These two examples highlight a vital problem in health care: adverse drug reactions cause about 100,000 deaths per year, according to one 1998 analysis.1 Clinicians' understanding of the interaction between specific drugs and a person's genetic makeup not only has serious clinical implications, it could also help in the development of tests that reduce the trial and error involved in drug selection. Although the field of pharmacogenomics–pharmacogenetics is relatively new, it's generating an enormous amount of information that nurses can put into practice at the bedside. (See Primer on Pharmacogenetics and Pharmacogenomics.2-9) Early research describing the effects of genetic variation on drug response focused on ethnic differences. With the completion of the Human Genome Project, however, it's now possible to associate such variation with precise genetic origins. This creates new opportunities for patient teaching regarding drug selection, which nurses must understand in order to provide patients with an adequate "interpretation of selective genetic and genomic information or services," as required by the American Nurses Association's Essentials of Genetic and Genomic Nursing: Competencies, Curricula Guidelines, and Outcome Indicators.10

The terms pharmacogenetics and pharmacogenomics are often used interchangeably, but they have different meanings. Pharmacogenetics concerns the relationship between a person's genetic makeup and her or his response to medication. Pharmacogenomics refers to the general investigation of all genes that determine drug behavior in the body.2

The starting point for pharmacogenomics research was the identification of variable drug responses in individuals, and then in populations, through the study of geographic, racial, and ethnic differences.3 It's now understood that the manner in which people respond to drugs is influenced by many genes, which exhibit slight variations, or polymorphisms, and that it's possible to test for polymorphisms, which may predict disparities in drug response.2

* Allele: any of the alternative forms of a gene, occurring at a given locus, each inherited from one parent.4

* DNA bases: one of the four essential, nitrogen-containing building blocks that contribute to the structure of DNA nucleotides: adenine (A), guanine (G), cytosine (C), and thymine (T).5

* Genome: a cell's DNA, including mitochondrial DNA and the nuclear chromosomes, of which only an estimated 3% encode proteins.4, 6

* Genotype: the genetic makeup of the entire genome or of specific genes or regions of genes; also, an individual's genetic identity.4, 6

* Pharmacokinetics: the body processes involved in the absorption, distribution, metabolism, and excretion of medications.7

* Polymorphisms: DNA variants that occur within a specific population at a frequency greater than 1%.6

* Single nucleotide polymorphisms (SNPs; pronounced "snips"): polymorphisms involving the substitution of a single nucleotide, or DNA base (adenine, thymine, cytosine, or guanine), for one of the other three. About 90% of polymorphisms are SNPs.6, 8

* Substrate: a drug that binds to and is metabolized by an enzyme. A drug can be a substrate of more than one enzyme.9

Many ongoing research efforts are focused on identifying significant SNPs and associated adverse drug reactions. But DNA sequencing is a very slow and expensive process. SNPs occur roughly once in every 100 to 300 of the 3 billion bases that make up the human genome. Once identified, the SNPs must be analyzed for their propensity to cause disease or variation in drug response.8

SNPs AND LIVER ENZYME FUNCTION

SNPs can affect all cells, including those in the liver that create and secrete the enzymes in the cytochrome P-450 (CYP) enzyme system, which metabolize and break down more than 30 classes of drugs,8 or roughly 60% of common prescription drugs.9 Variations in the genes that encode the CYP enzymes can increase or decrease a person's ability to metabolize drugs.8

The CYP enzymes have been classified into numerous families and subfamilies identified by a sequence of letters and numbers, such as CYP1A2, CYP2B6, CYP2C9, CYP2C19, CYP2D6, and CYP3A4. All of these enzymes play major roles in drug metabolism.9, 11 For example, the CYP2D6 enzymes are involved in the metabolism of antidepressants, antipsychotics, antiarrhythmics, [beta]-blockers, pain medications, and antiemetics, and the CYP2C19 enzymes metabolize many anticonvulsants, proton pump inhibitors, benzodiazepines, and antimalarials.9, 11

Whereas enzymes that are less active than normal can cause drugs to stay within a person's body for too long, creating drug overdoses, enzymes that are more efficient than normal can eliminate a drug too quickly. Based on the activity level of CYP enzymes, patients may be classified into four metabolizer categories: poor, intermediate, extensive, and ultrarapid.7, 12 Clinical drug trials use CYP genotyping in delineating patient populations, and pharmaceutical companies use it to determine how the various CYP enzymes break down their chemical compounds.

