Is personalized medicine a dream or a reality?

http://informahealthcare.com/lab ISSN: 1040-8363 (print), 1549-781X (electronic) Crit Rev Clin Lab Sci, Early Online: 1–11 ! 2014 Informa Healthcare USA, Inc. DOI: 10.3109/10408363.2014.950407

REVIEW ARTICLE

Critical Reviews in Clinical Laboratory Sciences Downloaded from informahealthcare.com by UMEA University Library on 09/07/14 For personal use only.

Is personalized medicine a dream or a reality? Bridget L. Morse and Richard B. Kim Department of Medicine, Division of Clinical Pharmacology, Schulich School of Medicine and Dentistry, Western University, London, ON, Canada

Abstract

Keywords

Personalized medicine is an emerging field with a goal of applying genomic information as a predictor of disease risk as well as individualization of drug therapy. For optimization of drug therapy, significant progress has been made in the past decade in linking genetic variation in genes associated with drug disposition to prediction of drug response and adverse reactions. For most drugs in clinical use, the interplay of many factors, including genetics, demographics, drug–drug interactions, disease states and the environment, result in the interindividual variability observed during drug therapy. Broadly speaking, such determinants of drug response are mediated through modulation of drug concentrations reflective of pharmacokinetic factors, as well as drug targets, often referred to as pharmacodynamics. It is clear that for personalized medicine to become clinically meaningful, genomic as well as clinical and environmental influences must be considered together. We show, for a number of drugs in clinical use, that genomics-guided treatment options not only are becoming feasible but are also on the cusp of showing superiority in terms of clinical outcomes as well as cost-benefit. One of the most widely studied drugs with regard to genomics-guided dosing options is the oral anticoagulant, warfarin. Genetic polymorphisms in the gene encoding cytochrome P450 2C9 (CYP2C9) and those in the target gene responsible for the warfarin anticoagulant effect, vitamin K epoxide reductase (VKORC1), account for much of the variability in the warfarin maintenance dose; however, routine genotyping in warfarin therapy remains controversial. We will outline the importance of understanding all of the variables that mediate warfarin response as the prerequisite to successful utilization of genotype-guided warfarin therapy. Similarly, HMG Co-A reductase inhibitors, commonly known as statins, also display wide interindividual variability in plasma concentration, response and toxicity due in part to polymorphisms in transporter genes, including SLCO1B1 and ABCG2. Genetic factors are also important considerations in treatment with other therapeutic agents discussed, including clopidogrel and tamoxifen. Implementation of personalized medicine-based treatment options for these and other drugs, the pharmacokinetics or pharmacodynamics of which are impacted by functional genetic variations, will require overcoming a number of challenges, including cost, turnaround time, and demonstration of clinical benefit, as well as better training of health care professionals about genomics in general, and pharmacogenomics in particular.

Clopidogrel, HMG Co-A reductase inhibitors, interindividual variability, pharmacogenomics, pharmacokinetics, pharmacodynamics, tamoxifen, warfarin History Received 19 January 2014 Revised 9 May 2014 Accepted 28 July 2014 Published online 2 September 2014

Abbreviations: ACCP: American College of Chest Physicians; BCRP: breast cancer resistance protein; CPIC: Clinical Pharmacogenetics Implementation Consortium; CYP: cytochrome P450; ESRD: end-stage renal disease; INR: international normalized ratio; LDL: low-density lipoprotein; OATP: organic anion transporting polypeptide; P2Y12 receptor: ADP receptor expressed on the surface of platelets; RCT: randomized controlled trial; SNP: single nucleotide polymorphism; SNRI: serotonin–norepinephrine reuptake inhibitor; SSRI: selective serotonin reuptake inhibitor; VKOR: vitamin K epoxide reductase (encoded by VKORC1 gene); VTE: venous thromboembolism; WRAPID: Warfarin Regimen using A Pharmacogenetics-guided Initiation Dosing

Referees: Dr. Matthias Schwab, Professor and Chair of Clinical Pharmacology, Head of the Dr. Margarete Fischer-Bosch-Institute of Clinical Pharmacology, Stuttgart, and Department of Clinical Pharmacology, Institute of Experimental and Clinical Pharmacology and Toxicology, University Hospital, Tu¨bingen, Germany, and Dr. Shinya Ito, Division Head, Clinical Pharmacology and Toxicology, The Hospital for Sick Children, and Professor, Medicine, Pharmacology & Pharmacy, Department of Paediatrics, University of Toronto, Toronto, Canada. Address for correspondence: Dr Richard B. Kim, University Hospital, 339 Windermere Rd, Rm B9-132, London, ON N6A 5A5 Canada. E-mail: [email protected]

2

B. L. Morse & R. B. Kim


Introduction Interindividual variation in drug responsiveness has been an ongoing obstacle to the effective and optimal use of therapeutic agents. Unpredictable differences in drug response can manifest as severe drug toxicity in some patients and loss of drug efficacy in others. In addition, unexpected drug toxicity or loss of efficacy lead to greater financial burden in terms of the overall health care costs; some reports indicate that adverse drug reactions cost more annually than the medications themselves1. Personalized medicine acknowledges the notion that not all patients requiring drug therapy are going to respond to that agent in an identical manner, due to difference in genetic makeup as well as dietary and environmental influences. The hopes and goals of personalized medicine have been: (1) to identify therapeutic agents for which interindividual variability is a key issue for effective treatment; (2) to address/identify factors leading to interindividual variability of these agents; (3) to treat each patient with the right drug and right dose on an individualized basis by considering all these factors; and (4) to aid in the prevention of adverse reactions and thereby more effectively cure or ameliorate disease progression. This review will summarize clinically relevant factors associated with interindividual variability, describe the progress in personalized medicine of therapeutic agents of current controversy, and address the challenges and strategies for implementation of personalized medicine in clinical care.

Determinants of interindividual variability in drug response Differences in pharmacokinetics versus pharmacodynamics Pharmacokinetics is often referred to as ‘‘what the body does to the drug’’, i.e. a drug’s absorption, distribution, metabolism

Crit Rev Clin Lab Sci, Early Online: 1–11

and elimination properties. Conversely, pharmacodynamics is often referred to as ‘‘what the drug does to the body’’, i.e. the mechanism(s) by which a drug elicits its effect. Effective drug response requires a myriad of both pharmacokinetic and pharmacodynamic processes to occur in a predictable manner (Figure 1). Once a drug is administered, it must reach sufficient systemic exposure in terms of blood levels and then be distributed from the blood to its site of action. There are a few exceptions to this, including drugs administered directly to the site of action, e.g. topical ointment administration, or drugs which exert most of their effects soon after intestinal absorption and prior to significant systemic exposure, e.g. the statin class of lipid lowering drugs, which work by targeting liver HMG-CoA reductase in the liver. However, the majority of therapeutic agents require adequate systemic exposure for effective drug response, and therefore require desirable pharmacokinetic properties that are mostly dependent on drug-metabolizing enzymes and drug transporter proteins expressed in organs such as the intestine, liver and kidney. Not surprisingly, alterations in the function of these proteins can alter drug response because of changes in the overall concentration of the drug at its site of action. Drug response can also be directly affected through alterations in pharmacodynamic components, e.g. target/receptor density. Therefore, genetic variations that alter the expression or function of genes involved in drug disposition or drug targets are now widely recognized as sources of variability in drug responsiveness. Additional intrinsic factors (Figure 2) that likely reflect prevalence of common genetic variations include ethnicity or race. Moreover, clinical conditions such as age, renal and hepatic function that are known to alter drug disposition or response, as well as the impact of disease, must also be considered. In order to truly offer personalized medicine-based patient care, in addition to the intrinsic factors noted above, extrinsic factors such as diet, environmental chemicals (e.g. smoking), alcohol and drug–drug interactions must also be taken into account.

