Abstract
Many practitioners who have not had pharmacogenomic education are required to apply pharmacogenomics to their practices. Although many aspects of pharmacogenomics are similar to traditional concepts of drug-drug interactions, there are some differences. We searched PubMed with the search terms pharmacogenomics and pharmacogenetics (January 1, 2005, through December 31, 2019) and selected articles that supported the application of pharmacogenomics to practice. For inclusion, we gave preference to national and international consortium guidelines for implementation of pharmacogenomics. We discuss special considerations important in the application of pharmacogenomics to assist clinicians with ordering, interpreting, and applying pharmacogenomics in their practices.
Abbreviations and Acronyms:
CDS (clinical decision support), CPIC (Clinical Pharmacogenetics Implementation Consortium), CYP (cytochrome P450), DPWG (Royal Dutch Association for the Advancement of Pharmacy-Pharmacogenetics Working Group), EHR (electronic health record), FDA (Food and Drug Administration), INR (international normalized ratio)Pharmacogenomic practice information has expanded in the literature during the past several years. We searched PubMed with the search terms pharmacogenomics and pharmacogenetics (January 1, 2005, through December 31, 2019) and selected articles that supported the application of pharmacogenomics to practice. For inclusion, we gave preference to national and international consortium guidelines for implementation of pharmacogenomics. Although there are numerous scientific publications, application of this science remains relatively new to most medical practitioners. Many health care providers have not had pharmacogenomic education yet are increasingly expected to apply this expanding discipline to their practices. Although approximately 87% of medical nursing, pharmacy, and health science schools’ didactic programs have added pharmacogenomics to their curricula according to a recent survey, this educational gap has not been entirely closed by recent education of new practitioners.
1
,2
In addition, the amount and quality of education may have limitations. In the past, of medical students in the United States and Canada who have had pharmacogenomic education, 57% considered pharmacogenomic instruction to be “poor” or “not at all adequate” at their medical school and 76% considered it “poor” or “not at all adequate” at most medical schools.3
Evaluations have demonstrated that although pharmacogenomic information is important to practitioners, many do not know how to implement it in their practices.4
,5
Many practicing pharmacists also have had limited pharmacogenomic education in their professional programs. Although pharmacy professional societies and schools support pharmacogenomic curricula development,
6
especially for new students,7
there are a limited number of postgraduate specialty pharmacy residencies,8
fellowships, and clinical pharmacology training grants9
for advanced pharmacogenomic education available nationally. Past evaluations of practicing pharmacists also demonstrate the overall importance of implementation into practice. In an assessment of pharmacists’ educational needs, 85% of respondents agreed that pharmacists should be required to be knowledgeable about pharmacogenomics. In contrast, 63% of respondents felt that they could not accurately apply the results of pharmacogenomic tests to drug therapy selection, dosing, or monitoring.10
Several organizations of experts have assembled to assist with the integration of pharmacogenomics into practice, including the Clinical Pharmacogenetics Implementation Consortium (CPIC),
11
the Royal Dutch Association for the Advancement of Pharmacy-Pharmacogenetics Working Group (DPWG),12
the Canadian Pharmacogenomics Network for Drug Safety,13
and expert practice groups in medical societies. Pharmacogenomic dosing guidelines for implementation from these groups can be found on the PharmGKB website (https://www.pharmgkb.org/).14
Although the provided recommendations are extremely helpful, these often address single gene-drug pairs (Table) or a single pharmacologic issue at a time. Because drug therapy is often complex, with many patients taking several medications concurrently, considering each drug problem individually can be challenging and time-consuming for the busy practitioner, especially outside his or her area of practice.TableExamples of Drug-Gene Pairs
| Drug | Associated gene |
|---|---|
| Abacavir | HLA-B∗57:01 |
| Allopurinol | HLA-B∗58:01 |
| Carbamazepine | HLA-B∗15:02 and HLA-A∗31:01 |
| Citalopram | CYP2C19 |
| Clopidogrel | CYP2C19 |
| Codeine | CYP2D6 |
| Escitalopram | CYP2C19 |
| Fluoropyrimidines | DPYD |
| Fluoxetine | CYP2D6 |
| Fluvoxamine | CYP2D6 |
| Paroxetine | CYP2D6 |
| Simvastatin | SLCO1B1 |
| Tacrolimus | CYP3A5 |
| Tamoxifen | CYP2D6 |
| Tramadol | CYP2D6 |
| Thiopurines | TPMT and NUDT15 |
| Venlafaxine | CYP2D6 |
| Warfarin | CYP2C9 and VKORC1 |
To compensate for this, a growing number of pharmacogenomic testing laboratories are providing panel tests with reports that may include a few to many medications, often grouped by drug class or medical specialty. The practitioner must make the final treatment decision, and if the laboratory combines results from several genes, recommendations may not always match the existing national or international pharmacogenomic guidelines. Furthermore, these reports typically do not incorporate drug-drug interactions into medication recommendations. Difficulties with ordering tests, interpreting tests, and applying pharmacogenomics to complex clinical practice continue to be a challenge.
