Nyas_nar_020.tex

Genotypic Approaches to Therapy
in Children
A National Active Surveillance Network
(GATC) to Study the Pharmacogenomics of
Severe Adverse Drug Reactions in Children
COLIN J.D. ROSS,a,d BRUCE CARLETON,a,b,c,d DANA G. WARN,bSUNITA B. STENTON,b,c,d SHAHRAD ROD RASSEKH,eAND MICHAEL R. HAYDENa,d aDepartment of Medical Genetics, University of British Columbia (UBC), Centrefor Molecular Medicine and Therapeutics, Vancouver, British Columbia, CanadabPharmaceutical Outcomes and Policy Innovations Programme, Children’s andWomen’s Health Centre of British Columbia, British Columbia, CanadacFaculty of Pharmaceutical Sciences, University of British Columbia, BritishColumbia, Canadad Child and Family Research Institute, British Columbia, CanadaeDivision of Pediatric Hematology, Oncology, and BMT, Department ofPediatrics, University of British Columbia, British Columbia, Canada ABSTRACT: A striking failure of modern medicine is the debilitating and
lethal consequences of adverse drug reactions (ADRs), which rank as one
of the top 10 leading causes of death and illness in the developed world
with direct medical costs of 137–177 billion annually US$ in the USA.
Although many factors influence the effect of medications (e.g., age, or-
gan function, drug interactions), genetic factors account for 20% to 95%
of drug response variability and play a significant role in the incidence
and severity of ADRs. The field of pharmacogenomics seeks to identify
genetic factors responsible for individual differences in drug efficacy and
ADRs. Pharmacogenomics has led to several genetic tests that provide
clinical dosing recommendations. The Genetic Approaches to Therapy in
Children (GATC) project is a national project established in Canada to
identify novel predictive genomic markers for severe ADRs in children.
An ADR surveillance network has been established in eight of Canada’s
major children’s hospitals, serving up to 75% of all Canadian children.
The goal of the project is to identify patients experiencing specific ADRs
and matched controls, collect DNA samples, and apply genomics-based
Address for correspondence: Michael Hayden, Centre for Molecular Medicine and Therapeutics, 950 West 28th Avenue, Vancouver, BC, Canada V5Z-4H4. Voice: 604-875-3535; fax: 604- 875-3819.
Ann. N.Y. Acad. Sci. 1110: 177–192 (2007). C 2007 New York Academy of Sciences.
doi: 10.1196/annals.1423.020
ANNALS OF THE NEW YORK ACADEMY OF SCIENCES
technologies to identify ADR-associated genetic markers with the goal of
preventing serious ADRs in susceptible children.
KEYWORDS: pharmacogenomics; adverse drug reaction; ADR surveil-
lance
ADVERSE DRUG REACTIONS (ADRS)
A striking failure of modern medicine is the debilitating and lethal conse- quences of ADRs, which rank as one of the leading causes of death and illnessin the USA,1 claiming 100,000–218,000 lives,2,3 and costing 137–177 billionUS$ each year.1,3,4 A paradox of modern drug development is that clinicaltrials provide evidence about efficacy and preliminary safety at a standardizeddose in a large population, while physicians treat individual patients who oftendiffer in their response to drug therapy. ADRs account for 7% of all hospi-tal admissions,1,5 yet retrospective review of ADR reporting shows that lessthan 5% of ADRs are reported.6 Although many factors influence the effectsof medications (e.g., age, organ function, drug interactions), genetic factorsaccount for 20–95% of drug response variability.
PHARMACOGENOMICS TO REDUCE THE OCCURRENCE
OF SEVERE ADRS
The goal of pharmacogenomics is to avoid ADRs and maximize drug ef- ficacy for individual patients.7 Pharmacogenomic studies are performed inpopulations of subjects treated with a specific drug to identify genetic vari-ants that predict drug response and the occurrence of adverse reactions. Onceidentified and validated, a genetic variant is incorporated into a diagnostic testthat will predict a patient’s response to a specific drug. Pharmacogenomicscan improve the benefits and reduce the risks of medications by determiningwhat patients are most likely to respond favorably to a specific medicationand by predicting which individuals are at greater risk for an ADR.8–10 Thisinformation can be used to adjust an individual’s dosage according to his orher likely response to treatment.
EXAMPLES OF PHARMACOGENOMIC TESTING
CURRENTLY IN CLINICAL PRACTICE
Pharmacogenomics has led to several genetic tests that provide clini- cal dosing recommendations. For example, Irinotecan , an anticancer drug,is associated with severe and potentially fatal diarrhea and neutropenia in20–35% of patients.11,12 A specific variant in the UGT1A1 gene reduces ROSS et al.
