CHAPTER 4 Pharmacogenetics and Pharmacogenomics
Individual patient differences in drug responsiveness are well recognized by health care professionals. Understanding the basis for these differences is of major clinical and economic importance because of the high frequency of therapeutic failure and adverse reactions to drugs. Patients may receive inadequate or suboptimal benefit, suffer adverse effects from drug treatment, or both. This chapter highlights pharmacogenetic and pharmacogenomic principles and provides illustrative examples where these principles can be applied to optimize therapeutic benefit and minimize adverse effects.
Pharmacogenetics is the branch of pharmacology that seeks to understand the genetic basis for differences in drug responsiveness among humans. The ability to select the safest and most effective drug and dose for a patient based on the patient’s pharmacogenetic profile should simplify the process of adjusting the therapeutic regimen to achieve the desired clinical response. Pharmacogenetics is defined by the U.S. Food and Drug Administration (FDA) as the investigation of the role of variations in DNA sequence on drug response. The term pharmacogenomics—sometimes used interchangeably with pharmacogenomics—is defined by the FDA as the investigation of variations in DNA and RNA characteristics as related to drug response. Pharmacogenomics may also refer to the application of genomic information toward the discovery and development of drugs with new and more specific targets.
FDA labeling regulations now stipulate “if evidence is available to support the safety and effectiveness of the drug only in selected subgroups of the larger population with a disease, the labeling shall describe the evidence and identify specific tests needed for selection and monitoring of patients who need the drug.” So far, the FDA has recommended changes in the labeling for 6-mercaptopurine, irinotecan, and tamoxifen to include pharmacogenetic information on treatment outcome. It has also been estimated more recently that 25% of all outpatients receive at least one prescribed medication that contains pharmacogenetic information in the drug label.18 This information is not generally helpful at present because testing for most pharmacogenetic variants is not yet routine in clinical practice and most general practitioners would be incapable of using such information even if it were currently freely available. The broader field of “personalized medicine” also includes more rational targeting of drugs, such as restricting the use of trastuzumab to treat tumors based on their tumor phenotype (i.e., only those tumors that overexpress human epidermal growth factor receptor 2, the ligand for trastuzumab). This approach to therapy is sometimes included under the umbrella of pharmacogenetic testing.
Pharmacogenetics and pharmacogenomics are areas of intense interest and development within the biotechnology and pharmaceutical industries.49 Many pharmaceutical companies are beginning to genotype patients in premarket clinical trials to exclude individuals who are predicted to experience adverse effects or therapeutic failure. This concept is illustrated in Figure 4-1. A consortium that includes major pharmaceutical companies is assembling a high-density map of the single nucleotide polymorphisms (SNPs) that exist throughout the human genome to facilitate the construction of pharmacogenetic profiles that predict drug responsiveness. SNPs occur, on average, about once every 1000 bases in the 3 billion–base human genome. A website (www.ncbi.nlm.nih.gov/SNP) sponsored by the National Center for Biotechnology Information maintains an updated listing of these SNPs.
FIGURE 4-1 Diagram illustrating the strategy of selecting patients for drug therapy based on response to the drug. Shaded figures, responders who lack a genetic predisposition for toxicity; white figures, inadequate responders; black figures, patients predisposed to toxicity because of a genetic trait.
This chapter presents basic principles and mechanisms that account for pharmacogenetic differences in drug therapy and toxicity. The pharmacogenetic examples included are of historical or clinical interest (or both); they are not intended to be exhaustive because, with the sequencing of the human genome, tabular listings of pharmacogenetic traits of clinical interest fall rapidly out of date. For more comprehensive and contemporary information, readers are directed to monographs,30,59 chapters,22,23 and general reviews11,13,15,25,45,61 on pharmacogenetics and pharmacogenomics, and to the regularly updated Pharmacogenomics Knowledge Base (www.pharmgkb.org), which forms part of the National Institutes of Health–funded Pharmacogenetics Research Network.21
The genome determines the structure, configuration, tissue distribution, subcellular compartmentalization, and concentration of endogenous proteins. In most cases, for a drug to produce a therapeutic or toxic response, it must interact with one or more proteins, which are subject to genetic variation in humans. Genetic differences in plasma proteins may affect the affinity and the extent of drug binding. Genetic differences in the enzymes that metabolize a drug may confer differences in the concentrations of the parent compound, its active metabolites, and toxic derivatives. Genetic differences in cell membrane proteins or drug transporter proteins may influence drug absorption, distribution, and excretion. Finally, patients may have cell surface or intracellular drug receptors mediating therapeutic or adverse effects that are genetically more or less abundant or sensitive than is the norm.