Figure 1. Overview of pharmacokinetic and pharmacodynamic processes. Adapted from Jusko et al.93.

Personalized medicine


DOI: 10.3109/10408363.2014.950407

3

Figure 2. Factors leading to interindividual variability in drug response.

Hot topics in personalized medicine Warfarin Warfarin is the most highly prescribed oral anticoagulant in the United States, with over 33 million prescriptions dispensed in 20112. Although several new oral anticoagulants have emerged recently, these new therapies have limitations, including irreversibility, uncertainty in therapeutic monitoring endpoints, narrow indications and cost3. Given the wide body of evidence that support warfarin efficacy for many indications, and the limitations of new oral anticoagulants that warfarin lacks, it is unlikely that warfarin therapy will become obsolete. Warfarin’s primary limitation, however, is its narrow therapeutic index that results in substantial interindividual variability in warfarin dosage requirements. This has have been known since its early use4, and currently remains the primary concern of warfarin treatment. Warfarin is a vitamin K antagonist that elicits its anticoagulant effect through inhibition of vitamin K epoxide reductase (VKOR, encoded by the VKORC1 gene); this enzyme catalyzes the rate-limiting step in the reduction of vitamin K epoxide to vitamin K, an essential cofactor for the activation of clotting factors II, VII, IX, and X5. Warfarin is dispensed as a racemic mixture, with S-warfarin having 2- to 5-times greater anticoagulant activity compared to R-warfarin5–7. Clearance of warfarin from the body occurs almost solely through hepatic metabolism, with cytochrome P450 (CYP) 2C9 being the primary enzyme responsible for S-warfarin metabolism and CYPs 1A2, 3A4 and 2C19 being responsible for metabolism of R-warfarin5. The effect of warfarin is measured by the international normalized ratio (INR), the ratio of prothrombin time in the test compared to control samples; for most indications, the target INR for warfarin therapy is 2.0–3.0. Traditionally, warfarin has been dosed on an individual basis through trial-and-error, with dose increases/decreases made per INR values. Considering the high rate of adverse events associated with warfarin use,

the need for more predictive dosing has become evident, and the development of clinical algorithms and computer-based programs for predicting warfarin dose has become a subject of interest, particularly algorithms that combine both the non-genetic and genetic factors that contribute to warfarin interindividual variability. Non-genetic factors related in interindividual variability The first relationship described that addressed warfarin interindividual variability was that of the therapeutic warfarin dose with age; a decrease in warfarin dosage with an increase in age was first observed more than three decades ago8. Indeed, in more recent analyses, age continues to appear as a significant factor relating to warfarin dose9–11. The reasons for the decreased therapeutic dose in the elderly are unclear; however, it likely involves changes in both the pharmacokinetic and the pharmacodynamic properties of warfarin7. Other variables commonly included in clinical algorithms are weight/body mass index, gender, smoking status (due to CYP induction), and indication for warfarin treatment. Race also plays a major role in warfarin interindividual variability, with African-American patients being particularly difficult to treat. This is likely due to additional genetic and non-genetic factors in this population, and separate clinical algorithms have been developed for this subset of patients12,13. Because warfarin is cleared by the liver, caution is clearly warranted when warfarin is used in subjects with hepatic impairment. A more recently acknowledged and often overlooked issue concerning warfarin interindividual variability is that due to differences in renal function. Although warfarin is excreted negligibly into the urine, it is known that physiologic changes due to chronic kidney disease can lead to changes in hepatic metabolism14. It was demonstrated previously that warfarin clearance is indeed lower in subjects with end-stage renal disease (ESRD)15, and many studies have

4




Table 1. Factors related to interindividual variability of commonly used therapeutic agents. Therapeutic agent

Clinical characteristics

Genes of interest

Interacting drugs

Atorvastatin

Age

SLCO1B1, ABCG2

Fluvastatin Pitavastatin Pravastatin Simvastatin Rosuvastatin Warfarin

Age Age Age, gender Age Age, race Age, indication, race, smoking status, renal function Age, diabetes

ABCG2 SLCO1B1 SLCO1B1 SLCO1B1 SLCO1B1, ABCG2 CYP2C9, VKORC1

OATP1B1 inhibitors (cyclosporine, gemfibrozil, rifampin), CYP3A4 inhibitors (protease inhibitors, cyclosporine) Cholestyramine OATP1B1 inhibitors OATP1B1 inhibitors, cholestyramine OATP1B1 inhibitors, CYP3A4 inhibitors OATP1B1/BCRP inhibitors (protease inhibitors) Amiodarone, NSAIDs, aspirin, antiplatelets, CYP inhibitors, antibiotics CYP2C19 inhibitors (proton pump inhibitors, calcium channel blockers) CYP2D6 inhibitors (selective serotonin reuptake inhibitors)

Clopidogrel Tamoxifen

Vitamin D status/seasonal changes

CYP2C19 CYP2D6, CYP3A4

demonstrated a significant risk of bleeding with warfarin therapy in these subjects16. Interestingly, our recent evaluation of clinical and genetics characteristics related to warfarin dose demonstrated a significant positive correlation between dose and glomerular filtration rate9. A recent study evaluating warfarin dose requirements in subjects with ESRD found that these subjects needed a 24% lower dose compared to subjects with normal renal function, and that subjects with ESRD spent less time in the therapeutic range17. Yet another study found the need for significantly lower maintenance doses of warfarin even in subjects with moderate renal dysfunction18. Although renal function is not included in many clinical algorithms, it is also not generally evaluated as a potential factor due to negligible warfarin renal excretion. However, renal function should not be ignored when determining warfarin dose, and particular care should be taken in subjects with severe renal dysfunction. Given warfarin’s narrow therapeutic index as well as the role of multiple drug-metabolizing enzymes in warfarin clearance, warfarin is particularly susceptible to interindividual variability through drug–drug interactions. Specifically, since the metabolism of the more active isomer involves primarily one enzyme, CYP2C9, inhibitors of this enzyme can lead to drastic changes in warfarin response. The interaction of most concern is that with amiodarone, because this drug inhibits CYP2C9, along with other CYPs, and because amiodarone and warfarin are often used for the same indication of atrial fibrillation and are therefore commonly co-prescribed. Chronic treatment with other CYP inhibitors can also warrant dosage decreases due to the effects on both warfarin enantiomers7. Pharmacodynamic interactions should also be considered, and treatment with other drugs that increase the risk of bleeding should be used with caution (Table 1). Genetic factors related in interindividual variability Evidence for the role of genetic factors in the interindividual variability of warfarin is considerable, and includes both pharmacokinetic and pharmacodynamic components. The first suggestion that genetics might be relevant to warfarin treatment was with the discovery of the R144C polymorphism in the CYP2C9 gene (a variant now termed CYP2C9*2), which demonstrated approximately 10-fold lower hydroxylation

capacity of S-warfarin compared to the wild-type allele19. It was then quickly verified that this polymorphism was relevant in vivo, with the retrospective confirmation of significantly lower doses in subjects heterozygous for the R144C polymorphism (now termed *1/*2) compared to homozygous wildtype carriers (*1/*1)20. Another single-nucleotide polymorphism (SNP) in CYP2C9 (I359L, now known as CYP2C9*3) was also discovered, as well as several polymorphisms in the VKORC1 gene. Both retrospective and prospective studies have demonstrated the need for decreased warfarin dose in subjects carrying any of the CYP2C9 or VKORC1 variants; in addition, the independent effects of each variant lead to the requirement for even lower doses in subjects carrying more than one variant allele21,22. Our study also demonstrated that, during treatment initiation, the VKORC1 haplotype affected the time to the first INR in the therapeutic range, and VKORC1 and CYP2C9 polymorphisms affected the time to the first INR 44; this is arguably the most important timeframe for prevention of warfarin-related adverse effects23. We also reported a significant relationship between polymorphisms in both genes and the maintenance dose, as reported in previous studies21–24. Furthermore, more than one study has evaluated the contribution of genetic variables to interindividual variability in the warfarin dose compared to that using clinical variables alone. These studies reached similar conclusions: clinical variables alone accounted for 12–22% of the variability in warfarin dose, whereas the addition of genetic variables to clinical variables was able to account for 45–54% of interindividual variability in the dose10,22. In one of these evaluations, the VKORC1 promoter SNP -1639 G4A alone accounted for 25% of the variability10. Finally, our prospective evaluation of the Warfarin Regimen using A Pharmacogenetics-guided Initiation Dosing (WRAPID) algorithm, derived using both clinical and genetic variables, also showed that the use of this algorithm was able to eliminate the effects of CYP2C9*2, *3 and VKORC1 -1639 polymorphisms on the endpoints of time to therapeutic INR and time to INR44 that we had observed previously in patients dosed according to standard care24. Although the above studies leave little doubt regarding the role of genetic polymorphisms as the basis for variability in warfarin dose, what remains unclear is the value of obtaining genetic information for dosing during the initiation phase of