the inactivation of Irinotecan metabolite and therefore increases the riskof potentially lethal neutropenia and diarrhea.13 In 2004, the Irinotecan la-bel was modified to indicate the role of UGT1A1∗28 polymorphism andrecommendations for lower starting doses in patients homozygous for thepolymorphism.14–17 Azathioprine and its active metabolite 6-mercaptopurine has been a first-line therapy for cancer and autoimmune diseases for over 50 years;18 however, vari-ants in the thiopurine methyltransferase (TPMT) gene reduce the inactivationof azathioprine metabolites, resulting in more active azathioprine metabolitesand leading to severe or fatal myelosuppression and infection in homozygouscarriers of these variants (0.5–1% of people).19–21 Azathioprine-induced toxic-ity can be avoided with TPMT genotyping,22,23 and in 2003 the FDA revised theazathioprine label to inform clinicians about increased risk of severe myelosup-pression for TPMT activity–deficient genotypes and to provide TPMT testingoptions.24 Warfarin is the most commonly prescribed oral anticoagulant drug in the United States, with an estimated 2 million people taking the drug on any givenday,25 but the narrow therapeutic window makes it challenging to determinea patient’s ideal dose. Too high a dose may lead to serious risks of excessivebleeding and intracranial hemorrhage (affecting 5–35% of patients), whilea subtherapeutic dose may lead to the dangerous formation of blood clots(affecting 1–8% of patients).26–28 Variants in the CYP2C9 and VKORC1 genesaccount for 57–63% of variance in warfarin dose,29,30 and in 2005 the FDAvoted to revise the warfarin label with recommendations for pharmacogenetictesting.31 The average cost per warfarin-induced bleeding event is $15,988,32and warfarin pharmacogenetic testing would save an estimated $1.1 billionin U.S. healthcare spending each year, while preventing 17,000 strokes and85,000 serious bleeding incidents.33 THE GATC PROJECT: GENOTYPIC APPROACHES
TO THERAPY IN CHILDREN
The GATC project is a nationwide program established in Canada to identify novel predictive genomic markers of severe ADRs in children for evidence-based individualized drug therapy in children. The underlying hypothesis ofthe project is that genetic polymorphisms in drug metabolism genes cause asignificant portion of concentration-dependent ADRs in children. The ultimategoal of the project is to develop genotype-based dosing guidelines to predictsafety and avoid severe ADRs in children.
SIGNIFICANCE OF ADRS IN CHILDREN
The GATC project aims to address the significant problem of severe ADRs in children. Children are at a greater risk for ADRs, yet there is a remarkable lack ANNALS OF THE NEW YORK ACADEMY OF SCIENCES
of understanding of causation. An estimated 15% of pediatric hospitalizationsresult in ADRs, and 28% of these ADRs are severe.34,35 Health Canada’srecords indicate that 1,193 ADRs were reported in children between 1998 and2002; however, this voluntary reporting identifies fewer than 5% of ADRs.6 Inthe U.S., an estimated 25,000–54,000 children die each year from ADRs, andan estimated 2,500–5,400 children die each year in Canada.1 More than 75% ofpharmaceuticals licensed in North America have never been tested in pediatricpopulations and are used without adequate guidelines for safety or efficacy.36Until recently, it was assumed that children responded to medications as “smalladults.” Clinical practice focused on adjusting childrens’ dosages to accountfor smaller body mass, with the assumption that clinical effects would beequivalent to those observed in adults. However, clinicians now recognize thathuman drug metabolizing systems may in fact develop or change with age.55Children are metabolizing adult-designed, tested, and approved drugs whilebeing equipped with immature or inefficient systems.
GATC ADR SURVEILLANCE
The objectives of the GATC project are to identify patients that experience specific ADRs, collect DNA samples from those patients and their matchedcontrols, and apply genomics-based technologies to identify ADR-associatedgenetic markers. The first goal of the project is to identify patients experiencingsevere ADRs and collect DNA samples from those children. An ADR surveil-lance network has been established in eight of Canada’s major children’s hos-pitals, serving up to 75% of all Canadian children (FIG. 1). The project has alsoestablished collaborations with the Canadian Pediatric Surveillance Program(CPSP), a nationwide network of 2,300 pediatricians across Canada, whichbegan their ADR surveillance in January 2004, as well as with the CanadianC-17 Research Network, a network of all 17 pediatric oncology departmentsacross Canada.
TOWARD A GLOBAL NETWORK OF ADR SURVEILLANCE
The GATC project was approached by investigators at the University of South Australia and the Women’s and Children’s Hospital in Adelaide whohave traveled to Vancouver and are exploring the possibility of becomingthe first international GATC surveillance site. They are currently applyingfor peer-reviewed funding for GATC participation and we have been activelycollaborating with them in this initiative. In the U.S., Dr. Steven Leeder atChildren’s Mercy Hospital, Kansas City, MO, has established an active ADRsurveillance clinician position in the hospital and will join the GATC projectin 2007.
ROSS et al.
FIGURE 1. Canada-wide GATC surveillance network for severe ADRs in children.
STEPS TO ESTABLISH AN ADR SURVEILLANCE NETWORK
The creation of a nationwide hospital-based ADR surveillance network is a complex task that requires: (1) identifying and collaborating with cliniciansand key management and safety personnel at each institution; (2) local/nationaladvertising for ADR surveillance clinicians; (3) performing interviews and ref-erence checks for potential personnel; (4) creation and execution of contractualsite service agreements; (5) acquisition, preparation, and shipping of computerhardware, study, and DNA sampling supplies; (6) organizing courier accountsfor each surveillance clinician for biological sample transport; (7) establishingsecure network connections at each site for communication of study data; (8)authorization from local hospital information systems departments for connec-tion of study laptop computers to local servers; (9) setting up ADR surveillanceclinicians with desk space, printer access, pager, telephone, and internet access;(10) coordinating billing and invoicing processes for each site; (11) creationof a GATC Reference and Training Manual for surveillance clinicians and siteinvestigators; (12) coordinating local ethics board applications; (13) setting uplaboratory agreements for blood draws and safe storage of blood samples; and(14) creation of the custom ADR case ascertainment clinical database.