Clinically, CYP genotyping may be used to screen patient populations known to have SNPs affecting specific CYP enzymes and impairing drug clearance. Asians, for example, are more likely to exhibit impaired clearance of drugs that are substrates of the enzymes CYP2C19, such as warfarin, and CYP2D6, such as risperidone (Risperdal). Delayed elimination may cause such drugs to reach toxic blood levels. As a result, this population has a disproportionate number of adverse reactions to neuroleptic medications, including extrapyramidal effects such as tremors or rhythmic muscle spasms (tardive dyskinesia).7, 9 Recognizing an ethnic tendency toward poor tolerance to neuroleptic medications and warfarin,13 clinicians often prescribe these drugs at lower doses in Asian patients, which appears to produce the desired effectiveness. Ethnic variation in drug response may be significant when the original drug tests and studies are performed in white populations, the majority of whom do not have a SNP affecting the CYP2C9, CYP2C19, and CYP2D6 enzymes.7, 14

Another confounding variable is that a person may be missing certain genes or have multiple active copies of a specific gene. This apparently affects substrates of the CYP2D6 enzyme most frequently.7 If the gene is absent, patients don't metabolize the drug, but patients with one to three or more active copies very quickly metabolize the enzyme's substrates,7 which include the prodrugs codeine and tramadol.11 With certain drugs, being able to anticipate such variation could be extremely useful.

NEW PHARMACOGENOMIC TESTS

Several companies manufacture tests that determine patients' DNA sequencing from a blood sample. The first such test cleared for clinical use, the AmpliChip CYP450 Test, became available in the United States in 2005. Using microarray chip technology, the test creates a report indicating which CYP2D6 and CYP2C19 variations are present and whether the patient is a poor, intermediate, extensive, or ultrarapid metabolizer of each enzyme's substrates.15 It can help determine the metabolism involved in about 25% of prescription drug classes.16

The list of known CYP substrates is continually growing. The P450 Drug Interaction Table on the Web site of the Indiana University School of Medicine (available at http://bit.ly/19UX4g ) is updated frequently and provides links to the names of the primary studies identifying the listed substrates.11 Nurses can use the information to evaluate a patient's response to medication and assist in determining whether the patient is a candidate for DNA sequence testing, particularly in cases suggesting underdosing or overdosing caused by aberrant metabolism.

This explosion of investigations in pharmacogenomics has required the National Center for Biotechnology Information to catalog and describe data on both SNPs and microarray information.2 Pharmacogenomic screening may eventually eliminate from drug intervention trials participants who would have a poor response (no effect or an adverse effect) to the tested drug. Excluding such populations might allow research to more accurately determine a drug's efficacy in specific patient populations, thereby improving its chances of approval. That would require nurses to determine which populations had been excluded from drug intervention trials prior to administering newly approved drugs.

PRESCREENING FOR SPECIFIC DRUGS

Pharmacogenomic tests have now been developed specifically to screen patients for potentially poor responses to psychotropic drugs,9, 17 as well as to drugs such as warfarin, statins, tacrine (Cognex), codeine, azathioprine (Imuran), and irinotecan (Camptosar), which are discussed below.

