Elizabeth L. Black, MD, Blue Mountain Family Health, Clarkston, WA
Brian T. Hocum, PharmD, CGP, Adjunct Faculty Member, Washington State University College of Pharmacy, Spokane, WA; Regional Pharmacist Liaison, Genelex Corporation, Seattle, WA
Kevin J. Black, MAT, Clarkston, WA
Joseph Kitzmiller, MD, PhD, FCP, Colleges of Engineering and Medicine, Associate Director, Center for Pharmacogenomics Director, Clinical Pharmacology Fellowship, College of Medicine, The Ohio State University, Columbus, OH
With the rapidly increasing number of drugs available to the primary care physician’s armamentarium, the rational and judicious use of pharmacogenetics (PGx) can improve drug selection by increasing the likelihood of effectiveness and reduce harmful side effects.
- Adverse drug events contributed to 13.5 million
outpatient and ED visits over a recent 3-year period, with the elderly particularly vulnerable.
- The increased utilization of health care resources may
be contributing up to 13% of the total spending on healthcare in the United States.
- The FDA refers to alleles that influence drug effectiveness and toxicity as "pharmacogenetic biomarkers.” PGx biomarkers are further classified as either pharmacokinetic (PK) or pharmacodynamic (PD). PK biomarkers affect how the body absorbs, distributes, metabolizes, and excretes drugs. PD biomarkers are less well understood. These biomarkers affect the action of drugs at the molecular level.
- Clinical scenarios commonly targeted for pharmacogenomic investigation include statin, warfarin anticoagulation, clopidogrel, pain management, and a host of psychotropic medications.
Advancing technology has always challenged physicians in their practice of medicine. New research, techniques, and treatments can improve the prevention and management of disease, but not without confusion and occasional controversy. The addition of pharmacogenetic testing to the armamentarium of primary care providers (PCPs) presents just such a challenge.
Pharmacogenetics and pharmacogenomics are very similar terms that often are used interchangeably. Authors tend to use pharmacogenomics when discussing broader research about the relationship between the genome and pharmacotherapy, such as in genome wide association studies (GWAS). Pharmacogenomics usually applies to a population. On the other hand, pharmacogenetics more often is used when talking about specific genes and their influence on specific drugs. An example of this would be the study of how cytochrome P450 2C9 (CYP2C9) and VKORC1 polymorphisms affect warfarin pharmacokinetics and pharmacodynamics.1 Pharmacogenetics deals with individuals. This article will distinguish the two terms and will focus on pharmacogenetics, which will hereafter be abbreviated, PGx.
Pharmaceutical management is becoming the standard of care for many medical conditions. Many evidence-based standards for quality in the treatment of certain disease states support the introduction of particular drug therapies in the presence of known diagnoses. Even if specific medications or classes of medications are not defined by quality standards, defined management goals may not be achieved without medication.
The rationale for promoting evidence-based standards is clear to most physicians, although the benefit of such programs is debated. Research presents convincing evidence that the risks of many disease states are significantly reduced by the introduction of pharmaceutical agents and adherence to treatment goals. However, these drugs do not come without their own risks.
Many management recommendations are drawn from well-designed studies, but these focus on results in broad populations rather than individuals. Medications shown to be effective in these studies may be less effective in a particular patient, resulting in failure to reach desired treatment goals. Medications can also cause unintended effects. While most often inconvenient or uncomfortable, adverse drug reactions (ADRs) can be very dangerous.
ADRs and/or treatment failure may erode patient confidence in their physician, or perhaps in the validity of the evidence used as a basis for their recommendation. This can lead to mistrust of pharmaceutical treatment as a whole and can reduce patient motivation to meet health management goals. This fear may contribute to the preference of some individuals for alternative therapies. Although many of these therapies have poor evidence of effectiveness, they often are embraced as an alternative to the anecdotal failures of recommended medications.
The emergence of PGx testing offers promise in mitigating some risks associated with medication therapy. Testing for known genetic variants that affect drug metabolism can potentially enhance therapeutic response to medication, reduce ADRs, and optimize treatment of disease. While this can positively affect both disease-specific outcomes and patient satisfaction, PGx testing has its own complications. Testing may indicate that a commonly available medication is less advisable for a particular patient, but alternatives may not be readily accessible. Also, the novelty of the technology may lead to patient uncertainty regarding the significance and implications of genetic testing results. Most health care professionals also feel uncertain about the utility of PGx testing and discussing it with their patients.2
Consider the following hypothetical case:
History of Present Illness: GS is a 46-year-old female office administrator who presents for follow-up on type 2 diabetes, depression, hypertension, and hyperlipidemia. Recent fasting lab work is available for review. She is taking all medications as prescribed. She reports that her depression has recently worsened in conjunction with some stress at home and at work. She is taking citalopram, and her symptoms were previously well-controlled on that medication, but over the last few months, she has had increased dysphoria, anxiety, and anhedonia. She is interested in an antidepressant dose increase to help control her symptoms. She is having some trouble affording Crestor as it is tier 3 on her insurance. She asks if a lower-cost generic would be appropriate for her.
Past Medical History: hypertension, diabetes, hyperlipidemia, major depression, anxiety
- Lisinopril 20 mg daily
- HCTZ 25 mg daily
- Metformin 500 mg BID
- Rosuvastatin 10 mg daily
- Citalopram 20 mg daily
- Lorazepam 2 mg TID prn
Recent Lab Results:
- TSH and CBC WNL
- CMP WNL except fasting
- LDL: 110 mg/dL
- CrCl: 80 ml/min
Framingham’s Risk Score: 20%
In the absence of additional information, a PCP might simply try an alternative statin and see if it was tolerated at a sufficient dose to achieve the desired result. The dosage of citalopram could be increased to see if she achieves better results. However, the potential for both tolerance and effectiveness will not be known until after therapy is changed.