TRAINING OF CLINICAL SURVEILLANCE PERSONNEL
Training of surveillance personnel includes: ADR identification, reporting, patient enrollment, ethical issues, obtaining informed consent and child assent,advertising the project within the institutions, linkage with other healthcare ANNALS OF THE NEW YORK ACADEMY OF SCIENCES
professionals in the institutions and data transfer. Each surveillor and siteinvestigator is provided with the “GATC Training and Reference Manual,”a 65-page binder outlining procedures and protocols: project advertisement,local laboratory setup for blood draws, DNA sample requirements and stability,DNA sample collection instructions (blood, buccal swabs, and saliva), shippinginstructions, instructions for data entry into the ADR database, target drug lists,and an extensive reference list on the subjects of ADRs, pharmacogenomics,and ADR surveillance.
MAINTAINING AN ACTIVE ADR SURVEILLANCE NETWORK
Once the personnel requirements and the logistics of the network were ad- dressed, the basic infrastructure for GATC was in place. The next steps ofnetwork development focused in two key areas: engaging hospital clinicians inthe project and establishing effective support and communication among thesurveillors within the network. GATC active surveillance was implemented us-ing education and presentation methods that highlighted the project goals andobjectives to various audiences, including hospital administrators, physicians,pharmacists, and other healthcare providers. The overall response to the projectis positive, and support has been offered by numerous clinicians. A requestfor ADR referrals is communicated during department meetings with all inpa-tient and outpatient specialty areas and pharmacy departments. However, themost significant challenge encountered by the surveillors has been develop-ing strategies for keeping the project forefront in clinician’s minds. Healthcareenvironments are pressed for resources, and focused primarily on providingtreatment and interventions to improve the immediate health of children whilepromoting efficient discharge of patients. Successful strategies employed tokeep the GATC project paramount in the daily routines of healthcare profes-sionals include: frequent meetings with key clinicians, posting project adver-tisements throughout public hospital areas, and personally attending follow-upclinics. Support within the network of clinicians is also a challenge as teammembers reside in multiple cities and various time zones. Communication bye-mail and telephone occurs regularly, and monthly group teleconferences areconducted. These serve to promote group problem-solving of both individualand common issues and challenges. The network also holds annual meetingsin Vancouver, which furthers the collaborative nature of this team.
OBSTACLES ENCOUNTERED AND OVERCOME
BY THE NETWORK
With the implementation of the GATC active surveillance network came obstacles that required thoughtful resolution. Some obstacles encountered ROSS et al.
include: lengthy ethical reviews and study approvals that delay the implemen-tation of the study in some sites by several months, significant privacy concernsin regards to patient referral processes and ADR patient access protocols, anddifficulties both initiating and maintaining connections to the study’s virtualprivate network (VPN). The VPN is an essential GATC element as it providesa secure route for anonymous data transfer from individual sites to the studycenter server.
CLINICAL ADR DATA
A secure, password-protected database was created for collection of demo- graphic and clinical information from cases and controls. The ADR databasecaptures relevant clinical information about patients, ADRs, suspected drugs,concurrent medications, past and current medical conditions, and ancestry, aswell as other relevant patient medical information. The project has addressedthe complex process of evaluating possible confounding factors, such as dis-ease state or interacting drugs, and clinical judgment by the ADR surveillorsand clinicians. In the clinical database, the causality of the ADR is gradedusing two instruments: the World Health Organization Collaborating Centrefor International Drug Monitoring Causality Assessment Algorithm, and theNaranjo ADR Probability Scale in order to provide a validated assessment ofADR–drug association before predictive biomarkers are identified. Accurateand detailed clinical data are critical factors in the discovery of ADR-associatedbiomarkers. Szoeke et al. (2006) recently highlighted the limitations of previ-ous pharmacogenomic studies that lacked detailed prospective case ascertain-ment and were missing ancestry data, comorbid conditions, medication doses,and concurrent medication. The GATC surveillance clinicians are trained tocollect these data and capture them within our clinical database. The GATCproject is therefore addressing many of the limitations of previous pharmaco-genetic ADR–drug association studies.
EXAMPLE: SURVEILLANCE NETWORK IDENTIFIES THREE
SIMILAR CASES OF SEVERE IMMUNE- MEDIATED ADRS
Interactions between drugs and the immune system occur as inadvertent consequences of the protective function of the immune system, with drugmolecules or drug-carrier haptens being recognized as “nonself ” by the im-mune system.48 The first severe immune-mediated reaction the GATC projectidentified was in a 10-year-old who previously suffered a life-threatening ana-phylactic reaction to penicillin and presented to a pediatric hospital with a ANNALS OF THE NEW YORK ACADEMY OF SCIENCES
severe reaction to trimethoprim-sulfamethoxazole (cotrimoxazole). Cotrimox-azole is an antibiotic prescribed to treat an upper respiratory tract infection.
The child subsequently developed a fever, mouth lesions, nonpurulent con-junctivitis, and some erythematous rash on the body. The child was diagnosedwith Stevens Johnson Syndrome (SJS), a serious and life-threatening skin con-dition, and was admitted to hospital for supportive care. Shortly thereafter, thechild was transferred to intensive care and placed on a respirator for 5 days.