DOI: 10.3109/10408363.2014.950407

warfarin therapy. The US Food and Drug Administration approved the addition of genetic information to warfarin labeling in 2007; this currently states ‘‘CYP2C9 and VKORC1 genotype information, when available, can assist in selection of the initial dose of warfarin’’7. Guidelines for warfarin starting doses according to genotype were subsequently added, and are very similar to those of WRAPID. However, the routine use of genetic information for guiding initial warfarin dosing is not recommended by the American College of Chest Physicians (ACCP) because of the lack of large, randomized controlled trials (RCTs) comparing genetics-based dosing to standard care25, while the Clinical Pharmacogenetics Implementation Consortium (CPIC) strongly encourages the inclusion of genetic information, when available, by using published algorithms26. Interestingly, since the ACCP published its guidelines in 2012, results of two large RCTs have recently been made available, albeit with little clarification on the issue at hand. In a cohort of 455 patients, Pirmohamed et al.27 concluded that genotype-guided dosing was associated with a higher percentage of time in the therapeutic INR range compared to standard care (67.4 versus 60.3%) and with a lower incidence of excessive anticoagulation (INR 44). However, in a cohort of 1015 patients, Kimmel et al.28 found that the mean time spent in the therapeutic range was almost identical in the two groups (45.2 versus 45.4%), and concluded that genotype-based dosing did not improve anticoagulation control when compared to dosing based upon clinical variables alone. A clear difference between the two RCTs is the treatment of the control group – the latter study used clinical variables for dose adjustment, whereas the former used ‘‘standard care’’ (likely trial-and-error at most sites). Moreover, potential weaknesses in the study by Kimmel et al.9 were the large proportion of venous thromboembolism (VTE) patients who do not benefit as much from genotypebased dosing, and the inclusion of African-Americans for whom novel genetic predictors were not considered; both likely contributed to the negative findings. Given the results from these and previous studies, it can be concluded that warfarin dosing based upon algorithms that incorporate at least clinical variables related to warfarin interindividual variability should now be a part of ‘‘standard care’’; inclusion of genetic variables as well would result in equal if not better anticoagulation control. The most relevant question that remains is whether or not the addition of genetic information to warfarin dosing is cost-effective, a question that applies to personalized medicine for most therapeutic agents (discussed below). Statins (HMG Co-A reductase inhibitors) Due to their established utility in the prevention of major cardiovascular events in patients at risk, statins remain one of the most highly prescribed classes of drugs in the US, with simvastatin and atorvastatin accounting for almost 150 million dispensed prescriptions in 20112. In previous years, the choice of statin and statin dose for an individual patient was based largely on a ‘‘trial-and-error’’ technique, that is, titration to lipid response and presence of adverse effects, the most common being statin-induced myopathy. The goals of personalized medicine in statin therapy include addressing factors related to the incidence of myopathy during statin treatment as well as those related to the therapeutic effect.


5

Statin drugs elicit their lipid-lowering effect via inhibition of the HMG Co-A reductase enzyme that is present in the liver. This enzyme is responsible for the reduction of HMGCoA to mevalonic acid, the rate-limiting step for cholesterol biosynthesis29. Decreasing hepatic cholesterol production results in a compensatory increase in the expression of hepatic low-density lipoprotein (LDL) receptors, which leads to an increased removal by the liver of LDL from the blood, thereby decreasing circulating total and LDL cholesterol. There are currently seven HMG Co-A reductase inhibitors on the market in the US: atorvastatin, fluvastatin, lovastatin, pitavastatin, pravastatin, simvastatin and rosuvastatin. When administered orally, statin drugs reach their site of action in the liver prior to reaching the systemic circulation (Figure 1), and most of these drugs are highly extracted by the liver upon this first pass. Because of this first-pass extraction into the site of action, plasma concentrations of statins may not necessarily correlate with lipid-lowering effects – most of the drug may not even reach the systemic circulation. Conversely, statins must reach the systemic circulation in order to arrive at their primary site of toxicity, the muscle, and plasma concentrations of statins do correlate with the incidence of myopathy. The high extraction of statins by the liver is largely due to active uptake by transporter proteins, primarily OATP1B1 (OATP, organic anion transporting polypeptide), which is encoded by the gene SLCO1B1. Following entry into the liver, many statins are metabolized primarily by CYP enzymes: atorvastatin, lovastatin and simvastatin by CYP3A4, fluvastatin by CYP2C9 and to a lesser extent CYP3A4, and pitavastatin by CYP2CP. Fluvastatin, lovastatin and simvastatin are also partially metabolized by CYP2D6. Pravastatin is primarily eliminated unchanged through biliary and renal excretion. Rosuvastatin undergoes minor metabolism via CYP2C9, but it is largely excreted unchanged through biliary excretion. Known factors that affect interindividual variability in statin response are largely related to statin pharmacokinetics and the involvement of drug metabolizing enzymes such as CYP3A4, as well as drug efflux and uptake transporters. The evidence for interindividual variability due to differences in pharmacodynamic components is at this point inconclusive; however, interindividual differences in several targets relating to lipid metabolism have been evaluated30. Non-genetic factors related in interindividual variability As with warfarin, one of the clear clinical characteristics related to statin dose is patient age, with the elderly appearing particularly sensitive to myopathy31,32. This effect appears consistent among the different statins. Race also plays a major role in interindividual variability of some statins. When evaluated in Caucasians and Asians residing in the same environment, it was found that the pharmacokinetics of rosuvastatin were drastically different, with about a two-fold decrease in clearance in Asian subjects33. This difference, which remained even when known genetic differences were accounted for, suggested that the difference in rosuvastatin disposition between the two groups related to non-genetic or unknown genetic factors. Caution is warranted with dosing rosuvastatin at 420 mg/day in Asian subjects34. As all statins are substrates for CYP enzymes and/or drug transporters, all

6


are also subject to interindividual variability due to drug–drug interactions, which may be statin-specific depending on the proteins involved in their elimination (Table 1). Interactions can also occur with other cholesterol-lowering drugs, e.g. bile acid sequestrants that directly bind these drugs35,36.