Bridging the gap in provider understanding of PGx is essential to the future of primary care. PCPs are ideally positioned to counsel patients in a manner that will make this technology most clinically meaningful. Not only are the vast majority of prescriptions written in the primary care setting,3 but an estimated 60% of office visits related to adverse drug events (ADEs) take place in the primary care setting.4 Patients are also more likely to report ADEs in that environment.4 Further, patients have indicated a preference for discussing PGx test results with their PCP.5
GS presents with a common constellation of issues. Cost concerns conflict with efforts to reach therapeutic goals and reduce associated risks from known chronic conditions. She also presents with a psychological complaint common to primary care. The very familiarity of this scenario illustrates how PCPs are in an ideal position to use additional information from PGx testing to guide decision making to facilitate positive results, or avoid potential hazards. One of those hazards is the risk of adverse drug reaction to medication.
ADEs and ADRs
ADEs may be classified as any injury resulting from drug use. They contributed to an estimated 13.5 million outpatient visits between 2005 and 2007, including emergency department (ED) and physician office visits.4 The elderly are particularly vulnerable in this regard. An estimated 100,000 ED visits related to ADEs among Americans 65 and older resulted in hospital admissions between 2007 and 2009.6 Polypharmacy also increases the risk of ADEs. Older Americans have the highest rate of health care resource utilization in relation to ADEs, but the highest absolute number of such visits occurred among 45- to 64-year-old patients. In fact, once data are adjusted for comorbidities and number of medications, the effect of age on the rate of visits related to ADEs is greatly reduced.4 This indicates that the problems of medication management are not exclusive to the geriatric population. The increased utilization of health care resources associated with ADEs may contribute to up to 13% of the total spending on health care in the United States.7 ADRs are a special type of ADE that occur at commonly prescribed doses, making them of particular interest in pharmacogenomics and PGx.8
Reducing the incidence of ADRs may reduce burden on the health care system overall and produce cost savings by preventing a portion of drug-related hospitalizations. Achieving this result will rely on PCPs’ understanding of pharmacogenomics and applied PGx.
Tables 1 and 3 show that three of GS’s medications are associated with genetic variants that may affect the outcome of therapeutic changes. One of these, citalopram, is associated with a variant that carries an FDA-recommended dosage limitation for patients with a known genetic variant. This genetic variant is associated with increased risk of QTc prolongation and Torsades de Pointes, a potentially severe complication. Information like this is currently widely available and an understanding of how to use it will help PCPs make therapeutic decisions that are more likely to avoid ADRs and improve therapeutic outcomes.
Knowledge, Confidence, and Attitudes Toward PGx
PGx is an area where PCPs report a low index of confidence. In one national survey, the majority of PCPs reported that they do not feel well informed about PGx testing. Although more than half indicated that they received genetics training in medical school, most felt that the training was inadequate to prepare them to use PGx testing in their clinical decision-making. The lack of confidence in using PGx test results may be a reason why many PCPs responding to the survey reported that they have never ordered PGx testing.2 A multi-specialty survey of U.S. physicians shows that primary care is not unique in this regard. With the sole exception of oncology specialists, the vast majority of physicians surveyed did not regularly order genetic testing, citing lack of information.9
While PCPs indicate uncertainty about how to use PGx testing in clinical practice, there is broad acknowledgement of its potential utility. In the multi-specialty survey, 97.6% of respondents believed that genetics may influence a patient’s response to drug therapies.9 In the PCP survey, almost two-thirds of respondents agreed that PGx testing represents a valuable potential tool to predict risk of ADRs or likelihood of efficacy. Among physicians who have utilized PGx testing, reducing drug toxicity and improving effectiveness were cited as significant observed benefits to patients.2
PCPs envision a strong role in the future of using PGx information in clinical practice. Most of the PCPs surveyed felt that informing patients of PGx testing availability and recording PGx testing results in patient records should be a responsibility of the PCP. More than half felt that PCPs should also be responsible for informing their patients of PGx testing results. Beyond that point, however, uncertainty again emerges, as less than half of surveyed PCPs felt that they should be primarily responsible for determining how PGx results should be used in medication management.9
Principles of PGx
SNPs, Alleles, Genotypes and Phenotypes. Different forms of genes that are passed on from parent to child are called alleles. The combination of alleles an individual inherits determines his or her genotype, and the expression of these alleles determines his or her phenotype. Genetic variation arises from the introduction of mutations, or alterations in the DNA sequence, in these alleles. The most commonly identified mutations are single nucleotide polymorphisms, also called SNPs. A particular SNP may or may not result in changes in protein regulation, expression, or activity.
When an identified SNP negatively affects protein function it is termed a loss-of-function allele. Someone with one (heterozygote) or two (homozygote) loss-of-function alleles will have less overall protein expression and/or activity compared to someone with two normal-function alleles. When an SNP is identified that positively affects protein function, it is termed a gain-of-function allele. The presence of a gain-of-function allele, or duplication of a normal function allele, may result in increased protein expression and/or enhanced activity. These genotypes can have a direct impact on numerous metabolic functions, including how individuals respond to certain drugs at a cellular level.
The effect that genetic variations have on drug metabolism is characterized by well-established phenotypes. A poor metabolizer is an individual with two inactive or loss-of-function alleles. In patients with this phenotype, drugs may not be metabolized efficiently. This can result in increased drug concentrations that can reach toxic levels. Ultrarapid metabolizers have gene duplicates and therefore increased drug metabolism. This can result in subtherapeutic drug levels at doses that would likely be effective in normal metabolizers. Figure 1 illustrates the consequences that these genetic variations can have on drug metabolism and therefore effectiveness and toxicity. Loss- and gain-of-function alleles may also result in altered response to medications due to abnormal binding at its site of action or receptor.