Serological testing for Mycoplasma pneumoniae IgM was reactive. In addi-tion, the GATC surveillance network identified two other children of the samegender, of similar ages but different ethnicities, who were also known to haveexperienced an allergic reaction to penicillin with skin rashes and who hadexperienced severe reactions due to cotrimoxazole that same year. One childsuffered an SJS reaction and the other was diagnosed with drug hypersensitiv-ity syndrome (DHS). The mean length of hospital stay for the three patientswas 9 days (2–16 days). ADRs mediated by the immune system account for adisproportionate number of fatal and serious adverse reactions, and constitute amajor clinical problem for patients and physicians.48 Sulfonamide-containingdrugs, including the combination drug cotrimoxazole, account for the majorityof drug-related episodes of bullous exfoliative rashes, such as is found in SJSreactions.40 The GATC surveillance network continues to vigilantly monitor all hospi- tal admissions for similar cases. Of note, Health Canada’s ADR MonitoringProgram (CADRMP) received eight reports of cotrimoxazole-induced SJSin 10 full years of passive reporting, compared to the GATC active surveil-lance network’s identification of these three cases in one year. This differ-ence further highlights the benefits of active ADR surveillance versus thetraditional passive surveillance that most national ADR monitoring agenciesemploy.
PRIORITIZATION OF TARGET DRUGS
During the past year the surveillance network identified several immediate opportunities to focus the project’s efforts to reduce the occurrence of severeADRs in children. The project will continue to enroll participants with seriousADRs to any drug, but we recognize that these priority drugs warrant additionalfocus in year 2. In addition to the high-priority ADRs, lists of other high-priority target drugs for neurology, psychiatry, gastroenterology, neonatology,dermatology, and oncology have been generated for the ADR surveillancenetworks. The flexibility of GATC active surveillance enables new drugs to bequickly added to the target drug list as new products come on the market, andspecific ADRs are identified and prioritized by the pharmacogenomics teamand surveillance network.
ROSS et al.
GENOMIC ANALYSES
The second goal of the GATC project is to apply genomics-based technolo- gies to identify ADR-associated genetic markers. Genetic markers are com-pared between patients that suffered ADRs with samples from controls thatreceived the same drugs but did not suffer an ADR. These drug-matched casesand controls are genotyped using the proven and cost-effective technology ofthe Illumina GoldenGate SNP genotyping assay. The samples are genotypedfor a panel of SNP markers, which was designed to capture the genetic varia-tion of 220 key drug metabolism genes (i.e., phase I and II drug metabolismenzymes, drug transporters, drug targets, drug receptors, transcription factors,ion channels, and other disease-specific genes related to the physiologicalpathway of ADRs). This panel builds upon extensive research of GATC inves-tigator, Dr. Michael Phillips, and consists of 1,536 HapMap-based tag SNPsand 1,536 functional SNPs. HapMap-based tag SNPs were selected becausethe sheer number of SNPs in the human genome (>11 million unique SNPs)makes it unfeasible to examine all of the SNPs in a candidate gene, and moreimportantly, direct assay of all existing common polymorphisms is unneces-sary because the genotypes at many of these sites are strongly correlated. Thefunctional SNPs of the panel cause nonsynonymous amino acid changes or areknown to cause changes in enzyme activity or function. This panel of SNPsrepresents an unparalleled opportunity to discover the genetic basis of commonvariants that influence drug metabolism.
INITIAL RESULTS
After 1.5 years, the GATC ADR surveillance network identified, enrolled and collected DNA samples from over 430 severe ADR cases (TABLE 1). Themost common ADR identified by the GATC project in the first year wassevere skin rash (n = 104). Various drug classes were implicated in thesesevere skin rashes including antibiotics (n = 56), anticonvulsants (n = 33),and chemotherapy agents (n = 11). The GATC project focused significantattention in children’s cancer treatment centers and identified several severeADRs including cisplatin ototoxicity (n = 31), anthracycline cardiotoxicity(n = 28), vincristine neurotoxicity (n = 20), L-asparaginase anaphylaxis (n =17) and hypertriglyceridemia (n = 5), and methotrexate leukoencephalopathy(n = 7) (TABLE 1). Over the last 32 years, the “passive” ADR surveillancesystem of Health Canada received four reports of SJS suspected of beingassociated with ibuprofen, while the GATC project had identified three casesof suspected ibuprofen-induced SJS in less than 1 year. The network has provento be highly effective. Preliminary genomic analyses have identified geneticvariants that are highly associated with severe ADRs. Two examples of casesenrolled in GATC are described here.
ANNALS OF THE NEW YORK ACADEMY OF SCIENCES
TABLE 1. Most common severe ADRs identified by the GATC network
EXEMPLARY CASE REPORT 1:
ANTHRACYCLINE-INDUCED CARDIOTOXICITY
A previously healthy 10-year-old presented to a children’s hospital in February 2005 with an abdominal mass. Biopsy confirmed the diagnosis ofa neuroblastoma and the child began a chemotherapy protocol that includeddoxorubicin. A cumulative anthracycline dose of 300 mg/m2 was given andtolerated well until August, when just prior to the last cycle of chemotherapy,the child came to the hospital for a routine CT scan. During the CT scan, thechild became suddenly unwell, developed ventricular tachycardia (250 bpm),and went into cardiogenic shock. The child was intubated and rushed to theintensive care unit (ICU). An echocardiogram showed marked cardiac dysfunc-tion with a percentage of shortening fraction (SF) that dropped from normalon previous evaluation (high 30s) to 13, with virtually no cardiac output. Thechild could not maintain a blood pressure despite use of inotropic and othermedications, and was therefore placed on extracorporeal membrane oxygena-tion (ECMO, heart–lung bypass machine). The child spent the next few weeksin the ICU. The last cycle of chemotherapy was not given on account of thissevere ADR. This child is currently in cancer remission 6 months post treat-ment, and continues on oral cardiac medications with significant limitationsto physical activity.
