A Meta-Analysis
`S Fried, E Kozer, I Nulman, TR Einarson, G Koren
`The Risk of Severe Depression, Psychosis or Panic Attacks
`with Prophylactic Antimalarials
`CR Meier, K Wilcock, SS fick
`
`Erratum
`
`Drug
`SafetY.
`
`145-172
`
`173-184
`
`185-196
`
`197-202
`
`203-213
`
`213
`
`PHARMACY LIBRARY
`UNIVERSITY OF WISCONSIN
`MAR 0 8 2004
`
`Madison, WI 53705
`
`Drug Safety is indexed in Index Medicus, Medline, EMBASE/Excerpta Medica, Current Contents/Clinical Medicine, BIOSIS® Database,
`International Pharmaceutical Abstracts (IPA), Elsevier BIOBASE/Current Awareness in Biological Sciences, Sociedad Iberoamericana de
`lnformaci6n Cientifica (SIICJ databases and Chemical Abstracts. Individual articles are available through the ADONIS document
`delivery service and on-line via the World Wide Web through Ingenta. Further details are available from the publisher.
`
`Roxane Labs., Inc.
`Exhibit 1016
`Page 002
`
`

`
`!!a~ .
`a 1s
`
`INTERNATIONAL
`
`F.Y. Aold, Wmnipeg, MN, Canada
`a. lllal'lnwarth, Bordeaux, Frl!J¥:1!
`~ J. ~~ London, Englani:l •
`. N.L llenowllz, San Francisco, CA, USA
`J·,~· ,Bergmann, Paris, France
`{~;.C, ~~ paptuml, l!r)glanQ
`'i ~· ~kley, Wod~ ACT, Aulltr&lia t
`P.A. Chyka" Memphls, TN, USA
`D.~. 'crou-., Dune~, NllW Zealand
`· -.o. Day, syl~y, N5Wi ft.~~.tt.ua .
`• N. ·~,S0uthamp~~~
`t.R. Edwards, UpPsala, Sweden
`~C.G. Egbert~, l;ltrecht,
`The Netherlands
`
`L ~. Kogarah, NSW, Australia
`J.·R. Lciporte, Barcelona, Spain
`D.H. Lawlon, Glasgow, Scotland
`I.J. Llpworlh, Dutldee, Scotland
`T. utovllz, Wa~ DC, USA
`R. MeYboom, 's-Hertogenbosch.
`The Netherlandii
`
`G.N. v~ I:.ondam Jm21and
`':f!T.i ·
`~~t~ . k•IQ
`~ ·~• II
`LV. Wilton, Southampton, England
`M. Winder, Hannover, Germany
`tC.K. WQng, London, England
`!•
`
`Drug
`SafetY-
`
`Aim and Scope: Drug Safety advances the rational use of pharmacotherapy by
`providing a programme of review articles offering guidance for safe and
`effective drug utilisation and prescribing.
`The Journal includes:
`• Leading / current opinion articles providing an overview of contentious or
`emerging issues
`• Definitive reviews on the epidemiology, clinical features, prevention and
`management of adverse effects of an individual drug or drug class when
`given at therapeutic dosages or following overdose
`• Benefit-risk assessments providing an in-depth review of adverse effects and
`efficacy data for a drug in a specific disease to place the benefit-risk relationship
`in clear perspective
`• Practical reviews covering drug use in particular 'at-risk' patient groups to
`achieve optimal outcomes
`• Concept reviews covering issues in pharmacovigilance, risk management
`and medication error prevention
`• Original research articles will also be considered for publication.
`All manuscripts are subject to peer review by international experts. Letters to
`the editor are welcomed and will be considered for publication.
`
`Editor: Rosie Stather
`
`Publication Manager: June B. Hearne
`
`Editorial Office and Inquiries: Adis International Ltd, 41 Centorian Drive,
`Private Bag 65901, Mairangi Bay, Auckland 1311, New Zealand. Information on
`the preparation of manuscripts will be provided to authors.
`
`E-mail: drugsafety@adis.co.nz
`
`http://www .adis.com/ drugsafety
`
`Drug Safety (ISSN 0114-5916) is published as 1 volume with 15 issues by Adis
`International Limited. Annual 2004 institutional subscription price: $US1985.