Genetic testing for GS yields the following information:
- CYP2D6 (*1/*1)
- CYP2C9 (*1/*1)
- CYP2C19 (*2/*2)
- CYP3A4 (*1/*1)
- CYP3A5 (*3/*3) Non-expressor
- SLCO1B1/OAT1B1 (*1/*1) Normal Transporter
- VKORC1 (A/A)
High Sensitivity to Warfarin
Types of PGx Biomarkers. The FDA refers to alleles that influence drug effectiveness and toxicity as "pharmacogenetic biomarkers.”10 PGx biomarkers are further classified as either pharmacokinetic (PK) or pharmacodynamic (PD). PK biomarkers affect how the body absorbs, distributes, metabolizes, and excretes drugs. Their effects on drug bioavailability, blood concentrations, and distribution into tissues are easy to measure11 and therefore they are well understood and studied. This class of biomarkers includes genes that code for drug-metabolizing enzymes such as CYP2D6, CYP2C9, and CYP2C19. Also included are genes that code for transporter proteins like OAT1B1 and P-glycoprotein. Drug-metabolizing enzymes like the cytochrome P450s biotransform drugs into metabolites more readily eliminated by the body or modified by other enzymes.12 Transporters function to move drugs in and out of cells and across barriers like the small intestine, liver, kidney, and brain. For this reason they are sometimes referred to as "gatekeepers.”13 They are also involved in directly eliminating drugs via biliary and urinary excretion. Two important biomarkers involved in drug transport include SLCO1B1, which codes for the OAT1B1 transporter, and ABCB1, sometimes called the "multidrug resistance gene,” which codes for p-glycoprotein.14 OAT1B1 is an influx transporter meaning it moves drugs into cells and p-glycloprotein is an efflux transporter meaning it moves drugs out of cells and into the intestinal tract, bile, blood, or urine.15
PD biomarkers are less well understood. These biomarkers affect the action of drugs at the molecular level. Their effects are harder to isolate and measure, and uncertainty of the exact mechanism of action for many drugs further limits research in this area. Just as there are different subtypes of PK biomarkers, there are also subtypes of PD biomarkers. Some impact drug response directly whereas others may play a more indirect role as a result of a genetic variance that affects the underlying disease process. Indirect PD biomarkers may still significantly influence the efficacy, toxicity, and/or laboratory values of treatment. An example of a direct effect would be opioid binding to the mu-opioid receptor. Genetic variations of the OPRM1 gene, which codes for the mu-opioid receptor, may affect the amount of pain attenuation achieved with opioids.11 An example of an indirect PD biomarker would be HLA-B*1502, which is strongly associated with carbamazepine use and the risk of Stevens-Johnson Syndrome (SJS) and toxic epidermal necrolysis (TEN), despite it not being involved in the drug’s known mechanism of action.16,17 APOE is an example of a PD biomarker that is associated with laboratory values. Certain variations in APOE are associated with greater LDL reductions in patients being treated for high cholesterol.18
While PGx biomarkers can increase physician knowledge of how patients will potentially respond to medications, some PK biomarkers may affect the metabolism of a drug in a manner that does not result in a PD difference, either good or bad. A list of significant PGx biomarkers may be found on the FDA website.10 Tables 1 and 2 describe some of the important PK and PD biomarkers.
GS Treatment plan:
Plan: #1 Prescribe simvastatin 40 mg daily, #2 Diet and exercise recommendations, #3 Change citalopram to paroxetine, #4 Increased psychologist sessions.
GS’s PGx results reveal important aspects with regard to current and future medication management. CYP2D6, CYP2C9, and CYP2C19 are all highly polymorphic. Together they metabolize approximately 75% of all hepatically metabolized medications. Currently, citalopram is the only medication GS is receiving that goes through one of these pathways, namely CYP2C19. Being normal at the CYP2D6 and CYP2C9 pathways means that choosing medications that go through these pathways can be administered at standard doses. Being a CYP2C19 poor metabolizer limits the dose or therapeutic option of medications that are substrates for this pathway. At 20 mg daily, citalopram was at a maximum daily dose per the product insert. Therefore, choosing paroxetine, an SSRI metabolized by the CYP2D6 pathway, creates therapeutic options for management of this patient’s depression.
CYP3A4 has many substrates, including several of the statins. Individuals with decreased expression/activity of CYP3A4 may require decreased doses to achieve similar therapeutic effects and acceptable risk compared to normal metabolizers. Because GS is a CYP3A4 normal metabolizer, standard doses of a statin metabolized by this pathway may be appropriate. SLCO1B1 is an influx transporter involved in the transport of many medications including many statins. GS’s normal SLCO1B1 results allow for the prescriber to more confidently initiate therapy with a SLCO1B1 transported statin that might otherwise result in statin intolerance due to inadvertent increased exposure of the drug. This information leads to the use of simvastatin 40 mg daily for cholesterol management.
Applied PGx. To illustrate how PGx information can affect pharmaceutical management in clinical practice, an illustration drawn from a class of medications that are both commonly used and significantly affected by known biomarkers may be useful. HMG-coenzyme A reductase inhibitors (statins) are an excellent class of drugs to illustrate the value and limitations of PGx in everyday primary care.
Cholesterol management is a pillar in primary care and statins are the most commonly prescribed pharmaceutical therapy in the United States.19 They also display extensive interpatient variation in blood levels.20 Although generally safe, this variation undoubtedly leads to toxicity in some individuals or decreased efficacy in others.21 Table 3 shows some of the PGx biomarkers associated with variances in statin therapy.
The significance of PK biomarkers is similar to PK drug interactions. For example, coadministration of grapefruit and statins is known to increase the blood levels of some statins and increase the risk of ADRs including myalgia and rhabdomyolysis.22 Coadministration of atorvastatin and grapefruit has resulted in up to a 50% increase in atorvastatin area under the curve (AUC).23
The mechanism for this interaction is intestinal inhibition of CYP3A4 by grapefruit. CYP3A4 is one of the enzymes that metabolizes atorvastatin into metabolites. The impaired metabolism of atorvastatin as a result of this inhibition results in increased bioavailability of the drug, therefore increasing the risk of ADRs. A loss-of-function allele for CYP3A4 may have similar effects. Indeed a clinical study observed a 78% decrease in atorvastatin dose requirements in CYP3A4*22 allele carriers compared to non-carriers (see Table 3).
In these two scenarios, the clinician faces a similar mechanism potentially affecting the therapeutic outcome of their prescribing decision. The obvious distinction is that while a patient can be advised not to consume grapefruit when atorvastatin is prescribed, a patient’s CYP3A4 loss-of-function allele is inherent. However, the clinician may either opt to decrease the atorvastatin dose or choose an alternative statin that does not undergo significant metabolism by CYP3A4 such as rosuvastatin, pitavastatin, pravastatin, or fluvastatin (see Table 4).