Anthracyclines are commonly used to treat childhood leukemia and solid tu- mors. Anthracycline-induced cardiotoxicity affects 6–7% of patients on stan-dard doses, with an increased risk in children <15 years, and a still higher ROSS et al.
risk in children <4 years of age55,56 with mortality rates greater than 50%.57Anthracycline-cardiotoxicity is a cumulative dose-dependent phenomenon.
However, severe cardiotoxicity in children may occur at any dose, and sig-nificant decreases in left ventricular ejection fraction have been documentedat <300 mg/m2, while some patients can tolerate doses higher than 1,000mg/m2 (Ref. 55). This observed heterogeneity could be explained by geneticsusceptibility and gene–environment interactions. Elucidation of these factorscould lead to more informed and patient-specific dosage individualization inthe future. A uniform lowering of dose would reduce the risk of cardiotoxicity,but would be more than offset by increased cancer-related morbidity.
The GATC project has focused efforts to identify children with anthracycline-induced cardiotoxicity and to identify anthracycline-ADR–associated genetic variants and initiated retrospective recruitment of patientswith anthracycline-induced heart failure.
Approximately 22% of hospital admissions for pediatric cancer are caused by ADRs.58 With the improved survival of these patients in past 40 years (from10% in the 1960s to more than 80% today), the number of cancer survivorswith permanent drug-related disabilities has increased accordingly. The GATCsurveillance network identified a high prevalence of severe ADRs in pediatriconcology clinics (>30% of identified ADRs) and recognized that pediatriconcology warranted additional focus.
EXEMPLARY CASE REPORT 2: CODEINE-INDUCED
INFANT MORTALITY
Codeine is the most widely prescribed opioid analgesic in pediatrics37 and the American Academy of Pediatrics considers codeine use in mothers compat-ible with breast feeding.38 Dr. Jim Cairns, the deputy chief coroner of Ontario,Canada, initially identified a lethal opioid overdose in a 13-day-old infant afterthe mother was given Tylenol 3 (a combination of acetaminophen and 30 mgcodeine) for obstetric pain relief. Initially, the mother took two tablets fourtimes a day but reduced the dose by half on day 7 postpartum because ofsomnolence and constipation. The mother expressed breast milk and storedit in the freezer because of poor feeding by the infant. Maternal milk fromthe last days of the baby’s life contained 87 ng/mL of morphine, a concentra-tion 10–20 times higher than expected, and the infant’s blood contained lethallevels of morphine.39 Dr. Gideon Koren, the GATC investigator in Toronto fol-lowed up the case and with the GATC’s Dr. Steven Leeder and discovered thatthe mother carried a cytochrome P450 2D6 (CYP2D6) gene duplication.39Codeine is an analgesic with central nervous system (CNS)-depressant ef-fects secondary to its biotransformation to morphine, a reaction catalyzed byCYP2D6.41 The CYP2D6 duplication genotype (ultrarapid metabolizer phe-notype) of the mother caused a substantially increased production of morphine ANNALS OF THE NEW YORK ACADEMY OF SCIENCES
from codeine and the accumulation of toxic levels of morphine in the mother’sbreast milk, which, when fed to the infant, resulted in lethal levels of mor-phine in the infant’s blood. The infant possessed two normal CYP2D6 alleles,but since the CYP2D6 pathway is incompletely developed in neonates, thisplaces infants at particular risk for dose-related adverse effects.42 Morphine isnormally eliminated through biotransformation to M3G and M6G by severalUDP-glucuronosyltransferases (UGT1A subfamily and UGT2B7).43 While most individuals are normal metabolizers of codeine to morphine, approximately 5.5% of people currently living in Europe are CYP2D6 ultra-metabolizers.44 Approximately 29% of Ethiopians, 10% of people with SouthEuropean ancestry, and 1–2% of people with North European ancestry carryCYP2D6 gene duplications with an ultrarapid metabolizer phenotype.45,46 Themother in this case was of Bangladeshi origin.
Significance
Codeine is the most widely prescribed opioid analgesic in pediatrics.47 In Canada, approximately 129,000 infants are exposed to codeine from theirbreastfeeding mothers each year, based on a total of 340,000 births per yearof which 73% of the infants are breastfed and approximately 52% of mothersreceive codeine after childbirth (Statistics Canada, 2006). It is possible thatin some cases this codeine-induced ADR may be mistaken for sudden infantdeath syndrome (SIDS) in young infants.39 The reputedly low incidence ofopioid-related adverse effects has made codeine popular for use in younger agegroups including neonates.49,50 Few clinical studies of the analgesic efficacyor adverse effects of codeine in children have been undertaken.
Removing codeine from the market is not the solution to this problem; it would leave the majority of new mothers without the benefit of this other-wise safe and effective medication. Rather, the solution may lie in identifyingthe individuals at risk for this ADR before taking codeine, so that they can beoffered safe alternatives (i.e., acetaminophen or NSAIDs) or a reduced doseof codeine that will lead to comparable drug exposure despite the differencesin genetic make-up. A clinical study is now under way to further validate thecodeine/morphine ADR-associated genotypes.