`Personal subscription rate: $US225. For 2004, subscription includes electronic
`access at no extra cost. (Further subscription information is given at the back of
`each issue.)
`Polley Statement: Although great care has been taken in compiling the
`content of this publication, the publisher and its servants are not responsible or
`in any way liable for the currency of the information, for any errors, omissions
`or inaccuracies, or for any consequences arising therefrom. Inclusion or
`exclusion of any product does not imply its use is either advocated or rejected.
`Use of trade names is for product identification only and does not imply
`endorsement. Opinions expressed do not necessarily reflect the views of the
`Publisher, Editor or Editorial Board.
`Copyright: © 2004 Adis Data Information BV. All rights reserved throughout
`the world and in all languages. No part of this publication may be reproduced,
`transmitted or stored in any form or by any means either mechanical or
`electronic, including photocopying, recording, or through an information
`storage and retrieval system, without the written permission of the copyright
`holder.
`The appearance of the code at the top of the first page of an article in this
`journal indicates the copyright owner's consent that copies of the article may be
`made for the personal or internal use of specific clients. This consent is given
`provided that the fee of $US31 per copy is paid directly to the Copyright
`Clearance Center Inc., 222 Rosewood Drive, Danvers, Massachusetts 01923,
`USA, for copying beyond that permitted by sections 107 or 108 of the US
`Copyright Law. This consent does not extend to other kinds of copying such as
`copying for general distribution, for advertising or promotional purposes, for
`creating new collective works, or for resale.
`
`Roxane Labs., Inc.
`Exhibit 1016
`Page 003
`
`

`
`LEADING ARTICLE
`
`Drug Safety 2004,27 (3), 145-172
`0114-5916/04/0003-0145/531.00/0
`
`© 2004 Adls Data Information BV. All rights reserved
`
`Pharmacogenetic Aspects of
`Drug-Induced Torsade de Pointes
`Potential Tool for Improving Clinical Drug Development
`and Prescribing
`
`Rashmi R. Shah
`Medicines and Healthcare products Regulatory Agency, London, United Kingdom
`
`Abstract
`
`Drug-induced torsade de pointes (TdP) has proved to be a significant iatro(cid:173)
`genic cause of morbidity and mortality and a major reason for the withdrawal of a
`number of drugs from the market in recent times. Enzymes that metabolise many
`of these drugs and the potassium channels that are responsible for cardiac
`repolarisation display genetic polymorphisms. Anecdotal reports have suggested
`that in many cases of drug-induced TdP, there may be a concealed genetic defect
`of either these enzymes or the potassium channels, giving rise to either high
`plasma drug concentrations or diminished cardiac repolarisation reserve, respec(cid:173)
`tively. The presence of either of these genetic defects may predispose a patient to
`TdP, a potentially fatal adverse reaction, even at therapeutic dosages of QT(cid:173)
`prolonging drugs and in the absence of other risk factors. Advances in pharmaco(cid:173)
`genetics of drug metabolising enzymes and pharmacological targets, together with
`the prospects of rapid and inexpensive genotyping procedures, promise to
`individualise and improve the benefit/risk ratio of therapy with drugs that have the
`potential to cause TdP. The qualitative and the quantitative contributions of these
`genetic defects in clinical cases of TdP are unclear because not all of the patients
`with TdP are routinely genotyped and some relevant genetic mutations still
`remain to be discovered.
`There are regulatory guidelines that recommend strategies aimed at uncovering
`the risk of TdP associated with new chemical entities during their development.
`There are also a number of guidelines that recommend integrating pharmacogene(cid:173)
`tics in this process. This paper proposes a strategy for integrating pharmacogene(cid:173)
`tics into drug development programmes to optimise association studies correlating
`genetic traits and endpoints of clinical interest, namely failure of efficacy or
`development of repolarisation abnormalities. Until pharmacogenetics is carefully
`integrated into all phases of development of QT-prolonging drugs and large-scale
`studies are undertaken during their post-marketing use to determine the genetic
`components involved in induction ofTdP, routine genotyping of patients remains
`unrealistic.
`Even without this pharmacogenetic data, the clinical risk ofTdP can already be
`greatly minimised. Clinically, a substantial proportion of cases of TdP are due to
`the use of either high or usual dosages of drugs with potential to cause TdP in the
`presence of factors that inhibit drug metabolism. Therefore, choosing the lowest
`effective dose and identifying patients with these non-genetic risk factors are
`important means of minimising the risk of TdP. In view of the common secondary
`pharmacology shared by these drugs, a standard set of contraindications and
`
`Roxane Labs., Inc.