PD biomarkers can be more challenging to use clinically.11 When evidence is sufficient, however, they can help to understand an individual’s overall sensitivity to a medication. For example, one study found patients’ response to atorvastatin 20 mg daily × 14 days varied based on their HMG coenzyme A reductase (HMGCR) genotype. HMGCR expresses phenotypically as the protein that serves as a receptor for statins. Results showed a 15-23% greater LDL reduction in HMGCR rs3846662 AA homozygotes compared to GG homozygotes (see Table 3).24 While this information identifies patients who are most likely to respond to statin therapy, it does not strictly predict efficacy of treatment. Perhaps an alternative statin would work better, but due to the limited amount of evidence it is still up to the prescriber to undergo a process of trial and error to determine the optimal treatment strategy.
As these brief examples illustrate, the clinical utility of PK biomarkers is much more straightforward since the routes of metabolism and transport of drugs are often well mapped. Simply adjusting dosage to compensate for the known effects or choosing a medication that utilizes an alternative pathway for metabolism and/or transport may avoid interactions.
PGx testing may in turn help a physician achieve optimal pharmacotherapy more efficiently and with less risk of ADRs.25 As physician knowledge expands, protocols may be developed that will enable PGx testing to become standard of care.
PCP Role in PGx
Getting Educated. To begin incorporating PGx testing into practice, the first step is education for physicians and the health care team. Medical schools in the United States and Canada are beginning to incorporate pharmacogenomics into their curriculum, but do not sufficiently prepare students to confidently utilize PGx in practice. A recent study determined that 82% of U.S. and Canadian medical schools incorporated pharmacogenomics into their curriculum, yet only 28% had more than 4 hours of didactic coursework on the subject and only 29% had plans to expand the curriculum within the next 3 years. Students feel that this is not adequate to prepare them; 57% considered pharmacogenomics instruction at their own school as "poor” or "not at all adequate” while 76% considered it "poor” or "not at all adequate” at most medical schools.26
There is more pharmacogenomics training available in schools of pharmacology. A study from 2010 determined that approximately 90% of schools included pharmacogenomics in their PharmD curricula compared to 39% as reported in 2005. Topic coverage was < 10 hours for 40.6%, 10-30 hours for 42.0%, and 31-60 hours for 14.5% of colleges and schools of pharmacy. Fewer than half were planning to increase course work over the next 3 years. 27
Although the need for ongoing education for future PCPs in pharmacogenomics is significant, PCPs already in practice must rely on resources outside of the classroom for information. Again, the need for suitable sources of professional education is largely unmet. While some physicians report learning of PGx testing and its clinical implications through organized CME such as professional meetings, grand rounds, and professional journals,2 some physicians also report less formal educational resources such as drug labeling information, communication with professional colleagues, and the Internet as primary sources of information.9
Implementation Considerations. Even with increased education, providers and clinical groups will have to carefully consider strategies for the assimilation of PGx into clinical practice. While specialty providers like oncologists may deal with PGx routinely in their practice, there is debate over how PGx testing should be implemented in the primary care setting.
The utilization of resources outside of the primary care setting may be beneficial in this regard. Clearly, pharmacology programs offer more education in the field, and pharmacists are well positioned to partner in the clinical integration of PGx by assisting with pharmaceutical management. Models of PGx testing and information delivery have been proposed that center around pharmacists and genetic counselors,28 but this may not meet the needs of patients who would prefer working with their PCPs in assimilating PGx testing results.5 Some PCPs feel that pharmacists should take primary responsibility for determining appropriate medication and dosing for patients dealing with significant PGx results. A collaborative approach, however, may be the most beneficial for the patient and serve to provide support for PCPs as they become more familiar with PGx testing and its implications in primary care. Previous studies have shown improved outcomes as the result of coordinated care between providers and pharmacists.29,30 This may suggest an ideally partnered approach to applied PGx.
PCPs may want to consider developing streamlined communication with a cooperative, multispecialty care team that includes pharmacists as an integral, rather than incidental, part of care delivery. Genetic counselors can be consulted where they are available. Emerging models of care, such as patient-centered medical homes (PCMHs) and accountable care organizations (ACOs), may be well suited to this kind of collaborative and personalized approach. PCPs can then ensure that testing done either through the primary care office or ordered by another provider is entered into the electronic health record (EHR) and that the patient is appropriately counseled about available testing, his/her individual results, and the implications those results may have for health and treatment.
Developing Protocols. Determining which patient populations to select as candidates for PGx testing will also be an important consideration. There are many proposed strategies with regard to this. Identifying risk due to factors like age, high-risk comorbid medical conditions and polypharmacy, or the presence of medications that are likely to be influenced by varying genotypes are all potential approaches. The association of both age and polypharmacy with ADRs is well established.4 Many medications have been proposed as high-priority for the potential application of PGx testing. Examples include statins, opioids, and anticoagulants. These medications are high priority because of the prevalence of use and their importance as a class in the management of common high-risk conditions. The known high frequency for ADRs, intolerance, and dependency are also important factors.
The targeting of individual patients on the basis of known risk factors may be more reactive than truly proactive. Trials are underway to develop and test broader approaches to PGx testing. These studies seek to identify potential genetic factors that may impact future care, well before known risk factors are present. Some of these models may have a significant impact on the use of PGx testing in primary care, as they take a truly preventive approach to the use of this technology.31
Identifying and utilizing patients’ individual genotyping for the purpose of collaborative communication with other members of the health care team will not be possible without consideration of how this information will be documented and shared. Consistent protocols will be of utmost importance in this effort. It would be preferable for genetic results to be integrated into the EHR as structured data. This would allow EHR systems to cross-reference other information in the medical record such as medication and problem lists. Tools within the EHR system, such as computerized decision support (CDS), would then be able to flag certain medications or disease states that might be affected by particular genotypes. Indeed, development and testing of such CDS support tools for PGx information is already being done.32 Furthermore, this structured data could then be shared with members of the health care team. If PGx information were to become standard of care, we may even begin to see it incorporated into the continuity of care document (CCD) standards that currently form part of the interoperability criteria for the CMS meaningful use incentive program.