The GATC project has proven that the creation of a nationwide active ADR surveillance network is indeed possible, but does require significant time, en-ergy, and planning to implement and sustain. Early success of an active surveil-lance network is evident in the substantial GATC enrollment of more than 430cases of ADRs and over 1,800 matched controls in the first year of operation.
ROSS et al.
Clinical trials provide evidence about the efficacy and preliminary safety of adrug at usual doses in a particular subset of our population, but physicians treatindividual patients who vary widely in their responses to drug therapy. Conse-quently, ADRs are one of the top 10 leading causes of death and illness in thedeveloped world. Genetic factors account for a significant level of variationin drug response and can play a significant role in the incidence and severityof ADRs. Pharmacogenomics seeks to identify genetic factors responsible fordifferences in drug efficacy to improve drug safety. Pharmacogenomics hasled to several DNA-based tests to improve drug selection, optimize dosing,and minimize the risk of toxicity. Advances in higher throughput genotyp-ing and sequencing technologies and the availability of international publiclyavailable databases, such as the International HapMap Project, are enablingnew advances in pharmacogenomics. In 2005, the FDA released guidelinesfor pharmacogenomics in a report: “Guidance for Industry PharmacogenomicData Submissions.”51 The report focuses on encouraging the use of pharma-cogenomics in drug development, encouraging industry to voluntarily sharepharmacogenomic data, and evaluation and validity of biomarkers as indica-tors of drug response.51,52 Recently, other regulatory agencies, both in Canadaand in Europe, have now provided draft guidelines for industries on usingpharmacogenomics in clinical trials and drug submissions.53,54 ACKNOWLEDGMENT
We would like to acknowledge the outstanding contributions from the mem- bers of the GATC ADR Surveillance Network: Anne Smith, Bobby Dhami,Kyla Harris, Cheri Nijssen-Jordan, David Johnson, Shanna Chan, Kevin Hall,Becky Malkin, Michael Rieder, Facundo Garcia Bournissen, Miho Inoue,Shinya Ito, Gideon Koren, Elaine Wong, Regis Vaillancourt, Pat Elliot Miller,Pierre Barret, Denis Lebel, Jean Francois Bussiers, Carol-anne Osborne, Dar-lene Boliver, and Margaret Murray. The GATC project has received supportfrom Genome Canada, the Eli Lilly and Pfizer companies, Canadian GeneticDiseases Network, Merck, B.C. Provincial Health Services Authority, GenomeB.C., Janssen Ortho, B.C. Child and Family Research Institute, Canadian GeneCure Foundation, Illumina, and IBM.
REFERENCES
1. LAZAROU, J., B.H. POMERANZ & P.N. COREY. 1998. Incidence of adverse drug reactions in hospitalized patients: a meta-analysis of prospective studies. JAMA
279: 1200–1205.
2. ERNST, F.R. & A.J. GRIZZLE. 2001. Drug-related morbidity and mortality: updating the cost-of-illness model. J. Am. Pharm. Assoc. (Wash.) 41: 192–199.
ANNALS OF THE NEW YORK ACADEMY OF SCIENCES
3. WHITE, T.J., A. ARAKELIAN & J.P. RHO. 1999. Counting the costs of drug-related adverse events. Pharmacoeconomics 15: 445–458.
4. PHILLIPS, K.A., D. VEENSTRA, S. VAN BEBBER, et al. 2003. An introduction to cost-effectiveness and cost-benefit analysis of pharmacogenomics. Pharmacoge-
nomics 4: 231–239.
5. PIRMOHAMED, M., S. JAMES, S. MEAKIN, et al. 2004. Adverse drug reactions as cause of admission to hospital: prospective analysis of 18 820 patients. BMJ
329: 15–19.
6. FLETCHER, A.P. 1991. Spontaneous adverse drug reaction reporting vs. event mon- itoring: a comparison. J. R. Soc. Med. 84: 341–344.
7. WOLF, C.R., G. SMITH & R.L. SMITH. 2000. Science, medicine, and the future: pharmacogenetics. BMJ 320: 987–990.
8. TATE, S.K. & D.B. GOLDSTEIN. 2004. Will tomorrow’s medicines work for every- one? Nat. Genet. 36: S34–S42.
9. GOLDSTEIN, D.B., S.K. TATE & S.M. SISODIYA. 2003. Pharmacogenetics goes ge- nomic. Nat. Rev. Genet. 4: 937–947.
10. SADEE, W. & Z. DAI. 2005. Pharmacogenetics/genomics and personalized medicine. Hum. Mol. Genet. 14 Spec No. 2: R207–R214.
11. BEUTLER, E., T. GELBART & A. DEMINA. 1998. Racial variability in the UDP- glucuronosyltransferase 1 (UGT1A1) promoter: a balanced polymorphism for
regulation of bilirubin metabolism? Proc. Natl. Acad. Sci. USA 95: 8170–
8174.
12. IYER, L., C.D. KING, P.F. WHITINGTON, et al. 1998. Genetic predisposition to the metabolism of irinotecan (CPT-11). Role of uridine diphosphate glucuronosyl-
transferase isoform 1A1 in the glucuronidation of its active metabolite (SN-38)
in human liver microsomes. J. Clin. Invest. 101: 847–854.