The primary care provider working in the clinic might be asked to order, interpret, or apply a pharmacogenomic test, typically without the assistance of specialty resources. To aid practitioners in implementation of pharmacogenomics, we identified some common questions and considerations for the application of pharmacogenomics to assist clinicians in working with these new concepts and technologies. Providers should note that many of the skills required in pharmacogenomics are similar to pharmacologic considerations currently used in daily clinical practice, and we highlight differences in this article.
Ordering Tests
Should I Order a Pharmacogenomic Test?
Routine pharmacogenomic testing for many medications is not required and may not be useful. In addition, many of these tests are not covered by many insurance plans. This usually requires the provider to determine coverage on a case-by-case basis. Insurers may take into consideration the cost of the test, preferred testing providers, the clinical indication provided, the patient's clinical history, or other variables. To facilitate insurance coverage determination, providers might encourage patients to contact their insurance company to inquire about test coverage benefits and any precondition criteria that must be met.
However, in some cases a pharmacogenomic test can predict an increased risk of severe drug toxicity or drug hypersensitivity. For example, individuals who are positive for the HLA-B∗58:01 allele have an increased risk of Stevens-Johnson syndrome and toxic epidermal necrolysis when treated with allopurinol.
15
These devastating dermatologic hypersensitivity reactions can result in significant morbidity and mortality.16
In severe cases, patients often require management in an intensive care unit or critical burn unit.17
Other examples of genes that can predict hypersensitivity reactions include HLA-B∗57:01 with abacavir18
and HLA-B∗15:02/HLA-A∗31:01 with carbamazepine.19
Chemotherapeutic toxicities secondary to genetic variants in metabolism can be severe and life-threatening. Therefore, consideration for testing should be given to DPYD for fluoropyrimidines,20
TPMT and NUDT15 for thiopurines,21
and UGT1A1 for irinotecan.22
Additional situations in which pharmacogenomic tests may be particularly useful are unexplained sensitivities or toxicities with multiple drugs not otherwise explained by drug interactions.
23
Sometimes metabolic patterns can be initially detected by considering the metabolism of drugs and the problems (eg, efficacy/adverse effects) that the patient has had in the past. For example, many drugs, including tramadol and metoprolol, are metabolized by cytochrome P450 (CYP) 2D6.24
A patient who describes inefficacy of tramadol in the past for pain and currently is taking a very low dose of metoprolol for blood pressure management or heart rate control has a drug response profile most consistent with that of a poor metabolizer of CYP2D6.25
In this case, the patient’s inability to activate the prodrug tramadol for analgesic effect26
or to appropriately metabolize metoprolol may be due to the presence of genetic variation in CYP2D6, which could be detected by testing.27
Pharmacogenomic testing may assist providers in sorting out complicated pharmacologic problems in individuals taking several medications but also requires providers to possess, or know where to access, knowledge of what drugs are substrates for individual CYPs or transporters.In addition to avoidance of severe drug toxicities and to explain a patient’s previous medication-related problems, pharmacogenomics may also be considered for patients who have been noncompliant or have self-adjusted their medication regimen because there may be an underlying problem that could be identified by pharmacogenomics. Finally, preemptive testing, where the testing is performed with the intention of guiding future prescription, presents an advantage over reactive testing in that genetic information is available at the point of care to guide decision making in real time, as opposed to potentially delaying drug therapy decisions with reactive pharmacogenetic testing.