Much of this could be accomplished by documenting testing through the use of ICD codes. These could be added to the patient record to reflect that genetic testing and counseling is being done, and then added to the problem list if genetic variants potentially effecting medication management are identified. In current ICD-9 nomenclature, v82.79 codes for "genetic screening NEC” and can be used during initial counseling and testing. This will transition to ICD-10 Z13.79, which is "encounter for other screening for genetic and chromosomal anomalies.” If a mutation or clinically significant variant is identified through testing, the ICD-9 code V83.89 "genetic carrier status” can be added, which will transition to Z14.8 "genetic carrier of other disease.” The implementation of ICD-10 may allow for more specific information to be directly coded into the patient record according to the needs of their care.
This ubiquity and ease of access to clinically significant PGx information may prevent patients from experiencing ADRs through the use of previously obtained genetic information. Also, as new information on pharmacologically significant genotypes becomes available, or if a patient develops a new condition that is affected by a known genetic factor, further testing could be performed and new information could be added into the patient’s individual record as needed.
Several years later GS developed atrial fibrillation requiring anticoagulation. Warfarin was chosen as the anticoagulant and the previously obtained PGx results, specifically CYP2C9 and VKORC1, were utilized to assist with dose optimization. Together these results indicated that a maintenance dose of approximately 3.5 mg daily may be sufficient to achieve an INR of 2.5. This helped both the prescriber and patient treat with more confidence.
While there is certainly debate about ordering pharmacogenetic testing specifically for dose optimization of warfarin, there is also considerable evidence that CYP2C9 and VKORC1 can help to reduce ADRs related to warfarin. This case demonstrates and supports the ongoing utility of PGx results in the medical record. A positive value may not be obtained with the use of PGx results for one drug (e.g., warfarin), but over the lifetime of treatment with multiple drugs, the continued use of these results can contribute to improved efficiency, efficacy, and cost effectiveness in applied pharmacotherapy.
Pitfalls. Open access to genetic information is directly opposed to the majority of patient opinion when it comes to how they want their genetic information recorded and used. Most patients currently want their genetic information very closely held and shared only with express consent.33 Current legislation in the form of the Genetic Information Nondisclosure Act reflects this concern, and limits how such information can be shared. It may be some time before PGx information may be freely used to benefit patient care.
The association between genetic variation and the potential for phenotypic expression in known disease states is another concern. Indeed, some PGx variations are also associated with increased risk for certain disease states, as illustrated in Table 5. This association presents an ideal opportunity to discuss the ethics of PGx testing with regard to informed consent. While the risks of many disease states as well the benefits of meeting treatment goals are well defined, the significance of some genetic variants are less clear. Informed consent becomes problematic when a patient is asked to make health care decisions weighing a known risk/benefit ratio against an unknown risk/benefit ratio. Patients may already be inclined to fear medication because of personal anxiety, somatization, anecdotal perception of personal or familial susceptibility to ADRs, and/or frightening reports through media sources and peer interactions regarding the risk of pharmaceuticals. Conflicting information might sway patients in favor of declining treatment for known high-risk conditions based on the theoretical genetic risk of ADR. This may result in an ethical violation of non-maleficence ("do no harm”).
While information regarding individual genotype should ideally provide patients with reassurance about which medications are safer and more efficacious for them, they have expressed reluctance to discuss information that did not result in clear and predictable advice.34 Developing a process for obtaining informed consent, preferably in writing, is highly advisable to optimally manage patient expectations, address potential concerns, and engage them in the use of PGx.
Patient engagement will be crucial as patients will need to participate actively in maintaining the most accurate information in their primary care setting as well as in other locations of care. Providing patients with a summary of their PGx results may be helpful if a patient is seeing providers not actively involved in cooperative health care teams. Surveys have indicated that patients may be willing to consider carrying results on a health alert card or some other device.33 This may provide a way to bypass patient concerns regarding the privacy of their genetic information, but relies on the patient to actively participate in their own care management.
A strong informed consent process would also provide an opportunity for PCPs to be clear on the specific testing options, costs, reimbursement, and procedures for testing in their area when designing protocols for PGx testing and counseling patients. The economics of health care raise significant concerns with regard to accurate and consistent use of genetic information in primary care. As a relatively new technology, PGx testing may not be covered by all insurance. This may create financial barriers to care. Of additional concern is the potential for increased cost of medications if it is discovered that a person’s genotype is not suitable for lower cost medications. It is not clear how responsive insurance carriers will be to authorizing preferred, but more costly, medications to patients based on genomic data.
The issue of insurance coverage raises a more direct concern in the management of PGx in primary care. Offices get communication throughout the course of the day from insurance carriers, pharmacies, and patients with requests for medication changes based on cost and formulary coverage. The PCP often manages these requests without adequate time for in-depth consideration or access to the full patient record. Physicians may be able to make the decision if one medication can be substituted for a lower cost medication of the same class, but they cannot be expected to know if the alternative medications will be suitable for the patient’s genotype, unless the data are available and they are educated as to how to use it. Furthermore, changes to medications made outside the office visit, either by primary care or by other providers, are not always consistently noted in the patient chart so that the new medications become part of their health record and can be considered in context of the patient’s genotype. This may result in patients being asked to return to the office to discuss any changes to their medication, which could in turn erode any potential savings to the health care system from reduced ADRs.
Increased knowledge in the area of genomic sciences is likely to have an expanding impact on medical care in the near future. PGx is already heavily influencing medication management in primary care, but much of this information is presented to physicians without an organized and proactive framework for applying it.
It is critical that physicians, particularly PCPs, seek to assume a leadership role in the implementation of genomic science in patient care. Ongoing education will help them use this information to manage medications and high-risk conditions optimally. The development of coordinated care models like ACOs and PCMH will also leverage the skill sets of a broader segment of the health care team and greatly benefit the implementation of this technology.
Despite significant barriers and pitfalls, creating solid protocols for implementation will yield optimal results for physicians and patients as genetic information takes a more prominent place in patient care. PCPs must become part of the development of these protocols for this technology to yield the greatest potential benefits.
This article is dedicated to Dr. Brian Robert Hocum, MD (June 18, 1949–July 27, 2014).
- Bhindi R, Ormerod O, Newton J, et al. Interaction between statins and clopidogrel: Is there anything clinically relevant? QJM 2008;101:915-925.