Warfarin is the principal oral anticoagulant used worldwide for long-term treatment and prevention of thromboembolic disease and one of the 20 most commonly used drugs in the United States.18 Because of its narrow therapeutic index and wide interindividual variability, however, warfarin is second only to insulin in requiring emergency treatment for adverse drug reactions.19 Recent research into polymorphisms affecting vitamin K epoxide reductase complex subunit 1 (VKORC1) and CYP2C9 liver pathways makes a strong case for prospectively genotyping warfarin candidates.18

Although patient responses to drugs vary, warfarin is atypical in that it's titrated so that all patients achieve the same goal: an increase in clotting time of two to three times. Nevertheless, therapeutic dosages may vary widely—up to tenfold in any given patient population.18, 20 Regulating the dosage can take weeks to months, during which time the patient risks clotting if the dose is too low and bleeding if the dose is too high. Nongenetic factors (age, sex, body mass index, vitamin K intake, and interacting drugs) are estimated to account for 30% of the dose variability, but the rest may be due to genetic factors.18

The elimination of warfarin is accomplished almost completely through metabolism by CYP liver enzymes, principally CYP2C9,20 although CYP2C19, CYP2C8, CYP2C18, and others also contribute.13 This helps explain why Asian patients, like Ms. Kim, who are known to exhibit impaired clearance of CYP2C19 substrates, require lower initiation and maintenance doses of warfarin. Ms. Kim's advanced age would put her at additional risk for excessive anticoagulation with warfarin because elderly patients are more sensitive to the drug and, in response, produce a greater prothrombin time. As patients age, lower doses usually maintain therapeutic levels.

Variant alleles of the CYP2C9 enzyme reduce the metabolism of warfarin, and such alleles are estimated to occur in 18% of the white population. A metaanalysis of nine studies demonstrated a practical application of this finding. White populations carrying at least one copy of the variant CYP2C9*2 or CYP2C9*3 alleles required dose reductions of 17% and 37%, respectively.13 Other variant alleles of CPY2C9 occur at lower frequencies in black populations.13

In addition to CYP2C9 variability, there is substantial variation in the gene VKORC1, which encodes the key enzyme involved in the synthesis of vitamin K, a necessary element in activating several clotting factors.20 The degree to which warfarin suppresses vitamin K synthesis depends on drug dose and the patient's VKORC1 genotype.

Usual doses of warfarin depress vitamin K synthesis by 30% to 50%.13 Together with clinical factors such as body size, age, and interacting drugs, CYP2C9 and VKORC1 genotypes can explain about 55% to 56% of the variability in therapeutic warfarin doses in white patients.13, 21

In 2007 the Food and Drug Administration (FDA) approved updated labeling for warfarin that explained more fully how genetic composition may affect drug metabolism.19 In 2008 two articles published in the New England Journal of Medicine alerted clinicians that, during the early weeks of warfarin administration, VKORC1 variation is the most important factor in modulating response to the drug. After the first two weeks, genetic variation in both VKORC1 and CYP2C9 contribute substantially to dose variability but not to the incidence of bleeding episodes.22, 23 Ideally, VKORC1 and CYP2C9 testing will soon be readily available during the prescribing period.24

The American Medical Association has recently published a brochure to raise awareness among prescribers of the effects of genotype on warfarin dosing,25 and the Barnes-Jewish Hospital at Washington University Medical Center, in conjunction with the National Institutes of Health, has constructed a Web site to assist clinicians in estimating therapeutic doses in patients new to warfarin (available at www.warfarindosing.org ). The interactive dosage calculator bases estimates on clinical factors such as age, sex, race, body weight, target international normalized ratio, interacting drugs, and (when available) CYP2C9 and VKORC1 genotypes.

One gene that has been identified as a response predictor for the statin drug class (used to lower lipids and slow the progression of coronary artery disease) and the drug tacrine (used in therapy for Alzheimer's disease) is apolipoprotein E (APOE) and its allelic variants, APOE[varepsilon]2, APOE[varepsilon]3, and APOE[varepsilon]4.26 APOE is a major binding protein for very-low-density lipoprotein and intermediate-density lipoprotein cholesterol, and it has 19 SNPs associated with it.27 A genetic receptor (or drug enzyme target) for the statin class of drugs is 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMG-CoA reductase), which has 33 SNPs associated with it.27

Evans and McLeod found that patients with the APOE allele variant APOE[varepsilon]4, which is associated with an elevated risk of cardiovascular disease,28 received the least benefit from statin therapy, whereas those with the APOE[varepsilon]2 allele had the greatest reduction in serum low-density lipoprotein (LDL) cholesterol.26 These effects, however, have not been replicated in other studies.26

Chasman and colleagues found that patients with two SNPs in the HMG-CoA reductase gene (SNP 12 and SNP 29) had 22% less reduction of total cholesterol after 24 weeks of statin therapy and that SNP 17 in the APOE gene was associated with baseline LDL cholesterol levels but not with changes in total serum lipids after 24 weeks of statin therapy.27 Findings from these studies suggest that genetic polymorphisms determine the effectiveness of statin therapy.