28
Preemptive testing may be considered for patients who request testing, when a family history indicates potential medication-related problems, or as a routine part of care in some health care systems.Should I Order a Single Gene Test or a Panel?
If testing is being ordered for a specific indication, testing options can be categorized as either a single gene or a panel test, each with their own considerations.
29
Depending on the details of the specific laboratory test, a single gene test may provide greater precision for a particular gene than when that same gene is tested as part of a multigene panel. For example, a test for a single gene may be performed by sequencing (such that all variants in the coding region of the gene could be detected) rather than by genotyping (where only a few predefined variants can be detected); alternatively, one pharmacogenomic testing laboratory may offer a targeted genotyping-based assay of the gene alone and test for more variants/alleles than another laboratory that includes the desired gene in a panel, potentially resulting in increased test sensitivity. In other cases, the test may be identical for that gene whether ordered alone or as part of a panel. Currently, there is significant variability among laboratories in which alleles are included in a test; however, efforts to define alleles that should be included at a minimum in clinical tests are underway and will be discussed later herein.30
,31
If a single gene test is selected and being ordered for a very specific purpose, there may be a greater chance that its cost will be covered by insurance rather than a nonspecific panel including the desired gene.32
Conversely, a panel test provides a comprehensive view of multiple genes. Information on multiple genes may be more beneficial when trying to use pharmacogenomics to sort out medication issues in a patient taking multiple drugs. Panel testing is also the most appropriate option in the preemptive setting because one cannot predict with certainty which medications a patient may require in the future.
28
In addition, if a panel test is performed when a single gene is required to answer the clinical question, the results can be used again for future clinical questions—in a sense the reactive test becomes a preemptive test in this scenario. Another advantage of panel testing is that testing for many genes is often multiplexed such that the cost may be similar to that of a single-gene test, or at least less than ordering several single genes separately. However, it is important to recognize that these panels are laboratory specific in which genes and alleles are included, with no standardization among different laboratories. Therefore, the provider must ensure that the desired gene(s) for the specific medication is part of the panel before ordering the test. When considering panel testing, providers should recognize that some laboratories use proprietary combinatorial algorithms to generate medication recommendations. Although these reports may bin medications into a red/yellow/green system that is easy to use, the rationale for how a medication was placed into a particular bin may be unknown to the ordering provider. Other laboratories rely on publicly available data in the Food and Drug Administration (FDA) (or other international) drug label, CPIC guidelines, and literature references to a single or a few genes at a time. These differences in interpretation strategies may lead to notable disagreement in medication recommendations among panels.33
Interpreting Tests
What Is the Quality of the Laboratory?
Laboratories in the United States that perform pharmacogenomic testing for clinical use must hold Clinical Laboratory Improvement Amendments certification and should be accredited by an agency such as the College of American Pathologists. Although direct-to-consumer services are available and patients may present with pharmacogenomic results from these tests, their incomplete test designs and their false-positive and false-negative rates suggest that results should be confirmed in a clinical laboratory before the clinical use of results for patient care.
34
Although there has been some controversy in the application of pharmacogenomics in clinical practice,35
the FDA currently warns that direct-to-consumer testing should not be used for making medical decisions, with the exception of FDA’s recent 510(k) clearance of 23andMe’s CYP2C19 pharmacogenetic test that no longer requires confirmatory testing.36
,37
Despite national and international efforts to standardize pharmacogenomic testing,
30
,38
,39
variability in testing and interpretation currently exist. This makes pharmacogenomic results different than many of the standard laboratory tests that a clinician will commonly use in his or her daily practice (eg, electrolytes, international normalized ratio [INR], liver function tests), where results are usually provided in a standard format, interpretations are consistent, and a reference range is provided to allow for comparisons among laboratories. Comparing pharmacogenomic results obtained from different laboratories may be challenging, and an understanding of the limitations of the test results in hand is essential for proper use.What Is the Precision of the Test?