- Haga SB, Burke W, Ginsburg GS, et al. Primary care physicians’ knowledge of and experience with pharmacogenetic testing. Clin Genet 2012;82:388-394.
- National Ambulatory Medical Care Survey: 2010 Summary Tables. 2010. Available at: http://www.cdc.gov/nchs/data/ahcd/namcs_summary/2010_namcs_web_tables.pdf. Accessed July 3, 2014.
- Sarkar U, Lopez A, Maselli JH, Gonzales R. Adverse drug events in U.S. adult ambulatory medical care. Health Serv Res 2011;46:1517-1533.
- Payne K, Fargher EA, Roberts SA, et al. Valuing pharmacogenetic testing services: A comparison of patients’ and health care professionals’ preferences. Value Health 2011;14:121-134.
- Budnitz DS, Lovegrove MC, Shehab N, Richards CL. Emergency hospitalizations for adverse drug events in older Americans. N Engl J Med 2011;365:2002-2012.
- The Network for Excellence in Health Innovation. A System-wide Approach to Improving Patient Medication Adherence for Chronic Disease. Thinking Outside the Pillbox 2009. Available at: http://www.nehi.net/publications/17-thinking-outside-the-pillbox-a-system-wide-approach-to-improving-patient-medication-adherence-for-chronic-disease/view. Accessed July 17, 2014.
- International drug monitoring: The role of national centres. Report of a WHO meeting. World Health Organ Tech Rep Ser 1972;498:1-25.
- Stanek EJ, Sanders CL, Taber KA, et al. Adoption of pharmacogenomic testing by US physicians: Results of a nationwide survey. Clin Pharmacol Ther 2012;91:450-458.
- U.S. Food and Drug Administration. Table of Pharmacogenomic Biomarkers in Drug Labeling. 2014. Available at: http://www.fda.gov/drugs/scienceresearch/researchareas/pharmacogenetics/ucm083378.htm. Accessed July 3, 2014.
- Pirmohamed M. Personalized pharmacogenomics: Predicting efficacy and adverse drug reactions. Annu Rev Genomics Hum Genet 2014;15:349-370.
- Wynn G, Oesterheld J, Cozza K, Armstrong S. Clinical Manual of Drug Interaction Principles for Medical Practice. Arlington, VA: American Psychiatric Publishing, Inc.; 2009.
- Dietrich CG, Geier A, Oude Elferink RP. ABC of oral bioavailability: Transporters as gatekeepers in the gut. Gut 2003;52:1788-1795.
- Ambudkar SV, Dey S, Hrycyna CA, et al. Biochemical, cellular, and pharmacological aspects of the multidrug transporter. Ann Rev Pharmacol Toxicol 1999;39:361-398.
- Kitzmiller JP, Binkley PF, Pandey SR, et al. Statin pharmacogenomics: Pursuing biomarkers for predicting clinical outcomes. Discov Med 2013;16:45-51.
- Leckband SG, Kelsoe JR, Dunnenberger HM, et al. Clinical Pharmacogenetics Implementation Consortium guidelines for HLA-B genotype and carbamazepine dosing. Clin Pharmacol Ther 2013;94:324-328.
- Carbamazepine [package insert]. Caraco Pharmaceutical Laboratories, LTDl, San Diego, CA; January 2008. A http://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=c13bc0b8-7900-4ef4-98ed-e1315a08d95d. Accessed July 5, 2014.
- Mega JL, Morrow DA, Brown A, et al. Identification of genetic variants associated with response to statin therapy. Arterioscler Thromb Vasc Biol 2009;29:1310-1315.
- National Center for Health Statistics. Health, United States, 2013: With Special Feature on Prescription Drugs. Hyattsville, MD. 2014.
- 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;6:400-408.
- Verschuren JJ, Trompet S, Wessels JA, et al. A systematic review on pharmacogenetics in cardiovascular disease: Is it ready for clinical application? Eur Heart J 2012;33:165-175.
- Cuciureanu M, Vlase L, Muntean D, et al. Grapefruit juice--drug interactions: Importance for pharmacotherapy. Revi Med Chir Soc Med Nat Iasi 2010;114:885-891.
- Lilja JJ, Kivisto KT, Neuvonen PJ. Grapefruit juice increases serum concentrations of atorvastatin and has no effect on pravastatin. Clin Pharmacol Ther 1999;66:118-127.
- Chung JY, Cho SK, Oh ES, et al. Effect of HMGCR variant alleles on low-density lipoprotein cholesterol-lowering response to atorvastatin in healthy Korean subjects. J Clin Pharmacol 2012;52:339-346.
- Zhou SF. Polymorphism of human cytochrome P450 2D6 and its clinical significance: Part II. Clin Pharmacokinet 2009;48:761-804.
- Green JS, O’Brien TJ, Chiappinelli VA, Harralson AF. Pharmacogenomics instruction in US and Canadian medical schools: Implications for personalized medicine. Pharmacogenomics 2010;11:1331-1340.
- Murphy JE, Green JS, Adams LA, et al. Pharmacogenomics in the curricula of colleges and schools of pharmacy in the United States. Am J Pharm Educ 2010;74:7.
- Mills R, Haga SB. Clinical delivery of pharmacogenetic testing services: A proposed partnership between genetic counselors and pharmacists. Pharmacogenomics 2013;14:957-968.
- Padiyara RS, D’Souza JJ, Rihani RS. Clinical pharmacist intervention and the proportion of diabetes patients attaining prevention objectives in a multispecialty medical group. J Manag Care Pharm 2011;17:456-462.
- Pape GA, Hunt JS, Butler KL, et al. Team-based care approach to cholesterol management in diabetes mellitus: Two-year cluster randomized controlled trial. Arch Intern Med 2011;171:1480-1486.
- Dolgin E. Preemptive genotyping trialed to prevent adverse drug reactions. Nat Med 2011;17:1323.
- Bielinski SJ, Olson JE, Pathak J, et al. Preemptive genotyping for personalized medicine: Design of the right drug, right dose, right time-using genomic data to individualize treatment protocol. Mayo Clin Proc 2014;89:25-33.