`Exhibit 1016
`Page 004
`
`

`
`146
`
`Shah
`
`warnings have evolved over the last decade. These include factors responsible for
`pharmacokinetic or pharmacodynamic drug interactions. Among the latter, the
`more important ones are bradycardia, electrolyte imbalance, cardiac disease and
`co-administration of two or more QT-prolonging drugs.
`In principle, if large scale prospective studies can demonstrate a substantial
`genetic component, pharmacogenetically driven prescribing ought to reduce the
`risk further. However, any potential benefits of pharmacogenetics will be squan(cid:173)
`dered without any reduction in the clinical risk of TdP if physicians do not follow
`prescribing and monitoring recommendations.
`
`Prescribing drugs during routine clinical practice
`is a relatively empirical trial and error process con(cid:173)
`sisting of selecting a drug and recommending a
`relatively rigid 'standard' dose schedule for every
`patient. These dose schedules, investigated during
`drug development, are based on population mean
`data and usually ignore the large interindividual
`variability that is present within a population.
`The International Conference on Harmonisation
`(ICH) guidelinef'l on 'Dose-Response Information
`to Support Drug Registration' recommends that in
`using dose-response information, the influences of
`various factors should be identified where possible.
`Pharmacokinetics and pharmacodynamics, the two
`components of a dose-response curve, are both sub(cid:173)
`ject to large interindividual variability. This vari(cid:173)
`ability anses from their modulation by factors such
`as age, gender, co-medications or the presence of
`concurrent diseases, e.g. renal or hepatic dysfunc(cid:173)
`tion. This variability also arises from genetic influ(cid:173)
`ences that regulate the expression of drug metabolis(cid:173)
`ing enzymes (pharmacokinetic variability) or the
`function of various pharmacological targets (phar(cid:173)
`macodynamic variability). The presence of variant
`alleles often exerts influences that usually far exceed
`those due to the other covariates stated above.
`It is estimated that the human genome has about
`50 000-1 00 000 functional single nucleotide poly(cid:173)
`morphisms (SNPs) [variations in the DNA in which
`a single base pair varies]. These SNPs give rise to
`variant alleles responsible for genetic polymorph(cid:173)
`isms within a population and may account for genet(cid:173)
`ically mediated interindividual differences in res(cid:173)
`ponse to clinically prescribed drugsPl The need to
`study genetically determined biochemical variations
`that characterise humans was first considered almost
`a century ago.f3.4J
`
`In terms of genetic influences on drug response,
`two models exist - high genetic and low environ(cid:173)
`ment model versus low genetic and high environ(cid:173)
`ment model. For many drugs with a shallow concen(cid:173)
`tration-response curve, genetic factors seem to mat(cid:173)
`ter only a little, while for others, genetic differences
`between individuals account for a very significant
`fraction of the overall variation in drug response. A
`typical example of an abnormal response that is
`almost exclusively genetically determined is the
`prolonged apnoea that follows· administration of
`suxamethonium chloride (a muscle relaxant) to indi(cid:173)
`viduals who inherit a variant form of plasma
`butyrylcholinesterase (designated atypical cholines(cid:173)
`terase). Subsequently polymorphism in the metabol(cid:173)
`ism of isoniazid by N-acetyltransferase 2 (NAT2)
`explained the susceptibility of some individuals to
`drugs metabolised by acetylation, e.g. peripheral
`neuropathy, hepatitis or poor anti-tuberculous res(cid:173)
`ponse following administration of isoniazid or
`haematological reactions or poor therapeutic res(cid:173)
`ponse to dapsone. Beginning in the late 1970s, ma(cid:173)
`jor advances in pharmacogenetics followed the dis(cid:173)
`covery of genetic polymorphism in enzymes that
`catalyse phase I metabolism (cytochrome P450
`[CYP]). This discovery not only explained further
`the individual susceptibility to drug reactions and
`lack of efficacy, but also provided a mechanistic and
`rationale basis for metabolic drug interactions.