13. JANNE, P.A., S. GURUBHAGAVATULA, B.Y. YEAP, et al. 2004. Outcomes of pa- tients with advanced non-small cell lung cancer treated with gefitinib (ZD1839,
“Iressa”) on an expanded access study. Lung Cancer 44: 221–230.
14. INNOCENTI, F., S.D. UNDEVIA, L. IYER, et al. 2004. Genetic variants in the UDP- glucuronosyltransferase 1A1 gene predict the risk of severe neutropenia of
irinotecan. J. Clin. Oncol. 22: 1382–1388.
15. MARCUELLO, E., A. ALTES, A. MENOYO, et al. 2004. UGT1A1 gene variations and irinotecan treatment in patients with metastatic colorectal cancer. Br. J. Cancer
91: 678–682.
16. ROUITS, E., M. BOISDRON-CELLE, A. DUMONT, et al. 2004. Relevance of different UGT1A1 polymorphisms in irinotecan-induced toxicity: a molecular and clinical
study of 75 patients. Clin. Cancer Res. 10: 5151–5159.
17. FOOD AND DRUG ADMINISTRATION. 2004. Advisory Committee for Pharmaceuti- cal Science, Clinical Pharmacology Subcommittee. http://www.fda.gov/ohrms/dockets/ac/04/briefing/2004–4079b1.htm.
18. REMY, C.N. 1963. Metabolism of thiopyrimidines and thiopurines. S-methylation with S-adenosylmethionine transmethylase and catabolism in mammalian tis-
sues. J. Biol. Chem. 238: 1078–1084.
19. WOODSON, L.C. & R.M. WEINSHILBOUM. 1983. Human kidney thiopurine methyl- transferase. Purification and biochemical properties. Biochem. Pharmacol. 32:
819–826.
20. WEINSHILBOUM, R.M., D.M. OTTERNESS & C.L. SZUMLANSKI. 1999. Methylation pharmacogenetics: catechol O-methyltransferase, thiopurine methyltransferase,
and histamine N-methyltransferase. Annu. Rev. Pharmacol. Toxicol. 39: 19–52.
ROSS et al.
21. MCLEOD, H.L., E.Y. KRYNETSKI, M.V. RELLING, et al. 2000. Genetic polymorphism of thiopurine methyltransferase and its clinical relevance for childhood acute
lymphoblastic leukemia. Leukemia 14: 567–572.
22. LENNARD, L., J.A. VAN LOON & R.M. WEINSHILBOUM. 1989. Pharmacogenetics of acute azathioprine toxicity: relationship to thiopurine methyltransferase genetic
polymorphism. Clin. Pharmacol. Ther. 46: 149–154.
23. MCLEOD, H.L., E.Y. KRYNETSKI, M.V. RELLING, et al. 2000. Genetic polymorphism of thiopurine methyltransferase and its clinical relevance for childhood acute
lymphoblastic leukemia. Leukemia 14: 567–572.
24. FOOD AND DRUG ADMINISTRATION 2003. Pediatric Oncology Subcommittee of the Oncologic Drugs Advisory Committee. http://www.fda.gov/ohrms/dockets/ac/03/minutes/3971M1.htm.
25. CHAN, Y.C., D. VALENTI, A.O. MANSFIELD & G. STANSBY. 2000. Warfarin induced skin necrosis. Br. J. Surg. 87: 266–272.
26. LINKINS, L.A., P.T. CHOI & J.D. DOUKETIS. 2003. Clinical impact of bleeding in patients taking oral anticoagulant therapy for venous thromboembolism: a meta-
analysis. Ann. Intern. Med. 139: 893–900.
27. PALARETI, G., N. LEALI, S. COCCHERI, et al. 1996. Bleeding complications of oral anticoagulant treatment: an inception-cohort, prospective collaborative study
(ISCOAT). Italian Study on Complications of Oral Anticoagulant Therapy.
Lancet 348: 423–428.
28. BUDNITZ, D.S., D.A. POLLOCK, K.N. WEIDENBACH, et al. 2006. National surveil- lance of emergency department visits for outpatient adverse drug events. JAMA
296: 1858–1866.
29. TAKAHASHI, H., G.R. WILKINSON, E.A. NUTESCU, et al. 2006. Different contribu- tions of polymorphisms in VKORC1 and CYP2C9 to intra- and inter-population
differences in maintenance dose of warfarin in Japanese, Caucasians and African-
Americans. Pharmacogenet. Genomics 16: 101–110.
30. VECSLER, M., R. LOEBSTEIN, S. ALMOG, et al. 2006. Combined genetic profiles of components and regulators of the vitamin K-dependent gamma-carboxylation
system affect individual sensitivity to warfarin. Thromb. Haemost. 95: 205–211.
31. FOOD AND DRUG ADMINISTRATION. 2005. Clinical Pharmacology Subcommittee Meeting of the Advisory Committee for Pharmaceutical Science, November 15.
32. FANIKOS, J., N. GRASSO-CORRENTI, R. SHAH, et al. 2005. Major bleeding compli- cations in a specialized anticoagulation service. Am. J. Cardiol. 96: 595–598.
33. MCWILLIAM, A., R. LUTTER & C. NARDINELLI. 2006. Health care savings from personalizing medicine using genetic testing: the case of warfarin. AmericanEnterprise Institute-Brookings Joint Center, Working Paper 06–23.