- Haga SB, Kawamoto K, Agans R, Ginsburg GS. Consideration of patient preferences and challenges in storage and access of pharmacogenetic test results. Genet Med 2011;13:887-890.
- Barash CI. Ethical Issues in Pharmacogenetics. 2013. Available at: http://www.actionbioscience.org/biotechnology/ethical_issues_in_pharmacogenetics.html. Accessed July 3, 2014.
- Ravyn D, Ravyn V, Lowney R, Nasrallah HA. CYP450 pharmacogenetic treatment strategies for antipsychotics: A review of the evidence. Schizophr Res 2013;149:1-14.
- Strattera(R) [package insert]. Eli Lilly and Company, Indianapolis, IN; February 2014. http://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=309de576-c318-404a-bc15-660c2b1876fb Accessed July 5, 2014.
- Rau T, Wuttke H, Michels LM, et al. Impact of the CYP2D6 genotype on the clinical effects of metoprolol: A prospective longitudinal study. Clin Pharmacol Ther 2009;85:269-272.
- Kirchheiner J, Nickchen K, Bauer M, et al. Pharmacogenetics of antidepressants and antipsychotics: The contribution of allelic variations to the phenotype of drug response. Mol Psychiatry 2004;9:442-473.
- Swen JJ, Nijenhuis M, de Boer A, et al. Pharmacogenetics: From bench to byte—an update of guidelines. Clin Pharmacol Ther 2011;89:662-673.
- Lundblad MS, Ohlsson S, Johansson P, et al. Accumulation of celecoxib with a 7-fold higher drug exposure in individuals homozygous for CYP2C9*3. Clin Pharmacol Ther 2006;79:287-288.
- Celebrex(R) [package insert]. G.D. Searle LLC Division of Pfizer Inc, NY, NY; July 2014. Available at: http://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=8d52185d-421f-4e34-8db7-f7676db2a226. Accessed July 5, 2014.
- Brandt JT, Close SL, Iturria SJ, et al. Common polymorphisms of CYP2C19 and CYP2C9 affect the pharmacokinetic and pharmacodynamic response to clopidogrel but not prasugrel. J Thromb Haemost 2007;5:2429-2436.
- 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;94:317-323.
- Plavix(R) [package insert]. Bristol-Myers Squibb/Sanofi Pharmaceuticals Partnership B, NJ; December 2013. Available at: http://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=01b14603-8f29-4fa3-8d7e-9d523f802e0b. Accessed July 5, 2014.
- Celexa(R) [package insert]. Forest Pharmaceuticals, Inc., St. Louis, MO; October 2013. Available at: http://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=4259d9b1-de34-43a4-85a8-41dd214e9177. Accessed July 5, 2014.
- Baldwin RM, Ohlsson S, Pedersen RS, et al. Increased omeprazole metabolism in carriers of the CYP2C19*17 allele; a pharmacokinetic study in healthy volunteers. Br J Clin Pharmacol 2008;65:767-774.
- Bae JW, Choi CI, Lee JH, et al. Effects of UDP-glucuronosyltransferase polymorphisms on the pharmacokinetics of ezetimibe in healthy subjects. Eur J Clin Pharmacol 2011;67:39-45.
- Chung JY, Cho JY, Yu KS, et al. Effect of the UGT2B15 genotype on the pharmacokinetics, pharmacodynamics, and drug interactions of intravenous lorazepam in healthy volunteers. Clin Pharmacol Ther 2005;77:486-494.
- Kim KA, Park PW, Park JY. Effect of ABCB1 (MDR1) haplotypes derived from G2677T/C3435T on the pharmacokinetics of amlodipine in healthy subjects. Br J Clin Pharmacol 2007;63:53-58.
- Pasanen MK, Fredrikson H, Neuvonen PJ, Niemi M. Different effects of SLCO1B1 polymorphism on the pharmacokinetics of atorvastatin and rosuvastatin. Clin Pharmacol Ther 2007;82:726-733.
- Ulvestad M, Skottheim IB, Jakobsen GS, et al. Impact of OATP1B1, MDR1, and CYP3A4 expression in liver and intestine on interpatient pharmacokinetic variability of atorvastatin in obese subjects. Clin Pharmacol Ther 2013;93:275-282.
- Kalliokoski A, Backman JT, Kurkinen KJ, et al. Effects of gemfibrozil and atorvastatin on the pharmacokinetics of repaglinide in relation to SLCO1B1 polymorphism. Clin Pharmacol Ther 2008;84:488-496.
- Niemi M. Transporter pharmacogenetics and statin toxicity. Clin Pharmacol Ther 2010;87:130-133.
- Yuan H, Huang ZJ, Liu JJ, et al. Influence on metoprolol antihypertensive effect of ß1-adrenoreceptor gene polymorphism and methylated modification. Cell Biol Int 2008;32:S7.
- Finkelstein Y, Bournissen FG, Hutson JR, Shannon M. Polymorphism of the ADRB2 gene and response to inhaled beta-agonists in children with asthma: A meta-analysis. J Asthma 2009;46:900-905.
- Blom JW, Doggen CJ, Osanto S, Rosendaal FR. Old and new risk factors for upper extremity deep venous thrombosis. J Thromb Haemost 2005;3:2471-2478.
- Simone B, De Stefano V, Leoncini E, et al. Risk of venous thromboembolism associated with single and combined effects of Factor V Leiden, Prothrombin 20210A and Methylenetethraydrofolate reductase C677T: A meta-analysis involving over 11,000 cases and 21,000 controls. Eur J Epidemiol 2013;28:621-647.
- Wu O, Robertson L, Twaddle S, et al. Screening for thrombophilia in high-risk situations: Systematic review and cost-effectiveness analysis. The Thrombosis: Risk and Economic Assessment of Thrombophilia Screening (TREATS) study. Health Technol Assess 2006;10:1-110.
- Varga EA, Kujovich JL. Management of inherited thrombophilia: Guide for genetics professionals. Clin Genet 2012;81:7-17.
- Zwisler ST, Enggaard TP, Noehr-Jensen L, et al. The antinociceptive effect and adverse drug reactions of oxycodone in human experimental pain in relation to genetic variations in the OPRM1 and ABCB1 genes. Fundam Clin Pharmacol 2010;24:517-524.