`At a pharmacokinetic level, use of pharmaco(cid:173)
`genetics has already resulted in great improvement
`of cancer therapy with mercaptopurine and azathio(cid:173)
`prine. These drugs are metabolised by thiopurine S(cid:173)
`methyltransferase
`(TPMT)
`that
`is expressed
`polymorphically in a population. At a pharmacody(cid:173)
`namic level, great advances have been made in
`uncovering the genetic and molecular bases of con-
`
`© 2004 Adls Data Information BV. All rights reserved.
`
`Drug Safety 2004: 27 (3)
`
`Roxane Labs., Inc.
`Exhibit 1016
`Page 005
`
`

`
`Pharmacogenetic Aspects of Drug-Induced Torsade de Pointes
`
`147
`
`genital long QT syndrome (LQTS). Identification of
`mutations at different genetic loci of ion channels
`has shown LQTS to be highly heterogeneous and
`have begun to explain the significant differences in
`the clinical features of individuals with it. Not only
`are there are gene-specific differences in the triggers
`for cardiac events but for some forms of LQTS,
`there are also gene-specific differences in response
`to changes in lifestyle and to therapy.
`Anecdotal observations of sporadic cases have
`raised the expectation that application of pharmaco(cid:173)
`genetics will result in the choice of the right drug at
`the right dose at the outset of therapy for each
`patient, thus maximising the probability of im(cid:173)
`proved efficacy and minimising the probability of an
`adverse drug reaction. There is, however, an urgent
`need to explore whether in clinical practice pharm(cid:173)
`acogenetics will deliver these anticipated benefits
`and to put these potential benefits in perspective
`with regard to non-genetic factors that also influence
`drug response.
`Over the last 10 years, the potentially fatal ad(cid:173)
`verse effect of many non-antiarrhythmic drugs on
`the QT interval of the surface ECG has attracted
`considerable clinical and regulatory interest.l5•71
`This paper focuses on drug-induced QT interval
`prolongation and torsade de pointes (TdP), to put in
`perspective the relative contributions of genetic and
`non-genetic factors in clinical practice.
`
`1. Drug-Induced Torsade de
`Pointes (TdP)
`
`l
`I I
`
`t I
`t
`l
`
`i
`
`val prolongation is observed with a large number of
`non-class III antiarrhythmic drugs. When prolonged
`excessively, it often leads (under the right circum(cid:173)
`stances) to potentially fatal ventricular tachyar(cid:173)
`rhythmias, particularly a polymorphic form known
`as TdP. Thus, the balance between the therapeutic
`antiarrhythmic and the potentially fatal proar(cid:173)
`rhythmic prolongations of the QT interval is a deli(cid:173)
`cate one.
`Since the measured QT interval varies with heart
`rate, it requires correction to obtain a rate-corrected
`QT interval - the QTc interval. Issues surrounding
`the measurement of QT interval, the rate correction
`formula that is most appropriate and the magnitude
`of QTc interval prolongation that is likely to induce
`TdP are complex and the reader is referred to other
`detailed reviews on this subject.l5·7l Quantitatively,
`the proarrhythmic threshold is not a sharp one and
`TdP can occur after only a minimal prolongation of
`the QTc interval. The risk of TdP, however, usually
`begins when the QTc interval is about 500ms and
`rises exponentially thereafter.
`Clinical manifestations ofTdP, which is usually a
`transient tachyarrhythmia, may include palpitation.
`When TdP is sustained, symptoms arising from im(cid:173)
`paired cerebral circulation such as dizziness, syn(cid:173)
`cope and/or seizures may manifest. TdP subsequent(cid:173)
`ly degenerates into ventricular fibrillation in about
`20% of casesl81 and, not uncommonly, sudden death
`may occur.l91 The overall mortality from TdP is of
`the order of 1 0-17%.fK,IOl
`Not surprisingly, this potentially fatal proar(cid:173)
`rhythmic effect of many non-antiarrhythmic drugs
`on the QT interval of the ECG has attracted consid(cid:173)
`erable clinical and regulatory interest over the last
`decade. Regulatory rejection of new drugs or restric(cid:173)
`tions in the use of many old and other new drugs
`over the last decade because of their 'QTc liability'
`has had a very profound influence on drug develop(cid:173)
`ment. There are justifiable clinical and regulatory
`expectations of a better pre-approval characterisa(cid:173)
`tion of new chemical entities (NCEs) for this poten(cid:173)
`tial risk. This expectation also applies to old drugs
`already on the market should any of them unexpect(cid:173)
`edly prove to be proarrhythmic. Among the many
`examples of such drugs are pimozide, terfenadine,
`thioridazine and cisapride. One important aspect in
`
`l.l Clinical and Regulatory Concern
`
`Drug-induced prolongation of the QT interval is
`a typical type A pharmacological adverse reaction;
`usually
`resulting
`from concentration-dependent
`block of potassium channels by most QT-prolong(cid:173)
`ing drugs. Despite its clinical and regulatory signifi(cid:173)
`cance, this potentially proarrhythmic effect of drugs
`on QT interval prolongation is not well understood.