Genetic factors related in interindividual variability There is one clear genetic component that causes interindividual differences in plasma concentrations of nearly all statins, namely polymorphisms in SLCO1B1, the gene encoding for the transporter, OATP1B1. OATP1B1 is expressed solely on the basolateral membrane of hepatocytes, and uptake by this transporter can be rate-limiting for the plasma clearance of its substrates, meaning that changes in OATP1B1 function can potentially affect plasma drug concentrations. In 2001, our group was the first to publish the identification of allelic variants in this gene, with the evaluation of 14 non-synonymous SNPs, including T521C (now referred to as SLCO1B1*5), with allele frequency of 15% in Caucasians. Since this SNP was identified, increased plasma exposure of atorvastatin, pitavastatin, pravastatin, simvastatin and rosuvastatin have been reported in variant carriers37. Its clinical relevance was further verified in a genome-wide association study, published in 2008, in which a single association was determined between the incidence of simvastatin-induced myopathy and the T521C SNP38; this confirmed that the increase in plasma concentrations elicited by decreased OATP1B1 transport translated to an increased risk of statin-related adverse effects. Interestingly, although all statins are OATP1B1 substrates, the magnitude of the effect of the T521C SNP on plasma exposure is statindependent, with the greatest effect being observed in subjects taking simvastatin, and with no effect in subjects taking fluvastatin39. This substrate dependence is likely due to compensatory hepatic uptake of some statins by other transporters such as OATP1B3 and/or the bile acid transporter, NTCP. Other polymorphisms in SLCO1B1 with relatively high allele frequencies include SLCO1B1*1b (A388G) and SLCO1B1*15 (A388G + T521C). Interestingly, the A388G polymorphism appears to have increased in vivo activity compared to wild-type, and has been demonstrated to result in decreased plasma concentrations of pravastatin and atorvastatin40–42. In vivo hepatic expression of OATP has also been demonstrated to be higher in carriers of A388G43. The effect of this polymorphism seems to be substrate-dependent, however, as one study demonstrated increased plasma concentrations of rosuvastatin33. Although the SLCO1B1*15 allele carries SNPs which appear to conflict in their effects on transport activity, the overall effect of this particular allele has been reported to be decreased in vivo activity, which causes similar increases in plasma concentrations of pravastatin, pitavastatin and rosuvastatin as seen among carriers of SLCO1B1*544. Many statins are also substrates for other transporters more ubiquitously expressed than OATP1B1. Rosuvastatin, which is mainly eliminated by biliary excretion, is transported by breast cancer resistance protein (BCRP). BCRP is encoded for by the ABCG2 gene, which is expressed at the canalicular side of hepatocytes (responsible for excretion into bile), and in the gut where it can prevent drug absorption.


Interestingly, studies evaluating SNPs in ABCG2 have demonstrated a relationship between genotype and both plasma exposure and lipid-lowering effects45–49. It is likely that decreased transport by BCRP leads to higher plasma and liver concentrations of rosuvastatin, thereby causing increased effects because of higher concentrations at the site of action. These effects are likely due to decreased transport in both the gut and liver, which results in increased bioavailability and decreased biliary excretion. Not surprisingly, polymorphisms in ABCG2 have also been linked to higher fluvastatin and atorvastatin exposure, both of which are also BCRP substrates45,50. Interestingly, however, our study and others demonstrated plasma concentrations of pravastatin to be unchanged in ABCG2 variant carriers, even though it is also transported by BCRP50,51. This lack of effect is again likely due to compensatory transport by other transporters, in this case, likely the multidrug resistanceassociated protein 2 (encoded by ABCC2 gene), for which pravastatin is also a substrate52. Transporter polymorphisms account for varying but clinically significant degrees of statin-associated adverse reactions that are related to genotype-dependent variation in statin uptake and elimination. For example, in our analysis of factors related to statin plasma concentrations, 38% of explainable atorvastatin variability could be accounted for by SLCO1B1 c.521T4C and SLCO1B1 c.388A4G, with positive and negative effects for each allele, respectively. However, for rosuvastatin, 88% of explainable variability could be accounted for by SLCO1B1 c.521T4C and ABCG2 c.421C4A, both of which had positive effects42. As some statins are eliminated primarily through metabolism by CYP450 enzymes, many analyses have also evaluated the effects of common polymorphisms in these genes on the efficacy and toxicity of statins, although conflicting results between pharmacokinetic and pharmacodynamic studies have raised questions about the clinical relevance of such SNPs53,54. In our analysis, 30% of the explainable variability in atorvastatin concentration was explained by a marker of CYP3A4 activity, 4-hydroxycholesterol42. This suggested that interindividual variability in metabolism may be a factor even for more hydrophobic statins like atorvastatin, and that other genetic (or non-genetic) factors affecting this variability need to be considered. Interestingly, in a recent study, a polymorphism in GATM, which encodes glycine aminidotransferase, an enzyme involved in the synthesis of creatine, and which is unrelated to statin metabolism, was significantly related to the incidence of statin-induced myopathy55. GATM was also upregulated following simvastatin exposure in vitro; thus, this gene is not only a potential novel genetic marker for statin myopathy, but also may serve as a mechanistic link between statin exposure and toxicity. Clopidogrel Non-genetic factors related in interindividual variability Clopidogrel is a thienopyridine commonly prescribed for the treatment and secondary prevention of acute coronary syndromes and for treatment of peripheral artery disease. Clopidogrel, itself a prodrug, forms an active thiol metabolite via a two-step process involving multiple CYP enzymes,


DOI: 10.3109/10408363.2014.950407

including CYP1A2, 2B6, 2C9, 2C19 and 3A456. It is the thiol metabolite which irreversibly binds to the P2Y12 receptor (ADP receptor expressed on the surface of platelets) blocking ADP-activated platelet aggregation for the life of the platelet5. CYP2C19 is involved in both steps of clopidogrel bioactivation, making drug–drug interactions with CYP2C19 a significant concern. Many proton pump inhibitors, particularly omeprazole, inhibit CYP2C19 and give rise to a potential interaction of relevance to clopidogrel’s antiplatelet effect. While short-term studies agree on omeprazole’s ability to mediate clopidogrel’s antiplatelet effect57,58, long-term studies regarding the result of the interaction on adverse cardiac events are conflicting59,60. It is still inconclusive if CYP2C19 drug interactions represent a clinically significant variable; however, it is suggested that concerned clinicians should consider other proton pump inhibitors, such as pantoprazole or lansoprazole, for subjects who require proton pump inhibitors during clopidogrel therapy61. Genetic factors related in interindividual variability Of the CYPs mediating the activation of clopidogrel to the active thiol metabolite, there exists a common loss-of-function polymorphism in CYP2C19, referred to as CYP2C19*2, with an allele frequency of 13% in Caucasians62. Other lossof-function alleles also exist for CYP2C19 (*3, *4, *5, *6, *7 and *8), with *3 only having a frequency of 41% in Asian populations63. A gain-of-function allele (*17) also exists, with allele frequencies of 4–18% depending on ethnicity64. Short-term studies have demonstrated that carriers of lossof-function variant alleles for CYP2C19 have lower concentrations of the active thiol metabolite and decreased antiplatelet effect following acute clopidogrel administration65–67. However, while the effects of the CYP2C19 genotype on antiplatelet effect may be significant, they do not appear to account for the majority of interindividual variability to antiplatelet response during therapy68. This may be one reason why the results of clinical studies evaluating the role of the CYP2C19 genotype on cardiovascular outcomes among subjects on clopidogrel therapy have been inconsistent. Recent meta-analyses do suggest that CYP2C19 variant carriers are at increased risk of cardiac events69,70, although it has also been suggested that the effect of the CYP2C19 genotype on clopidogrel efficacy and cardiovascular risk is indication-specific; the best evidence for genotype-based decision making exists for patients undergoing percutaneous coronary intervention following acute coronary syndrome71. Current CPIC guidelines indicate, when clopidogrel is used for this indication, that genetic information should be used, when available, in determining the choice of antiplatelet medication, and that in heterozygous or homozygous carriers of variant CYP2C19 alleles, alternative antiplatelet agents should be employed72. The fact that the consortium notes that currently known genetic and non-genetic factors do not appear to fully account for the mechanistic basis of the extent of interindividual variability in response to clopidogrel suggests that additional research is needed. Tamoxifen Although tamoxifen remains a primary anti-estrogen therapy for treatment and secondary prevention of estrogen-dependent