Every year, more than 100,000 people die in the U.S. because they carry “misspelled” genes that make medications either ineffective or deadly. Now doctors can test for the genes before prescribing … Imagine, a lawyer asking: “Doctor, did you know this drug would kill your patient? Did you know there is a test that would have predicted that? And why did you not give your patient the test?”4
Such statements widely read by the lay public (and their lawyers) underscore the need for dental professionals to understand the role of pharmacogenetic factors in drug responsiveness. Patient malpractice claims have already alleged negligence in the use of pharmacogenetic information.50
Proteins affect drug concentration (pharmacokinetics) and response (pharmacodynamics). Historically, genetic variation most often has been identified in pharmacokinetics, particularly in drug metabolizing enzymes.6 Genetic variation in the pharmacokinetic profile often necessitates a change in the dosage regimen of a drug, but not in its selection. Pharmacogenetic differences in drug target responsiveness28 are less well understood, but potentially will also have a significant impact on patient outcomes in the future. In these instances, certain drugs would be contraindicated for patients with particular genotypes. Just as drug prescribers are currently responsible for avoiding adverse “drug-drug” interactions, as described in Appendix 1, they increasingly will be held accountable for avoiding untoward “gene-drug” interactions in clinical practice. Genetic differences in pharmacokinetics and pharmacodynamics are anticipated for many, if not most, drugs, yielding important consequences for drug responsiveness, especially for agents with a narrow therapeutic index. Figure 4-2 illustrates the separate and combined influences of genetic polymorphisms in pharmacokinetics and pharmacodynamics.14
FIGURE 4-2 The potential consequences of administering the same dose of drug to individuals with genetic polymorphisms in both pharmacokinetics (drug-metabolizing enzymes) and pharmacodynamics (drug receptors). Active drug concentrations in the systemic circulation are determined by the individuals’ drug metabolism genotype, with (A) homozygous common (c/c) genotype converting 70% of a dose to the inactive metabolite, leaving 30% to exert an effect on the target receptor. B, For the patient with heterozygous (c/v) drug metabolism genotype, 35% is inactivated, whereas (C) the patient with homozygous mutant (v/v) drug metabolism inactivates only 1% of the drug dose, yielding the three drug concentration time curves. The drug response is further influenced by drug receptor genotypes. Patients with a c/c receptor genotype exhibit a greater therapeutic effect (solid lines) at any given drug concentration in comparison to those with a c/v receptor genotype, whereas those with v/v receptor genotypes are relatively refractory to drug effects at any plasma drug concentration. The combination of genetic polymorphisms in drug metabolism and receptor yields nine different theoretical patterns of drug effect. The therapeutic ratio (efficacy versus toxicity) ranges from very favorable in a patient with c/c genotypes for drug metabolism and drug receptor to very unfavorable in the patient with v/v genotypes. (It is assumed here that the toxic dose-response curve, shown in dotted lines, is not influenced by these polymorphisms.)
(Redrawn from Evans WE, Relling MV: Pharmacogenomics: translating functional genomics into rational therapeutics, Science 286:487-491, 1999.)
There are many gene-drug interactions with importance to dentistry. Codeine, one of the mostly commonly prescribed opioid analgesics for pain relief, is a prodrug and depends on its activation to morphine by CYP2D6, a drug-metabolizing enzyme that is known to exhibit a common genetic polymorphism in humans.48 As a result, codeine is an ineffective analgesic in a significant genetic subset (10%, depending on ethnic group) of the population. Genetic polymorphisms in opioid receptors or in second messenger systems mediating opioid receptor actions have also been observed. If a patient inherits a deficiency in CYP2D6 or the μ-opioid response system, it is unlikely that standard doses of codeine will be of therapeutic benefit. Increasing the dose of codeine to compensate for the genetic deficiency will most likely result not in analgesia but rather in an adverse reaction mediated by overstimulation of an alternative receptor that is responsive to codeine.