`QT interval prolongation per se is not proarrhythmic
`and does not influence cardiac performance. Class
`lll antiarrhythmic drugs are designed to prolong the
`rate corrected QT interval (QTc) and increase myo(cid:173)
`cardial refractory period, thereby exerting their ther(cid:173)
`apeutic benefit. Unfortunately, however, QTc inter-
`
`© 2004 Adis Data lnfarmatlan BV. All rights reserved.
`
`Drug Safety 2004: 27 (3)
`
`Roxane Labs., Inc.
`Exhibit 1016
`Page 006
`
`

`
`148
`
`Shah
`
`characterising this risk is the investigation of the
`role of genetic factors.
`In December 1997, the Committee for Proprie(cid:173)
`tary Medicinal Products (CPMP) of the European
`Union (EU) adopted a significant document con(cid:173)
`cerning pre-approval evaluation of an NCE for its
`potential to prolong the QT interval; 'Points to Con(cid:173)
`sider: The Assessment of the Potential for QT Inter(cid:173)
`val Prolongation by Non-Cardiovascular Medicinal
`Products' _[lll This document describes a strategy
`that EU regulators recommend for this purpose.
`There are initiatives currently in progress at the ICH
`level aimed at harmonising this strategy internation(cid:173)
`ally.
`
`l .2 Basic Electrophysiology
`
`The QT interval of the ECG reflects the duration
`of ventricular action potential that is determined by
`a delicate balance between inward and outward cur(cid:173)
`rents, especially during phases 2 and 3 of the action
`potential. Major ion currents involved during the
`depolarisation and repolarisation phases of a ven-
`
`tricular action potential are shown in figure 1. Re(cid:173)
`duction in the major outward current, mediated by
`the rapid component of the delayed rectifier potassi(cid:173)
`um channels (lKr), results in prolongation of the QT
`interval. Although reduction in lKr may result for
`many reasons, the most frequent cause at present is
`the administration of many clinically useful drugsPl
`Drugs reduce this current mainly by their effect on
`human KCNH2 gene (human ether-a-go-go-related
`gene [HERO]) a-subunits of the lKr channel.
`Two main hypotheses have been proposed to
`explain the underlying electrophysiological mecha(cid:173)
`nism for the induction of TdP. The two hypotheses
`are not mutually exclusive and may even be comple(cid:173)
`mentary. Delayed repolarisation gives rise to the
`development of early after-depolarisations (EADs)
`at the Purkinje fibre level. The emergence of these
`EADs is favoured by calcium loading during the late
`phase 2 of the prolonged action potential due to the
`first short cycle of the short-long-short series that
`generally precedes TdP. The amplitude of EADs is
`cycle-dependent and there is a strong correlation
`between the preceding RR interval and the ampli-
`
`Current
`
`Na• current
`
`L-type Ca2• current
`
`T-type Ca2• current
`
`Na• -Ca2• exchange
`
`I 10 1 (4-AP-sensitive)
`I TO 2 (Ca2•-activated)
`
`IKs
`
`IKr
`
`IKur
`
`lc1 or/~
`
`Inward rectifiers
`
`I, (pacemaker current)
`
`Probable clone
`
`SCN5A (hH1)*
`
`Dihydropyridine receptor*
`
`Na• -Ca2• exchanger
`
`Kv1.2, 1.4, 1.5, 2.1 +lor 4.2/3*
`
`- - - - - - - - - - - - KvLOT1 + lsK (minK)
`HERG
`
`-----====-------- CFTR (CI), ?TWIK/ORK1 (K) family
`-------=----- Kir2 (IK1), Kir3.1+3.4 (IK·Ach); Kir6 +SUR (IK-ATP)
`
`Possibly Kv1.5*
`
`Fig. 1. Cardiac ionic currents and respective ion channel clones responsible for generation of the action potential. Inward currents are drawn
`in blue outward currents in black. The amplitudes are not to scale (from Priori et al.,l12l with permission from Elsevier). • Subunits also
`identified.