7

breast cancer, significant interindividual variability exists in its efficacy. Like clopidogrel, tamoxifen itself is a prodrug that requires bioactivation to active moieties. Though tamoxifen metabolism is complex and involves multiple CYP enzymes, two active metabolites, 4-hydroxy tamoxifen and endoxifen, with high affinity for estrogen receptors, approximately 100-fold higher affinity than tamoxifen itself, are formed73. Of these two metabolites, endoxifen demonstrates much higher plasma exposure, which suggests that the effects of administered tamoxifen are due mostly to this moiety74. Endoxifen is formed via a two-step process, with the first mediated primarily by CYP3A4 and the second mediated primarily via CYP2D6 metabolism74. Non-genetic factors related in interindividual variability Because of the primary role of CYP2D6 in tamoxifen bioactivation, one of the key concerns for a lack of response to tamoxifen therapy is drug–drug interactions involving CYP2D6 inhibitors. One of the primary drug classes of concern is antidepressants because these agents inhibit CYP2D6 and are often prescribed in patients diagnosed with breast cancer. Considering the suggested role of endoxifen in the therapeutic effect of administered tamoxifen, plasma concentrations of this agent are likely highly relevant to the therapeutic effect. Numerous studies have shown that co-administration of antidepressants such as paroxetine, which inhibits CYP2D6, decreases plasma levels of endoxifen in patients on tamoxifen therapy74–77. One retrospective analysis demonstrated that the use of paroxetine was also associated with an increased risk of death in patients on tamoxifen, with the risk increasing with increase in duration of drug overlap78; however, other studies have failed to demonstrate a significant difference in outcome with concomitant use of CYP2D6 inhibitors79. Fortunately, not all selective serotonin reuptake inhibitor (SSRI) antidepressants have a similar degree of CYP2D6 interaction potential; thus, it is feasible to prescribe SSRIs which are less likely to interact with CYP2D6. Nonetheless, our recent study suggests that all SSRIs and the serotonin–norepinephrine reuptake inhibitor (SNRI), venlafaxine, do interact with tamoxifen metabolism, but not to the extent seen with paroxetine, fluoxetine or buproprion75. In this study, which evaluated both non-genetic and genetic factors related to interindividual variability in endoxifen plasma concentrations, along with CYP2D6 inhibition, interestingly we also observed a seasonal effect on endoxifen plasma concentrations and higher endoxifen plasma concentrations in subjects on vitamin D supplementation, which suggested that vitamin D status may be another non-genetic factor that requires consideration during tamoxifen treatment75. Genetic factors related in interindividual variability CYP2D6, which is highly polymorphic, has four phenotypes for in vivo CYP2D6 activity: poor metabolizers, intermediate metabolizers, extensive metabolizers and ultrarapid metabolizers80. While studies concur on the effect of CYP2D6 pharmacogenomics on endoxifen plasma concentration76,77,81, studies evaluating tamoxifen pharmacodynamics have conflicting results, which have led to


8


controversy on the role of the CYP2D6 genotype to diseaserelated outcomes82–84. Such inconsistencies may have resulted from a lack of appreciation of non-genetic factors noted in the previous sections as well as heterogeneity in the group of study subjects, doses of tamoxifen, variation in other chemotherapeutic agents, and pre- and post-menopausal status85,86. Incorrect genotyping is another major issue in the interpretation of such retrospective studies. Technical issues relating to genotyping may have occurred through the use of different sources of DNA (tumor versus host), and paraffin-embedded tissue for DNA extraction. In addition, insufficient numbers of variant alleles in CYP2D6 were assessed. The International Tamoxifen Pharmacogenomics Consortium addressed these issues of inconsistency in a recent meta-analysis in which they established three different sets of inclusion criteria for increasing the homogeneity of subjects enrolled in various tamoxifen studies85. They noted that in a cohort consisting of post-menopausal women with estrogen receptor positive breast cancer, who received 20 mg daily dose of tamoxifen for 5 years, the CYP2D6 genotype was significantly related to disease-free survival. It should be noted that, in addition to CYP2D6, there may be a role for other pharmacogenetic markers for tamoxifen pharmacokinetics and pharmacodynamics, including SNPs in drug transporters. In our recent study, we found that the ratio of endoxifen to tamoxifen was significantly correlated to the CYP2D6 genotype, in concurrence with previous studies. The CYP2D6 phenotype accounted for a moderate amount of variability (30%) in our study; however, differences in endoxifen plasma concentration could not completely be explained by the CYP2D6 genotype alone75. We also observed a significant relationship between the CYP3A4 genotype and both tamoxifen and endoxifen concentrations, with surprisingly higher levels of endoxifen in carriers of the variant CYP3A4*22 allele. This new finding, as well as the role of the CYP3A4 genotype in disease-related outcomes, needs to be confirmed. Other studies have focused on the enzymes, specifically sulfotransferase 1A1 (SULT1A1), that are responsible for the metabolism of tamoxifen’s active metabolites as opposed to their formation. SULT1A1 is involved in the degradation of endoxifen and 4-hydroxytamoxifen; however, no association between the SULT1A1 genotype and the tamoxifen effect or metabolite concentration has been found84, which suggests that formation and not elimination is likely the rate-limiting step that controls plasma concentrations of the active moieties. Our analysis, and others, also evaluated the role of drug transporter SNPs, as tamoxifen, endoxifen and 4-hydroxytamoxifen have been shown to be substrates for efflux transporters87–89. In our analysis, neither SNPs in MDR1 (encoding P-glycoprotein) nor ABCG2 (encoding BCRP) were significantly related to endoxifen concentrations. The lack of relationship with these transporter SNPs to outcomes was also demonstrated in another retrospective analysis90. Interestingly, however, in the latter study, the investigators did find a significant relationship between clinical outcomes and multiple SNPs in ABCC2 (encoding MRP2), with homozygous carriers of one SNP resulting in a hazard ratio of greater than 10 compared to wild-type. Although differences in MRP2 transport could plausibly alter the bioavailability or


clearance of tamoxifen and/or endoxifen due to its expression in the intestine, liver and kidney, interestingly no differences in trough plasma concentrations of tamoxifen or endoxifen were observed in this study. The investigators conclude that this SNP likely increases activity of MRP2 in the tumor itself, resulting in less active drug at the site of action.

Conclusions The issue of interindividual variability in disease treatment cannot be ignored. While there have been recent successes with respect to specific therapeutic agents, issues remain that will need to be addressed in the future. There has been much ‘‘hype’’ around the promise of a ‘‘genomics revolution’’ since the sequencing of the human genome in 2003 (http://www.genome.gov/10001772); however, it is clear that success in implementation of personalized medicine in the real-world patient care setting will require a multidisciplinary and long-term approach focused on measurable outcomes of relevance to drug safety, effectiveness and cost-benefit.

Future perspective Challenges to the implementation of personalized medicine The cost of health care continues to rise at a pace that is likely not sustainable for most developed countries. The emergence of genomics-guided treatment options that individualize therapy has enormous potential not only to offer safe and effective treatment options, but also to lower overall health care costs through the prevention of adverse drug reactions that often result in repeated clinic visits or hospitalizations. Although the cost of gene sequencing or genotyping had been widely touted as the hurdle that would prevent broader implementation of personalized medicine-based patient care, advances in genotyping technologies have made such arguments obsolete. Indeed, we are already entering the era of so-called direct-to-consumer marketing of personalized medicine. There are direct-to-consumer genetic testing services offered over the Internet, which provide a full panel of genetic markers for a low cost (e.g. $99.00 USD). However, in November 2013, US Food and Drug Administration asked a direct-to-consumer genetic testing company called 23andMe (www.23andme.com) to stop selling its personal genetics testing kits to consumers because of significant concerns regarding the clinical basis, relevance and accuracy of the genotyping results91. In addition to cost and accuracy of personalized genetic tests, another barrier to broader adoption of personalized medicine is the need for integrating not just genomic information, but also the patient’s clinical condition and severity of symptoms as well as other confounding variables such as diet (e.g. grapefruit juice effect) and drug–drug interactions in an expert and systematic fashion. Therefore, the acceptance of personalized medicine, at least for commonly prescribed medications, will take a concerted effort to demonstrate not only efficacy and safety but also cost-effectiveness. Moreover, greater effort will be needed to train the next generation of physicians, pharmacists and other health care professionals so that they become familiar with advances in genomics research and clinical

DOI: 10.3109/10408363.2014.950407

relevance. In addition, the creation of innovative technologies that can provide automated drug dosing and selection support and provider order entry systems as part of clinic or hospital electronic medical records will be needed. Moreover, great progress is being made in the field of ‘‘omics’’ technologies, where mRNA, protein expression and metabolome profiling along with genomic information are being integrated to create a more comprehensive strategy for personalized medicine in the next decade92.