Tacrine. There is a known response association between the APOE genotype and current medication for Alzheimer's disease.26 Now a study of patients with Alzheimer's disease indicates that more patients without the APOE[varepsilon]4 allele have an improved total response and cognitive response to 30 weeks of tacrine therapy than patients who have at least one APOE[varepsilon]4 allele (83% versus 40%, respectively). Additional studies of this phenomenon suggest that female patients with the APOE[varepsilon]4 allele have unknown gene interactions that may explain their relative lack of response to tacrine.26

Codeine. Since codeine is a prodrug (inert drug), it becomes an active analgesic only when converted into morphine during its metabolism by CYP2D6 enzymes. Codeine clearance, on the other hand, is accomplished through the CYP3A4 enzymes and renal activity. Typically, CYP2D6 enzymes convert only about 10% of the codeine a patient ingests into morphine before CYP3A4 enzymes and renal activity remove 80% of it from the body.29 If, however, a patient is an ultrarapid metabolizer of codeine (resulting from genetic variants that amplify CYP2D6 metabolism) and also has inhibited CYP3A4 activity or impaired renal function (resulting from other medications or comorbid conditions), the patient is at risk for opioid intoxication with central nervous system depression and subsequent respiratory inhibition—even when given small doses of codeine.29 In such circumstances, the value of genotyping for specific CYP enzymes is great. Another consideration with codeine is that 7% to 10% of whites have non-functional, mutant alleles in the CYP2D6 pathway that inhibit its conversion to morphine, making it ineffective for pain management at usual dosages.29

The P450 Drug Interaction Table indicates that the CYP2D6 pathway has several alleles that may inhibit or intensify drug metabolism.11 When a patient whose CYP2D6 genotype is unknown has an adverse reaction to a pain medication that uses this pathway—such as codeine, tramadol, or oxycodone (OxyContin)—the nurse may want to suggest the use of a substitute.

PHARMACOGENOMICS AND CANCER TREATMENT

Azathioprine, an important chemotherapy compound, is used to treat childhood acute lymphoblastic leukemia, inflammatory bowel disease, and autoimmune disorders. It's metabolized by a liver CYP enzyme called thiopurine methyltransferase (TPMT), which is associated with the CYP2B6 pathway.14 A rather small but significant number of whites have genetic polymorphisms in TPMT, causing thiopurine levels to become highly toxic during azathioprine therapy.8, 14 This has caused life-threatening, acute bone marrow failure in a number of patients, necessitating bone marrow transplantation in some cases.30

Irinotecan is used in the treatment of cancers of the brain, breast, colon and rectum, and lung. The active form of irinotecan, SN-38, is metabolized by the polymorphic enzyme UGT1A1. Some people, however, have a variant gene allele, called UGT1A1*28, which lowers the amount of UGT1A1 enzyme produced, causing SN-38 to rise above therapeutic levels. Patients with the UGT1A1*28 allele are more likely to have adverse effects, particularly fatal or severe neutropenia, when treated with irinotecan. In 2005 the FDA approved a UGT1A1*28 allele clinical test and irinotecan dosing guidelines for patients with the variant allele.31

In a study of 65 patients with rectal cancer (stages T2 to T4) who were treated preoperatively with 5-fluorouracil, those with at least one TSER*2 allele had a 38% higher rate of tumor shrinkage at the point of surgical resection, compared with those who had two TSER*3 alleles (genotype TSER*3/TSER*3).31 In an ongoing clinical trial, patients with the TSER*2 allele were treated with radiation and 5-fluorouracil (standard therapy), while patients with the TSER*3/TSER*3 genotype, which is predictive of 5-fluorouracil resistance, received standard therapy plus irinotecan. Preliminary trial results suggest improved response rates in both groups, supporting the value of genetic testing in directing cancer treatment.31