`
`© 2004 Adis Data lnforma~on BV. All rights reserved.
`
`Drug Safety 2004; 27 (3)
`
`Roxane Labs., Inc.
`Exhibit 1016
`Page 007
`
`

`
`Pharmacogenetic Aspects of Drug-Induced Torsade de Pointes
`
`149
`
`tude of EADs that follow. When the amplitude of
`the EADs reaches a critical threshold, a repetitive
`burst of electrical activity is triggered, which forms
`the basis of TdP. The other hypothesis proposes an
`increase in transmural dispersion of action potential
`duration
`throughout the ventricle and
`this may
`facilitate transient functional block. Drugs that
`block IKr cause significantly greater prolongation of
`the action potential duration in Purkinje fibres and
`the M-cells (special ventricular myocytes found in
`the mid-myocardial region) than in other myocytes
`layers. This is the consequence of relative scarcity of
`other major cardiac
`repolarising channels
`in
`Purkinje fibres and the M-cells. Thus, a uniform
`decrease in IKr function results in the transmural
`dispersion of action potential duration within the
`ventricular wall.
`
`1.3 QT Interval as a Surrogate of TdP
`
`Numerous clinical and experimental data have
`established that QTc interval prolongation is a major
`precursor of drug-induced TdP. It is, however, an
`imperfect (but at present the best available) surro(cid:173)
`gate marker of the risk of TdP. The potential for
`emergence of EADs and induction of TdP following
`prolongation of the QTc interval varies greatly. Not
`all drugs that prolong the QT interval to the same
`extent carry the same risk of causing TdP. The
`incidence of TdP is estimated to be 0.5-8.8% with
`quinidine[IJJ and 2.~.1% with sotaloJ.1 141 The inci(cid:173)
`dence is higher in combination preparations of
`sotalol that include a thiazide diuretic, which in(cid:173)
`duces hypokalaemia,D 5l and lower with racemic
`sotalol in contrast to S-sotalol because of the ~­
`adrenoceptor blocking activity of R-sotalol present
`in the former. Other ancillary properties of the drug
`(e.g. a- or ~-adrenoceptor or calcium channel block(cid:173)
`ing activities) greatly modify the risk of TdP at a
`given duration of QT intervaJ.1 16• 19l Selective, but
`not non-selective, IKr blockers are known to induce
`TdP in anaesthetised rabbits during ar-adrenoceptor
`stimulation.1 171 The use of ~-adrenoceptor blocking
`drugs, with the addition of a-blocking drugs when
`necessary, has been effective in the treatment of
`patients with congenital LQTS[18l and the response
`to adrenergic modulation in these patients appears to
`be genotype specific.f 19l
`
`Not every patient who has a prolongation of the
`QT interval to the same extent will develop TdP.
`The clinical outcome is modulated by not only the
`drug concerned and its plasma concentration but
`also a number of other host factors. Furthermore, the
`risk of developing TdP is not fixed. An individual
`can tolerate a QT-prolonging drug well for many
`months only to be at risk due to an inter-current
`event such as the development of hypokalaemia due
`to diarrhoea and vomiting, or introduction of an
`interacting co-medication.
`
`1.4 Drug Withdrawals Due to TdP
`
`Prenylamine, an effective antianginal drug, was
`the first drug to be withdrawn from the market (in
`1988) because of its high potential
`to cause
`TdP.[20l Table I lists drugs that have been withdrawn
`from various markets during the period 1990-200 I.
`These include eight drugs withdrawn because of
`their propensity to prolong QT interval with or with(cid:173)
`out TdP. Eleven (33%) of the 33 major drugs were
`withdrawn during this period because of their poten(cid:173)
`tial for drug interactions or prolongation of the QT
`interval and/or TdP. Two additional drugs (encai(cid:173)
`nide and flosequinan) were also removed from the
`market for their proarrhythmic potential.