- Lewis JP, Ryan K, O’Connell JR, et al. Genetic variation in PEAR1 is associated with platelet aggregation and cardiovascular outcomes. Circ Cardiovasc Genet 2013;6:184-192.
- Hajj A, Ksouda K, Peoc’h K, et al. KCNH2 polymorphism and methadone dosage interact to enhance QT duration. Drug Alcohol Depend 2014;141:34-38.
- Porcelli S, Fabbri C, Serretti A. Meta-analysis of serotonin transporter gene promoter polymorphism (5-HTTLPR) association with antidepressant efficacy. Eur Neuropsychopharmacol 2012;22:239-258.
- Hamberg AK, Wadelius M, Lindh JD, et al. A pharmacometric model describing the relationship between warfarin dose and INR response with respect to variations in CYP2C9, VKORC1, and age. Clin Pharmacol Ther 2010;87:727-734.
- Wang D, Guo Y, Wrighton SA, et al. Intronic polymorphism in CYP3A4 affects hepatic expression and response to statin drugs. Pharmacogenomics J 2011;11:274-286.
- Fiegenbaum M, da Silveira FR, Van der Sand CR, et al. The role of common variants of ABCB1, CYP3A4, and CYP3A5 genes in lipid-lowering efficacy and safety of simvastatin treatment. Clin Pharmacol Ther 2005;78:551-558.
- Stewart A. SLCO1B1 Polymorphisms and Statin-Induced Myopathy. PLoS Curr 2013;5.
- Link E, Parish S, Armitage J, et al. SLCO1B1 variants and statin-induced myopathy—a genomewide study. N Engl J Med 2008;359:789-799.
- Gao J, Cong HL, Mao YM, et al. [Effects of genetic variations of cholesteryl ester transfer protein on atorvastatin treatment efficacy and clinical outcomes in patients with coronary artery disease]. Zhonghua Yi Xue Za Zhi 2013;93:2195-2199.
- Gelissen IC, McLachlan AJ. The pharmacogenomics of statins. Pharmacol Res 2014:88C:99-106.
- Jiang XY, Zhang Q, Chen P, et al. CYP7A1 polymorphism influences the LDL cholesterol-lowering response to atorvastatin. J Clin Pharm Ther 2012;37:719-723.
- Iakoubova OA, Sabatine MS, Rowland CM, et al. Polymorphism in KIF6 gene and benefit from statins after acute coronary syndromes: Results from the PROVE IT-TIMI 22 study. J Am Coll Cardiol 2008;51:449-455.
- Oh J, Ban MR, Miskie BA, et al. Genetic determinants of statin intolerance. Lipids Health Dis 2007;6:7.
- Shitara Y, Sugiyama Y. Pharmacokinetic and pharmacodynamic alterations of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors: Drug-drug interactions and interindividual differences in transporter and metabolic enzyme functions. Pharmacol Ther 2006;112:71-105.
- Wilke RA, Ramsey LB, Johnson SG, et al. The clinical pharmacogenomics implementation consortium: CPIC guideline for SLCO1B1 and simvastatin-induced myopathy. Clin Pharmacol Ther 2012;92:112-117.
- Haerian BS, Haerian MS. OPRM1 rs1799971 polymorphism and opioid dependence: Evidence from a meta-analysis. Pharmacogenomics 2013;14:813-824.
- Linares OA, Daly D, Stefanovski D, Boston RC. The CYP2D6 gene determines oxycodone’s phenotype-specific addictive potential: Implications for addiction prevention and treatment. Med Hypotheses 2014;82:390-394.
- Lu Y, Qin X, Li S, et al. Quantitative assessment of CYP2D6 polymorphisms and risk of Alzheimer’s disease: A meta-analysis. J Neurol Sci 2014;343:15-22.
- Zhou LP, Yao F, Luan H, et al. CYP3A4*1B polymorphism and cancer risk: A HuGE review and meta-analysis. Tumour Biol 2013;34:649-660.
- Wu H, Kang H, Liu Y, et al. Roles of ABCB1 gene polymorphisms and haplotype in susceptibility to breast carcinoma risk and clinical outcomes. J Cancer Res Clin Oncol 2012;138:1449-1462.
- Painter JN, Nyholt DR, Krause L, et al. Common variants in the CYP2C19 gene are associated with susceptibility to endometriosis. Fertil Steril 2014;102:496-502.
- Ulman CA, Trevino JJ, Miller M, Gandhi RK. Fish odor syndrome: A case report of trimethylaminuria. Dermatol Online J 2014;20:21260.
- Genetics Home Reference: Your Guide to Understanding Genetic Conditions. UGT1A1 2014. Available at: http://ghr.nlm.nih.gov/gene/UGT1A1. Accessed July 7, 2014.
- Shahabi P, Siest G, Meyer UA, Visvikis-Siest S. Human cytochrome P450 epoxygenases: Variability in expression and role in inflammation-related disorders. Pharmacol Ther 2014; doi: 10.1016/j.pharmthera.2014.05.011. [Epub ahead of print].
- Kuang ZM, Huang ZJ, Li Y, et al. Revealing the contribution of Cytochrome P450 to salt-sensitive hypertension using DNA microarray. Eur Rev Med Pharmacol Sci 2013;17:3148-3156.
- Tiwari AK, Deshpande SN, Rao AR, et al. Genetic susceptibility to tardive dyskinesia in chronic schizophrenia subjects: I. Association of CYP1A2 gene polymorphism. Pharmacogenomics J 2005;5:60-69.
- Liu Q, Ramsey TL, Meltzer HY, et al. Sulfotransferase 4A1 haplotype 1 (SULT4A1-1) is associated with decreased hospitalization events in antipsychotic-treated patients with schizophrenia. Prim Care Companion CNS Dis 2012; doi: 10.4088/PCC.11m01293. Epub 2012 Jun 24.
- Ramsey TL, Meltzer HY, Brock GN, et al. Evidence for a SULT4A1 haplotype correlating with baseline psychopathology and atypical antipsychotic response. Pharmacogenomics 2011;12:471-480.