CONSENT, DISCLOSURE, AND PUBLIC POLICY

Most discussions in the health care literature have focused on the identification of people at risk for specific Mendelian genetic disorders and not on pretreatment genetic testing for adverse medication effects. As Williams and colleagues have pointed out, however, regardless of the purpose of a genetic test, results may reveal genetic alterations that could conceivably affect employment, insurance eligibility, or family relationships. Yet when patients undergo pretreatment genetic testing for susceptibility to adverse drug reactions, informed consent protocols used for Mendelian testing might not be followed.32 The risk and significance to patients from an absence of informed consent for such testing hasn't been determined, but as patient advocates, nurses will want to ensure that patients' decisions are well informed.

There is national consensus that health care professionals require education and training in order to translate genetic—and, now, genomic—information into practice. In the United States, surveys of nursing programs and related health care organizations have identified a number of approaches to improving nursing education. These include10, 32, 33

Understanding genetic contributors to drug response variability allows nurses to look for unexpected toxicities and predict which patients are more likely to respond optimally to a prescribed medication. Nursing issues raised by pharmacogenomic research are very broad, including liability, confidentiality, access to treatment, and discrimination. Nurses need to be knowledgeable about sources, retrieval, and uses of pharmacogenomic information. When drugs are prescribed for which pharmacogenomic testing is known to be useful, nurses need to inquire about the precise test protocols and outcomes, and educate patients accordingly. Ultimately, pretreatment screening for genetic and genomic variants may become routine.

For more than 23 additional continuing nursing education articles related to the topic of drug therapy and two related to the topic of genetics, go to www.nursingcenter.com/ce .

REFERENCES

1. Lazarou J, et al. Incidence of adverse drug reactions in hospitalized patients: a meta-analysis of prospective studies. JAMA 1998;279(15):1200–5. [Context Link]

2. National Center for Biotechnology Information. Just the facts: a basic introduction to the science underlying NCBI resources. Bethesda, MD; 2004. [Context Link]

3. Kudzma EC. Culturally competent drug administration. Am J Nurs 1999;99(8):46–51. [Context Link]

4. Roden DM, et al. Pharmacogenomics: challenges and opportunities. Ann Intern Med 2006;145(10):749–57. [Context Link]

5. Marieb EN. Human anatomy and physiology. 6th ed. New York: Pearson Education; 2004. [Context Link]

6. Wung SF. Genetic advances in coronary artery disease. Medsurg Nurs 2002;11(6):296–300. [Context Link]

7. Ng CH, et al. The emerging role of pharmacogenetics: implications for clinical psychiatry. Aust N Z J Psychiatry 2004;38(7):483–9. [Context Link]

8. Human Genome Project Information. Pharmacogenomics. U.S. Department of Energy Office of Science, Genome Management Information System. 2006. http://www.ornl.gov/sci/techresources/Human_Genome/medicine/pharma.shtml . [Context Link]

9. Howland RH. Personalized drug therapy with pharmacogenetics. Part 1: pharmacokinetics. J Psychosoc Nurs Ment Health Serv 2006;44(1):13–6. [Context Link]

10. Consensus Panel on Genetic/Genomic Nursing Competencies. Essentials of genetic and genomic nursing: competencies, curricula guidelines, and outcome indicators. 2nd ed. Silver Spring, MD: American Nurses Association; 2009. [Context Link]

11. Flockhart DA. Cytochrome P450 Drug Interaction Table. Indianapolis, IN: Indiana University School of Medicine, Division of Clinical Pharmacology; 2007. [Context Link]