`Drug-induced QT interval prolongation seems to
`be a modem 'epidemic'. This potentially fatal ad(cid:173)
`verse reaction is not only associated with cardio(cid:173)
`vascular drugs but also with over 90 non-cardio(cid:173)
`vascular drugs.Pl In one survey of 2194 cases of TdP
`in the US FDA database, the major drug classes
`involved were cardiac (26.2%), CNS (21.9%), anti(cid:173)
`infectives (19.0%) and antihistamines (11.6%). Of
`these, 92.8% were reported between 1989-98 in
`contrast to only 7.2% between 1969-88.1 101
`Since QT interval prolongation is not an ideal
`surrogate of the risk of TdP, withdrawals of drugs
`that simply prolong the QT interval illustrate the
`need to balance carefully the risks versus benefits
`and the availability of alternative agents. When
`found to induce TdP frequently, drugs such as
`prenylamine,
`terodiline
`terfenadine, astemizole,
`cisapride and Jevacetylmethadol were all withdrawn
`from the market when their benefit/risk ratio was
`determined to be adverse and alternatives were
`available. On the other hand, arsenic trioxide was
`approved and is still on market despite a known high
`
`·- 2004 Adls Data Information BV. All rights reserved.
`
`Drug Safety 2004; 27 (3)
`
`Roxane Labs., Inc.
`Exhibit 1016
`Page 008
`
`

`
`150
`
`Shah
`
`Table I. Reasons for the withdrawal of drugs from the market since
`1990
`
`Drug
`
`Dilevalol
`Triazolam
`Terodiline•
`Encainide
`Fipexide
`Temafloxacin
`
`Benzarone
`Remoxipride
`Alp idem
`Flosequinan
`
`Bendazac
`Soruvidine
`Chlormezanone
`
`Tolrestat
`Minaprine
`Pemoline
`Dexfenfluramine
`
`Year of
`withdrawal
`1990
`1991
`1991
`1991
`1991
`1992
`
`1992
`1993
`1993
`1993
`
`1993
`1993
`1996
`
`1996
`1996
`1997
`1998
`
`Fenfluramine
`
`1998
`
`Terfenadine•
`
`1998
`
`Bromfenac
`
`Ebrotidine
`Sertindole•
`
`Mibefradil
`
`Tolcapone
`Astemizole•
`Trovafloxacin
`Grepafloxacin•
`Troglitazone
`Alosetron
`Cisapride•
`
`1998
`
`1998
`1998
`
`1998
`
`1998
`
`1999
`1999
`1999
`2000
`2000
`2000
`
`Reason(s) for withdrawal
`
`Hepatotoxicity
`Neuropsychiatric reactions
`QTI prolongation and TdP
`Proarrhythmic effects
`Hepatotoxicity
`Hypoglycaemia, haemolytic
`anaemia and renal failure
`Hepatotoxicity
`Aplastic anaemia
`Hepatotoxicity
`Excess mortality, possibly
`due to arrhythmias
`Hepatotoxicity
`Myelotoxicity following Dl
`Hepatotoxicity and severe
`skin reactions
`Hepatotoxicity
`Convulsions
`
`Hepatotoxicity
`Cardiac valvulopathy and
`pulmonary hypertension
`Cardiac valvulopathy and
`pulmonary hypertension
`Dl, QTI prolongation
`and TdP
`Hepatotoxicity following
`prolonged administration
`Hepatotoxicity
`QTI prolongation and
`potential for TdP
`Rhabdomyolysis following Dl
`Concerns on potential Dl
`including risk of TdP
`Hepatotoxicity
`Dl, QTI prolongation and TdP
`
`frequency of TdP associated with its use. Likewise,
`pimozide and thioridazine continue to be available.
`More importantly, because of its low potential for
`extrapyramidal adverse effects and lack of reports of
`TdP despite marked prolongation in QT interval, it
`is planned to allow the re-introduction of sertindole.
`
`2. Clinical Implications
`of Pharmacogenetics
`
`To date, most pharmacogenetic studies have
`focussed on drug metabolising enzymes. The en(cid:173)
`zymes most frequently involved in the primary met(cid:173)
`abolism of drugs are CYP2C9, CYP2Cl9, CYP2D6
`and CYP3A4. The influence of pharmacogenetics in
`determining drug response is best illustrated by drug
`metabolising enzyme such as CYP2D6 and by phar(cid:173)
`macological target such as th~ potassium channels.