Declaration of interest The authors state no declarations of interest. R.B.K. is supported by the Wolfe Medical Research Chair in Pharmacogenomics and by grants from the Canadian Institutes of Health Research (MOP-89753) and the Drug Safety and Effectiveness Network (DSEN-PREVENT, FRN-117588), Academic Medical Organization of Southwestern Ontario Alternate Funding Plan Innovation Fund, the Cancer Care Ontario (CCO) Research Chair Award (Tier-1) in Experimental Therapeutics, and the Ontario Institute for Cancer Research (OICR) Translational Research Team grant.

References 1. Carleton BC, Smith MA. Drug safety: side effects and mistakes or adverse reactions and deadly errors? BC Medical J 2006;48: 329–33. 2. IMS Institute for Healthcare Informatics. The use of medicines in the United States: Review of 2011; 2012. Available from: http://www.imshealth.com/ims/Global/Content/Insights/IMS%20 Institute%20for%20Healthcare%20Informatics/IHII_Medicines_ in_U.S_Report_2011.pdf [last accessed 14 Jan 2014]. 3. Gong IY, Kim RB. Importance of pharmacokinetic profile and variability as determinants of dose and response to dabigatran, rivaroxaban, and apixaban. Can J Cardiol 2013;29:S24–33. 4. O’Reilly RA, Aggeler PM, Leong LS. Studies on the coumarin anticoagulant drugs: the pharmacodynamics of warfarin in man. J Clin Invest 1963;42:1542–51. 5. Lu MC. Antithrombotics, thrombolytics, antiplatelets, and coagulants. In: Lemke TL, Williams DA, eds. Foye’s Principles of Medicinal Chemistry. 7th ed. Baltimore: Lippincott Williams & Wilkins, 2013:841–76. 6. O’Reilly RA. Interaction of several coumarin compounds with human and canine plasma albumin. Mol Pharmacol 1971;7: 209–18. 7. Bristol Myers Squibb Company. COUMADIN prescribing information. Available from: http://www.accessdata.fda.gov/drugsatfda_docs/label/2011/009218s107lbl.pdf. 2011 [last accessed 14 Jan 2014]. 8. Routledge PA, Chapman PH, Davies DM, Rawlins MD. Pharmacokinetics and pharmacodynamics of warfarin at steady state. Br J Clin Pharmacol 1979;8:243–7. 9. Gong IY, Schwarz UI, Crown N, et al. Clinical and genetic determinants of warfarin pharmacokinetics and pharmacodynamics during treatment initiation. PLoS One 2011;6:e27808. 10. Gage BF, Eby C, Johnson JA, et al. Use of pharmacogenetic and clinical factors to predict the therapeutic dose of warfarin. Clin Pharmacol Ther 2008;84:326–31. 11. Lu Y, Yang J, Zhang H, Yang J. Prediction of warfarin maintenance dose in Han Chinese patients using a mechanistic model based on genetic and non-genetic factors. Clin Pharmacokinet 2013;52:567–81. 12. Perera MA, Cavallari LH, Limdi NA, et al. Genetic variants associated with warfarin dose in African–American individuals: a genome-wide association study. Lancet 2013;382:790–6. 13. Hernandez W, Gamazon ER, Aquino-Michaels K, et al. Ethnicityspecific pharmacogenetics: the case of warfarin in African Americans. Pharmacogenomics J 2014;14:223–8.


9

14. Naud J, Nolin TD, Leblond FA, Pichette V. Current understanding of drug disposition in kidney disease. J Clin Pharmacol 2012;52: 10S–22S. 15. Dreisbach AW, Lertora JJ. The effect of chronic renal failure on hepatic drug metabolism and drug disposition. Semin Dial 2003; 16:45–50. 16. Elliott MJ, Zimmerman D, Holden RM. Warfarin anticoagulation in hemodialysis patients: a systematic review of bleeding rates. Am J Kidney Dis 2007;50:433–40. 17. Kleinow ME, Garwood CL, Clemente JL, Whittaker P. Effect of chronic kidney disease on warfarin management in a pharmacistmanaged anticoagulation clinic. J Manag Care Pharm 2011;17: 523–30. 18. Limdi NA, Limdi MA, Cavallari L, et al. Warfarin dosing in patients with impaired kidney function. Am J Kidney Dis 2010;56: 823–31. 19. Rettie AE, Wienkers LC, Gonzalez FJ, et al. Impaired (S)-warfarin metabolism catalysed by the R144C allelic variant of CYP2C9. Pharmacogenetics 1994;4:39–42. 20. Furuya H, Fernandez-Salguero P, Gregory W, et al. Genetic polymorphism of CYP2C9 and its effect on warfarin maintenance dose requirement in patients undergoing anticoagulation therapy. Pharmacogenetics 1995;5:389–92. 21. D’Andrea G, D’Ambrosio RL, Di Perna P, et al. A polymorphism in the VKORC1 gene is associated with an interindividual variability in the dose-anticoagulant effect of warfarin. Blood 2005;105:645–9. 22. Carlquist JF, Horne BD, Muhlestein JB, et al. Genotypes of the cytochrome p450 isoform, CYP2C9, and the vitamin K epoxide reductase complex subunit 1 conjointly determine stable warfarin dose: a prospective study. J Thromb Thrombolysis 2006;22:191–7. 23. Schwarz UI, Ritchie MD, Bradford Y, et al. Genetic determinants of response to warfarin during initial anticoagulation. N Engl J Med 2008;358:999–1008. 24. Gong IY, Tirona RG, Schwarz UI, et al. Prospective evaluation of a pharmacogenetics-guided warfarin loading and maintenance dose regimen for initiation of therapy. Blood 2011;118:3163–71. 25. Holbrook A, Schulman S, Witt DM, et al. Evidence-based management of anticoagulant therapy: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:e152S–84S. 26. Johnson JA, Gong L, Whirl-Carrillo M, et al. Clinical Pharmacogenetics Implementation Consortium Guidelines for CYP2C9 and VKORC1 genotypes and warfarin dosing. Clin Pharmacol Ther 2011;90:625–9. 27. Pirmohamed M, Burnside G, Eriksson N, et al. A randomized trial of genotype-guided dosing of warfarin. N Engl J Med 2013;369: 2294–303. 28. Kimmel SE, French B, Kasner SE, et al. A pharmacogenetic versus a clinical algorithm for warfarin dosing. N Engl J Med 2013;369:2283–93. 29. Harrold M. Antihyperlipoproteinemics and inhibitors of cholesterol biosynthesis. In: Lemke TL, Williams DA, eds. Foye’s Principles of Medicinal Chemistry. Baltimore: Lippincott Williams & Wilkins, 2013:815–40. 30. Patel J, Abd T, Blumenthal RS, et al. Genetics and personalized medicine – a role in statin therapy? Curr Atheroscler Rep 2014;16: 384. 31. Joy TR, Hegele RA. Narrative review: statin-related myopathy. Ann Intern Med 2009;150:858–68. 32. Pasternak RC, Smith Jr SC, Bairey-Merz CN, et al. ACC/AHA/ NHLBI Clinical Advisory on the Use and Safety of Statins. Stroke 2002;33:2337–41. 33. Lee E, Ryan S, Birmingham B, et al. Rosuvastatin pharmacokinetics and pharmacogenetics in white and Asian subjects residing in the same environment. Clin Pharmacol Ther 2005;78:330–41. 34. AstraZeneca. CRESTOR Prescribing Information. Available from: http://www.accessdata.fda.gov/drugsatfda_docs/label/2013/ 021366s028s029lbl.pdf. 2013 [last accessed 14 Jan 2014]. 35. Hatanaka T. Clinical pharmacokinetics of pravastatin: mechanisms of pharmacokinetic events. Clin Pharmacokinet 2000;39: 397–412. 36. Scripture CD, Pieper JA. Clinical pharmacokinetics of fluvastatin. Clin Pharmacokinet 2001;40:263–81.