34. GONZALEZ-MARTIN, G., C.M. CAROCA & E. PARIS. 1998. Adverse drug reactions (ADRs) in hospitalized pediatric patients. A prospective study. Int. J Clin. Phar-
macol. Ther. 36: 530–533.
35. MARTINEZ-MIR, I., M. GARCIA-LOPEZ, V. PALOP, et al. 1999. A prospective study of adverse drug reactions in hospitalized children. Br. J. Clin. Pharmacol. 47:
681–688.
36. LEEDER, J.S. 2003. Developmental and pediatric pharmacogenomics. Pharmacoge- nomics 4: 331–341.
37. DE LIMA, J., A.R. LLOYD-THOMAS, R.F. HOWARD, et al. 1996. Infant and neonatal pain: anaesthetists’ perceptions and prescribing patterns. BMJ 313: 787.
38. AMERICAN ACADEMY OF PEDIATRICS, COMMITTEE ON DRUGS. 2001. Transfer of drugs and other chemicals into human milk. Pediatrics 108: 776–789.
ANNALS OF THE NEW YORK ACADEMY OF SCIENCES
39. KOREN, G., J. CAIRNS, D. CHITAYAT, et al. 2006. Pharmacogenetics of morphine poisoning in a breastfed neonate of a codeine-prescribed mother. Lancet 368:
704.
40. SHEPHERD G.M. 2003. Hypersensitivity reactions to drugs: evaluation and man- agement. Mt. Sanai J. Med. 70: 113–125.
41. SINDRUP, S.H. & K. BROSEN. 1995. The pharmacogenetics of codeine hypoalgesia.
Pharmacogenetics 5: 335–346.
42. BOUWMEESTER, N.J., B.J. ANDERSON, D. TIBBOEL, et al. 2004. Developmental phar- macokinetics of morphine and its metabolites in neonates, infants and young
children. Br. J. Anaesth. 92: 208–217.
43. STONE, A.N., P.I. MACKENZIE, A. GALETIN, et al. 2003. Isoform selectiv- ity and kinetics of morphine 3- and 6-glucuronidation by human UDP-
glucuronosyltransferases: evidence for atypical glucuronidation kinetics by
UGT2B7. Drug Metab. Dispos. 31: 1086–1089.
44. INGELMAN-SUNDBERG, M. 2004. Pharmacogenetics of cytochrome P450 and its applications in drug therapy: the past, present and future. Trends Pharmacol.
Sci. 25: 193–200.
45. BRADFORD, L.D. 2002. CYP2D6 allele frequency in European Caucasians, Asians, Africans and their descendants. Pharmacogenomics 3: 229–243.
46. GASCHE, Y., Y. DAALI, M. FATHI, et al. 2004. Codeine intoxication associated with ultrarapid CYP2D6 metabolism. N. Engl. J. Med. 351: 2827–2831.
47. DE LIMA, J., A.R. LLOYD-THOMAS, R.F. HOWARD, et al. 1996. Infant and neonatal pain: anaesthetists’ perceptions and prescribing patterns. BMJ 313: 787.
48. RIEDER M.J. 1993. Immunopharmacology and adverse drug reactions. Clin. Phar- macol. 33: 316–323.
49. HUSBAND, A.D. & A. DAVIS. 1996. Pain after tonsillectomy. Clin. Otolaryngol.
Allied Sci. 21: 99–101.
50. LLOYD-THOMAS, A.R. 1990. Pain management in paediatric patients. Br. J. Anaesth.
64: 85–104.
51. U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES. 2005. Guidance for in- dustry: pharmacogenomic data submissions. http://www.fda.gov/cder/guidance/6400fnl.pdf. FDA Guidance Documents.
52. SALERNO, R.A. & L.J. LESKO. 2006. Three years of promise, proposals, and progress on optimizing the benefit/risk of medicines: a commentary on the 3rd FDA-
DIA-PWG-PhRMA-BIO pharmacogenomics workshop. Pharmacogenomics J.
6: 78–81.
53. HEALTH CANADA. 2006. Draft Guidance Document: Submission of Pharmacoge- nomic Information March 2006. Health Canada Guidelines.
54. THE EUROPEAN FEDERATION OF PHARMACEUTICAL INDUSTRIES AND ASSOCIATIONS (EFPIA). 2006. EFPIA key messages surrounding pharmacogenetics externalrelease 2006. EFPIA.
55. LEEDER S. et al. 2000. Drug Metab. Rev. 32(S2): 281.
56. KREMER, L.C., E.C. VAN DALEN, M. OFFRINGA, et al. 2002. Frequency and risk
factors of anthracycline-induced clinical heart failure in children: a systemic
review. Ann. Oncol. 13: 503–512.
57. SILBER, J.H., R.I. JAKACKI, R.L. LARSEN, et al. 1993. Increased risk of cardiac dysfunction after anthracyclines in girls. Med. Pediatr. Oncol. 21: 477–479.
58. COHN, J.N. 1996. Prognosis in congestive heart failure. J. Card. Fail. 2: S225–229.
59. MITCHELL, A.A., P.G. LACOUTURE, J.E. SHEEHAN, et al. 1988. Adverse drug reac-
tions in children leading to hospital admission. Pediatrics. 82: 24–29.

Source: http://www.c17.ca/files/2013/2769/1308/Rassekh%20Publication%20Ann%20NY%202007.pdf