12. Lea DH. Tailoring drug therapy with pharmacogenetics. Nursing 2005;35(4):22–3. [Context Link]

13. Bristol-Myers Squibb Company. Coumadin tablets (warfarin sodium tablets, USP) crystalline; Coumadin for injection (warfarin sodium for injection, USP) [prescribing information]. 2007. http://www.accessdata.fda.gov/drugsatfda_docs/label/2007/009218s105lblv2.pdf . [Context Link]

14. Shastry BS. Pharmacogenetics and the concept of individualized medicine. Pharmacogenomics J 2006;6(1):16–21. [Context Link]

15. Roche Diagnostics. Background information: the technology behind AmpliChip microarrays. Basel, Switzerland: F. Hoffmann-La Roche Ltd.; 2007. [Context Link]

16. Roche Diagnostics. Product list: AmpliChip CYP450 test. F. Hoffmann-La Roche Ltd. 2006. http://www.roche.com/products/prod_diag_amplichip.htm . [Context Link]

17. de Leon J, et al. Clinical guidelines for psychiatrists for the use of pharmacogenetic testing for CYP450 2D6 and CYP450 2C19. Psychosomatics 2006;47(1):75–85. [Context Link]

18. Rettie AE, Tai G. The pharmacogenomics of warfarin: closing in on personalized medicine. Mol Interv 2006;6(4):223–7. [Context Link]

19. FDA News and Events. FDA approves updated warfarin (Coumadin) prescribing information. New genetic information may help providers improve initial dosing estimates of the anticoagulant for individual patients [press release]. 2007 Aug 16. http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/2007/ucm108967.htm . [Context Link]

20. Singh A, Emery J. Pharmacogenomics—the potential of genetically guided prescribing. Aust Fam Physician 2007;36(10):820–4. [Context Link]

21. Caldwell MD, et al. Evaluation of genetic factors for warfarin dose prediction. Clin Med Res 2007;5(1):8–16. [Context Link]

22. Schwarz UI, et al. Genetic determinants of response to warfarin during initial anticoagulation. N Engl J Med 2008;358(10):999–1008. [Context Link]

23. Shurin SB, Nabel EG. Pharmacogenomics—ready for prime time? N Engl J Med 2008;358(10):1061–3. [Context Link]

24. Diagnostic Testing and Intelligence Report. NEJM study probes pharmacogenomics of warfarin metabolism. Washington, DC: Washington G-2 Reports 2008. [Context Link]

25. American Medical Association, Arizona CERT, Critical Path Institute. Personalized health care report 2008: warfarin and genetic testing. Tucson, AZ: Arizona Center for Education and Research on Therapeutics; 2008. http://www.azcert.org/medical-pros/warfarin_brochure.pdf . [Context Link]

26. Evans WE, McLeod HL. Pharmacogenomics—drug disposition, drug targets, and side effects. N Engl J Med 2003;348(6):538–49. [Context Link]

27. Chasman DI, et al. Pharmacogenetic study of statin therapy and cholesterol reduction. JAMA 2004;291(23):2821–7. [Context Link]

28. Kulminski AM, et al. Health-protective and adverse effects of the apolipoprotein E [varepsilon]2 allele in older men. J Am Geriatr Soc 2008;56(3):478–83. [Context Link]

29. Gasche Y, et al. Codeine intoxication associated with ultrarapid CYP2D6 metabolism. N Engl J Med 2004;351(27):2827–31. [Context Link]

30. Mendible JC. Pharmacogenomics: medicines tailored just for you. Vancouver, BC: Suite101.com Media Inc. 2000. [Context Link]

31. Marsh S, McLeod HL. Pharmacogenomics: from bedside to clinical practice. Hum Mol Genet 2006;15 Spec No 1:R89–R93. [Context Link]

32. Williams JK, et al. Ethics, policy, and educational issues in genetic testing. J Nurs Scholarsh 2006;38(2):119–25. [Context Link]

33. Health Resources and Services Administration, Division of Nursing. Report of the Expert Panel on Genetics and Nursing: implications for education and practice. Expert Panel on Genetics and Nursing; 2000 Sept. 28–29; Washington, DC: U.S. Department of Health and Human Services, Bureau of Health Professions, Division of Nursing; 2000. [Context Link]