10


37. Kalliokoski A, Niemi M. Impact of OATP transporters on pharmacokinetics. Br J Pharmacol 2009;158:693–705. 38. Group SC, Link E, Parish S, et al. SLCO1B1 variants and statininduced myopathy – a genomewide study. N Engl J Med 2008; 359:789–99. 39. Niemi M, Pasanen MK, Neuvonen PJ. SLCO1B1 polymorphism and sex affect the pharmacokinetics of pravastatin but not fluvastatin. Clin Pharmacol Ther 2006;80:356–66. 40. Maeda K, Ieiri I, Yasuda K, et al. Effects of organic anion transporting polypeptide 1B1 haplotype on pharmacokinetics of pravastatin, valsartan, and temocapril. Clin Pharmacol Ther 2006; 79:427–39. 41. Mwinyi J, Johne A, Bauer S, et al. Evidence for inverse effects of OATP-C (SLC21A6) 5 and 1b haplotypes on pravastatin kinetics. Clin Pharmacol Ther 2004;75:415–21. 42. DeGorter MK, Tirona RG, Schwarz UI, et al. Clinical and pharmacogenetic predictors of circulating atorvastatin and rosuvastatin concentrations in routine clinical care. Circ Cardiovasc Genet 2013;6:400–8. 43. Nies AT, Niemi M, Burk O, et al. Genetics is a major determinant of expression of the human hepatic uptake transporter OATP1B1, but not of OATP1B3 and OATP2B1. Genome Med 2013;5:1. doi: 10.1186/gm405. 44. Gong IY, Kim RB. Impact of genetic variation in OATP transporters to drug disposition and response. Drug Metab Pharmacokinet 2013;28:4–18. 45. Keskitalo JE, Zolk O, Fromm MF, et al. ABCG2 polymorphism markedly affects the pharmacokinetics of atorvastatin and rosuvastatin. Clin Pharmacol Ther 2009;86:197–203. 46. Chasman DI, Giulianini F, MacFadyen J, et al. Genetic determinants of statin-induced low-density lipoprotein cholesterol reduction: the Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER) trial. Circ Cardiovasc Genet 2012;5:257–64. 47. Lee HK, Hu M, Lui S, et al. Effects of polymorphisms in ABCG2, SLCO1B1, SLC10A1 and CYP2C9/19 on plasma concentrations of rosuvastatin and lipid response in Chinese patients. Pharmacogenomics 2013;14:1283–94. 48. Tomlinson B, Hu M, Lee VW, et al. ABCG2 polymorphism is associated with the low-density lipoprotein cholesterol response to rosuvastatin. Clin Pharmacol Ther 2010;87:558–62. 49. Bailey KM, Romaine SP, Jackson BM, et al. Hepatic metabolism and transporter gene variants enhance response to rosuvastatin in patients with acute myocardial infarction: the GEOSTAT-1 Study. Circ Cardiovasc Genet 2010;3:276–85. 50. Keskitalo JE, Pasanen MK, Neuvonen PJ, Niemi M. Different effects of the ABCG2 c.421C4A SNP on the pharmacokinetics of fluvastatin, pravastatin and simvastatin. Pharmacogenomics 2009; 10:1617–24. 51. Ho RH, Choi L, Lee W, et al. Effect of drug transporter genotypes on pravastatin disposition in European- and African-American participants. Pharmacogenet Genomics 2007;17:647–56. 52. Matsushima S, Maeda K, Kondo C, et al. Identification of the hepatic efflux transporters of organic anions using doubletransfected Madin-Darby canine kidney II cells expressing human organic anion-transporting polypeptide 1B1 (OATP1B1)/ multidrug resistance-associated protein 2, OATP1B1/multidrug resistance 1, and OATP1B1/breast cancer resistance protein. J Pharmacol Exp Ther 2005;314:1059–67. 53. Gelissen IC, McLachlan AJ. The pharmacogenomics of statins. Pharmacol Res 2013. pii: S1043-6618(13)00269-7. doi: 10.1016/ j.phrs.2013.12.002. [Epub ahead of print]. 54. Lamba J, Hebert JM, Schuetz EG, et al. PharmGKB summary: very important pharmacogene information for CYP3A5. Pharmacogenet Genomics 2012;22:555–8. 55. Mangravite LM, Engelhardt BE, Medina MW, et al. A statindependent QTL for GATM expression is associated with statininduced myopathy. Nature 2013;502:377–80. 56. Fareed J, Jeske W, Thethi I. Metabolic differences of current thienopyridine antiplatelet agents. Expert Opin Drug Metab Toxicol 2013;9:307–17. 57. Sibbing D, Morath T, Stegherr J, et al. Impact of proton pump inhibitors on the antiplatelet effects of clopidogrel. Thromb Haemost 2009;101:714–19.