`Both exhibit genetic polymorphisms.
`
`2.1 Genetic Factors in Pharmacokinetics
`with Reference to CYP2D6
`
`Depending upon the ability of individuals to
`mediate CYP2D6-dependent hydroxylation of the
`(now obsolete) antihypertensive drug debrisoquine,
`two population phenotypes have been identified -
`extensive metabolisers (EMs) or poor metabolisers
`(PMs).l21l This polymorphism results from autoso(cid:173)
`mal recessive inheritance, in a simple Mendelian
`fashion, of alleles at a single locus mapped to chro(cid:173)
`mosome 22q G.l. Individuals heterozygous for the
`defective allele are EMs with some impairment in
`effecting this reaction, indicating a gene dose effect.
`Since the wild-type allele·(CYP2D6*1) responsible
`for normal functional capacity is dominant, only
`those individuals carrying two CYP2D6 inactivating
`alleles (e.g. CYP2D6*3, CYP2D6*4, CYP2D6*5 or
`CYP2D6*6) are phenotypic PMs.l22l Some pheno(cid:173)
`typically EM
`individuals
`inherit alleles
`(e.g.
`CYP2D6* 10 and CYP2D6* 17) that express the
`enzyme with reduced or altered affinity for certain
`CYP2D6 substratesP3-25l Within the EMs, there is
`another subgroup, termed the ultra-rapid metabolis(cid:173)
`ers, resulting from multiple copies of the alleles
`CYP2D6* 1 or CYP2D6*2 for normal metabolic ca(cid:173)
`pacity.126l Alleles CYP2D6*35 and CYP2D6*41 are
`also associated with ultra-rapid metabolism.
`
`Hepatotoxicity
`QTI prolongation and TdP
`Hepatotoxicity
`lschaemic colitis
`Dl, QTI prolongation
`and TdP
`Droperidol•
`QTI prolongation and TdP
`2001
`Levacetylmethadol• 2001
`Dl, QTI prolongation and TdP
`Rhabdomyolysis following Dl
`2001
`Cerivastatin
`a Drugs withdrawn specifically due to the risk of TdP.
`Dl =drug interactions; QTI = QT interval; TdP = torsade de pointes.
`
`© 2004 Adls Data Information BV. All ~ghts reserved.
`
`Drug Safety 2004; 27 (3)
`
`Roxane Labs., Inc.
`Exhibit 1016
`Page 009
`
`

`
`Pharmacogenetic Aspects of Drug-Induced Torsade de Pointes
`
`151
`
`Table II. Pharmacokinetic consequences of CYP2D6 polymor(cid:173)
`phism
`
`Consequences for PMs vs EMs
`Pharmacokinetic parameter
`2- to 5-fold
`Bioavailabillty
`2- to 6-fold
`Cmax
`AUG
`2· to 5-fold
`Half-life
`2- to 6-fold
`0.1- to 0.5-fold
`Metabolic clearance
`AUC ; area under the plasma concentration-time curve; Cmax ;
`peak plasma concentration; CYP ; cytochrome P450; EMs ;
`extensive metabolisers; PMs ; poor metabolisers.
`
`Although CYP2D6 accounts for only 2% of the
`total liver CYP content, it is responsible for the
`metabolism of well over 20% of the drugs eliminat(cid:173)
`ed by metabolic clearance.l271 CYP2D6 polymor(cid:173)
`phism is the most widely studied genetic polymor(cid:173)
`phism and CYP2D6 isozyme has been shown to
`control the oxidative biotransformation of well over
`60 drugs to date. These include antiarrhythmics, ~­
`blockers, antihypertensives, antianginals, antipsy(cid:173)
`chotics, antidepressants, analgesics as well as a
`number of other miscellaneous drugsP7•28l
`The pharmacokinetic consequences of polymor(cid:173)
`phism in CYP2D6, summarised in table II, are that
`relative to EMs, the PMs experience far greater
`exposure to the parent drugl29l while the reverse is
`true for the metabolites generated by this enzyme.
`PMs may of course activate alternative, otherwise
`dormant and possibly less effective, pathways and
`yield otherwise atypical metabolites.