An individual’s genotype is a genetic trait defined by the DNA sequences (i.e., alleles) inherited from the mother and the father. An individual can inherit two copies of the same allele (homozygous genotype) or a different allele from each of the parents (heterozygous genotype). The phenotype is a biologic or measurable expression of the genetic trait that depends on the level of penetrance of the gene, the accuracy and selectivity of the method used to measure it, and the influence of environmental factors in the expression of the trait. Historically, one of the most easily measured phenotypes was plasma drug concentration, which is probably why most of the initially identified pharmacogenetic traits were pharmacokinetic phenotypes. Determination of drug concentration is invasive, however, requiring administration of a drug or surrogate chemical and collection of blood samples over time. The drug concentration also depends to varying degrees on patient age, general health, nutritional status, and other factors such as exposure to enzyme inducers and inhibitors. Determination of a patient’s genotype is much less invasive because it does not require administration of a test drug or collection of blood samples over time. Instead, the genotype is determined from a small sample of DNA obtained easily from a buccal swab, hair follicle, or other ready source, and is not affected by age, general health, nutritional status, or other factors. For these same reasons, the prediction of drug response from a genetic test may not always be accurate or reproducible because of the influence of such nongenetic factors on drug response.
Many methods to determine the genotype have been developed in the past two decades, including restriction fragment length polymorphism analysis, allele-specific amplification, and DNA sequencing. Most methods rely on DNA amplification techniques based on the polymerase chain reaction that yield millions of copies of the specific target gene. New high-throughput methods promise to make the simultaneous determination of multiple genotypes readily available to health care professionals.52 Understanding the important and complex relationships between genotypes and phenotypes has fostered much research in functional genomics and proteomics.
A discussion of genetic polymorphisms in enzymes and receptors would be incomplete without consideration of the differences inherent between monogenic and polygenic phenotypes. Monogenic phenotypes derive from genetic variations in a single gene. Monogenic variation often separates populations into discontinuous (bimodal or trimodal) distributions of the phenotype. If the least commonly occurring phenotype arising from the monogenic variation has a frequency of greater than 1% in a population, it is termed a polymorphism. Different drugs or dosing regimens may be appropriate for specific phenotypes. Polygenic traits, in contrast, are phenotypes that derive from some combination of variations in multiple genes. In this case, clearly distinct or discontinuous phenotypes are not observed in a studied population. Instead, there is a unimodal, continuous, normal (Gaussian) distribution of the phenotype. A unimodal distribution of drug response is observed for most drugs metabolized by multiple enzymes, transported by multiple proteins, or acting through multiple receptors or second messenger systems. This unimodal distribution does not mean an absence of genetic variation in one or all of these proteins, but rather that multiple genes contribute to the overall drug response phenotype. Because each of the genes is potentially subject to genetic variation, the utility of genetic information in predicting therapeutic and toxic responses is considerably more complicated. Until recently, polygenic phenotypes were too complex to consider in optimizing drug therapy.
The frequency of specific alleles, genotypes, and phenotypes for drug metabolizing enzymes varies widely with ethnic origin.29 A similar variance is expected for drug receptors. Clinical trials are best conducted either with ethnically diverse study populations to capture differences among ethnic groups with ethnically defined subgroups to define effects in these groups precisely. Some genotyping methods were designed originally to identify only alleles prevalent in whites. With the documented ethnic heterogeneity within the human population, however, genotyping tests need to identify all relevant alleles of a particular gene regardless of ethnic frequency.
As indicated earlier, most of the pharmacogenetic traits identified to date occur in genes encoding drug metabolizing enzymes. It is anticipated that genetic polymorphisms may be identified in all drug metabolizing enzymes. Many of these genetic polymorphisms are already known to be important in therapeutics,63 and examples of historical and clinical interest are highlighted next.