58. Gilard M, Arnaud B, Cornily JC, et al. Influence of omeprazole on the antiplatelet action of clopidogrel associated with aspirin: the randomized, double-blind OCLA (Omeprazole CLopidogrel Aspirin) study. J Am Coll Cardiol 2008;51:256–60. 59. Rassen JA, Choudhry NK, Avorn J, Schneeweiss S. Cardiovascular outcomes and mortality in patients using clopidogrel with proton pump inhibitors after percutaneous coronary intervention or acute coronary syndrome. Circulation 2009;120: 2322–9. 60. Kreutz RP, Stanek EJ, Aubert R, et al. Impact of proton pump inhibitors on the effectiveness of clopidogrel after coronary stent placement: the clopidogrel Medco outcomes study. Pharmacotherapy 2010;30:787–96. 61. Bates ER, Lau WC, Angiolillo DJ. Clopidogrel-drug interactions. J Am Coll Cardiol 2011;57:1251–63. 62. Goldstein JA, Ishizaki T, Chiba K, et al. Frequencies of the defective CYP2C19 alleles responsible for the mephenytoin poor metabolizer phenotype in various Oriental, Caucasian, Saudi Arabian and American black populations. Pharmacogenetics 1997;7:59–64. 63. Desta Z, Zhao X, Shin JG, Flockhart DA. Clinical significance of the cytochrome P450 2C19 genetic polymorphism. Clin Pharmacokinet 2002;41:913–58. 64. Sim SC, Risinger C, Dahl ML, et al. A common novel CYP2C19 gene variant causes ultrarapid drug metabolism relevant for the drug response to proton pump inhibitors and antidepressants. Clin Pharmacol Ther 2006;79:103–13. 65. Kim KA, Park PW, Hong SJ, Park JY. The effect of CYP2C19 polymorphism on the pharmacokinetics and pharmacodynamics of clopidogrel: a possible mechanism for clopidogrel resistance. Clin Pharmacol Ther 2008;84:236–42. 66. Umemura K, Furuta T, Kondo K. The common gene variants of CYP2C19 affect pharmacokinetics and pharmacodynamics in an active metabolite of clopidogrel in healthy subjects. J Thromb Haemost 2008;6:1439–41. 67. Gong IY, Crown N, Suen CM, et al. Clarifying the importance of CYP2C19 and PON1 in the mechanism of clopidogrel bioactivation and in vivo antiplatelet response. Eur Heart J 2012;33: 2856–64. 68. Shuldiner AR, O’Connell JR, Bliden KP, et al. Association of cytochrome P450 2C19 genotype with the antiplatelet effect and clinical efficacy of clopidogrel therapy. JAMA 2009;302:849–57. 69. Jang JS, Cho KI, Jin HY, et al. Meta-analysis of cytochrome P450 2C19 polymorphism and risk of adverse clinical outcomes among coronary artery disease patients of different ethnic groups treated with clopidogrel. Am J Cardiol 2012;110:502–8. 70. Mao L, Jian C, Changzhi L, et al. Cytochrome CYP2C19 polymorphism and risk of adverse clinical events in clopidogreltreated patients: a meta-analysis based on 23 035 subjects. Arch Cardiovasc Dis 2013;106:517–27. 71. Johnson JA, Roden DM, Lesko LJ, et al. Clopidogrel: a case for indication-specific pharmacogenetics. Clin Pharmacol Ther 2012; 91:774–6. 72. Scott SA, Sangkuhl K, Stein CM, et al. Clinical Pharmacogenetics Implementation Consortium guidelines for CYP2C19 genotype and clopidogrel therapy: 2013 update. Clin Pharmacol Ther 2013; 94:317–23. 73. Johnson MD, Zuo H, Lee KH, et al. Pharmacological characterization of 4-hydroxy-N-desmethyl tamoxifen, a novel active metabolite of tamoxifen. Breast Cancer Res Treat 2004; 85:151–9. 74. Stearns V, Johnson MD, Rae JM, et al. Active tamoxifen metabolite plasma concentrations after coadministration of tamoxifen and the selective serotonin reuptake inhibitor paroxetine. J Natl Cancer Inst 2003;95:1758–64. 75. Teft WA, Gong IY, Dingle B, et al. CYP3A4 and seasonal variation in vitamin D status in addition to CYP2D6 contribute to therapeutic endoxifen level during tamoxifen therapy. Breast Cancer Res Treat 2013;139:95–105. 76. Jin Y, Desta Z, Stearns V, et al. CYP2D6 genotype, antidepressant use, and tamoxifen metabolism during adjuvant breast cancer treatment. J Natl Cancer Inst 2005;97:30–9. 77. Borges S, Desta Z, Li L, et al. Quantitative effect of CYP2D6 genotype and inhibitors on tamoxifen metabolism: implication for

DOI: 10.3109/10408363.2014.950407

78. 79.

80.


81.

82. 83.

84.

optimization of breast cancer treatment. Clin Pharmacol Ther 2006;80:61–74. Kelly CM, Juurlink DN, Gomes T, et al. Selective serotonin reuptake inhibitors and breast cancer mortality in women receiving tamoxifen: a population based cohort study. BMJ 2010;340:c693. Dezentje VO, van Blijderveen NJ, Gelderblom H, et al. Effect of concomitant CYP2D6 inhibitor use and tamoxifen adherence on breast cancer recurrence in early-stage breast cancer. J Clin Oncol 2010;28:2423–9. Sachse C, Brockmoller J, Bauer S, Roots I. Cytochrome P450 2D6 variants in a Caucasian population: allele frequencies and phenotypic consequences. Am J Hum Genet 1997;60:284–95. Murdter TE, Schroth W, Bacchus-Gerybadze L, et al. Activity levels of tamoxifen metabolites at the estrogen receptor and the impact of genetic polymorphisms of phase I and II enzymes on their concentration levels in plasma. Clin Pharmacol Ther 2011; 89:708–17. Goetz MP, Suman VJ, Hoskin TL, et al. CYP2D6 metabolism and patient outcome in the Austrian Breast and Colorectal Cancer Study Group trial (ABCSG) 8. Clin Cancer Res 2013;19:500–7. Regan MM, Leyland-Jones B, Bouzyk M, et al. CYP2D6 genotype and tamoxifen response in postmenopausal women with endocrine-responsive breast cancer: the breast international group 1–98 trial. J Natl Cancer Inst 2012;104:441–51. Kiyotani K, Mushiroda T, Nakamura Y, Zembutsu H. Pharmacogenomics of tamoxifen: roles of drug metabolizing enzymes and transporters. Drug Metab Pharmacokinet 2012;27: 122–31.


11

85. Province MA, Goetz MP, Brauch H, et al. CYP2D6 genotype and adjuvant tamoxifen: meta-analysis of heterogeneous study populations. Clin Pharmacol Ther 2014;95:216–27. 86. Kiyotani K, Mushiroda T, Zembutsu H, Nakamura Y. Important and critical scientific aspects in pharmacogenomics analysis: lessons from controversial results of tamoxifen and CYP2D6 studies. J Hum Genet 2013;58:327–33. 87. Teft WA, Mansell SE, Kim RB. Endoxifen, the active metabolite of tamoxifen, is a substrate of the efflux transporter P-glycoprotein (multidrug resistance 1). Drug Metab Dispos 2011;39:558–62. 88. Iusuf D, Teunissen SF, Wagenaar E, et al. P-glycoprotein (ABCB1) transports the primary active tamoxifen metabolites endoxifen and 4-hydroxytamoxifen and restricts their brain penetration. J Pharmacol Exp Ther 2011;337:710–17. 89. Callaghan R, Higgins CF. Interaction of tamoxifen with the multidrug resistance P-glycoprotein. Br J Cancer 1995;71:294–9. 90. Kiyotani K, Mushiroda T, Imamura CK, et al. Significant effect of polymorphisms in CYP2D6 and ABCC2 on clinical outcomes of adjuvant tamoxifen therapy for breast cancer patients. J Clin Oncol 2010;28:1287–93. 91. Annas GJ, Elias S. 23andMe and the FDA. N Engl J Med 2014; 370:985–8. 92. Chen R, Mias GI, Li-Pook-Than J, et al. Personal omics profiling reveals dynamic molecular and medical phenotypes. Cell 2012; 148:1293–307. 93. Jusko WJ, Ko HC, Ebling WF. Convergence of direct and indirect pharmacodynamic response models. J Pharmacokinet Biopharm 1995;23:5–8; discussion 9–10.

Personalized pain medicine: pipe dream or reality?

Noninvasive dentistry: a dream or reality?

Chemoprotection in glioblastoma therapy: reality or a dream?

Differential diagnosis by metabolic profile: a dream or reality?

Clostridium to treat cancer: dream or reality?

Adult Stem Cell Plasticity: Dream or Reality?

Integrated care in France: dream or reality?

PCORnet: turning a dream into reality.

Tuberculosis elimination: dream or reality? The case of Cyprus.

Universal health coverage in Rwanda: dream or reality.

Laparoscopic resection of pancreatic adenocarcinoma: dream or reality?

Personalized medicine in myelodysplastic syndromes: wishful thinking or already clinical reality?

Personalized treatments of cancer patients: a reality in daily practice, a costly dream or a shared vision of the future from the oncology community?

Akt signaling pathway in gastric carcinoma: A reality for personalized medicine?

Personalized medicine for breast cancer: it is a new day!

Personalized in vitro cancer modeling - fantasy or reality?

Personalized medicine or "one size fits all"?

Noninvasive prenatal diagnosis: from dream to reality.

Phalloplasty: The dream and the reality.

Narcolepsy - Between the dream and reality.

A world of cities: dream or nightmare?

Shared or induced obsessive compulsive disorder: Is it a reality?

Non-anticoagulant derivatives of heparin for the management of asthma: distant dream or close reality?

A central supply technician training program. From a dream to a reality.