Before even considering a specific test or the differences among tests, it is essential that the provider understand that pharmacogenomic tests are predictive in nature. Genes are often included in test panels either because they are known to encode a gene that metabolizes a medication
40
or because they were identified through a population-level association study.41
In some cases, functional genomic studies have been performed to determine the impact of specific genetic variants on enzyme function42
; however, other variables may also affect metabolism in vivo. In other cases, testing is based solely on association studies, and the mechanism underlying the association may or may not be well-established.43
In that case, some individuals will have the variant and not have the associated expected drug response. Furthermore, depending on whether the associated variant(s) is functionally significant, some associations may be population-specific.44
Putting the predictive nature of pharmacogenomic testing aside, if an individual had testing performed at more than 1 laboratory, the results may not be identical.33
,45
One reason that laboratories may report a different result is that the genetic alleles/variants included in the design of the pharmacogenomic test may differ.38
In today’s commonly used genotyping-based tests, if a genetic variant is not included in the test design, it cannot be detected,38
which may lead to false-negative results. For example, the CYP2C9∗5, ∗6, ∗8, and ∗11 alleles result in reduced function and are more common in individuals of African or African American descent. However, these alleles are often not included in clinical tests.29
Therefore, if an individual has a CYP2C9∗1/∗8 genotype, a laboratory that can detect this allele will report the correct genotype and the corresponding intermediate metabolizer phenotype; however, a laboratory that does not test/detect the ∗8 allele would report a CYP2C9∗1/∗1 genotype and corresponding normal metabolizer phenotype.29
A different therapeutic decision could be made based on the result. Warfarin is metabolized by CYP2C9, and if the patient has reduced metabolism, a dose reduction on initiation might be warranted. However, the practitioner may incorrectly initiate a higher-than-recommended dose if the pharmacogenomic results are reported as a normal CYP2C9 metabolizer. Therefore, the patient may be more likely to have an initial INR above the therapeutic range. The Association for Molecular Pathologists has initiated a project to recommend a minimum set of alleles that should be included in laboratory tests. Thus far, they have published recommendations for CYP2C19 and CYP2C9, with plans to create recommendations for additional genes. Tests may become more standardized over time as laboratories begin to adopt these recommendations.30
,31
Along with the lack of standardization of genes/alleles included on a pharmacogenomic panel, the phenotypic interpretation of the same genotype may differ among reporting laboratories. For example, due to remaining controversy in the field, CYP2D6∗1/∗4 is reported by some laboratories as an intermediate metabolizer, and other laboratories report this genotype as a normal metabolizer.
29
These differences are important for some drugs. The clinical recommendations for codeine do not change for these 2 phenotypes.46
However, differences in these phenotypes do change the clinical recommendations for tamoxifen.47
Because most clinical practitioners outside the laboratory would not know the nuances behind these genotype translations, it is important to understand how the laboratory is reporting the results and for which medications the results are intended to be used. Consensus recommendations for allele function from PharmVar48
and for genotype-to-phenotype translation from the CPIC and the DPWG were recently published, which may begin to resolve this controversy49
; however, until then, understanding these reporting differences may explain discrepancies in reports from different sources.In addition, the genetic interpretation from the same laboratory may change over time as a result of reclassification of an allele based on newer and more complete literature, or due to updates to test design, including additional alleles that were not part of an earlier assay design. Overall, when it comes to pharmacogenomic testing, the clinician needs to keep in mind that although the patient’s genetic makeup generally stays constant (in the absence of transplantation), their reported pharmacogenomic results may differ over time due to differences in test precision or updated interpretive information.
50
When questions arise, a phone call to the laboratory can be helpful in gaining a better understanding of the results.How Will the Results Be Reported?
A typical pharmacogenomic report will include a genotype and a predicted phenotype for each gene tested, but the format varies. Single-gene tests tend to be simple, focusing on the genotype (eg, CYP2C9∗3/∗3) and the phenotype (eg, poor metabolizer), and may or may not provide recommendations for specific medications. Depending on the laboratory and specific gene, some of these interpretations can be quite detailed, but often, the phenotype is enough information to assist the clinician familiar with the drug’s metabolism or action.
Multigene panels may include recommendations for several to many medications, may vary in the level of detail provided, and may or may not be based on proprietary algorithms that combine results for multiple genes to predict medication response. In contrast to more general gene-specific interpretations, medication-specific advice can be quite helpful to clinicians before ordering a drug because they might be very familiar with the medication they wish to prescribe but less so with its metabolic pathway. These reports with information on many medications and drug-gene interactions are designed to warn the clinician of a potential problem with a medication and to aid in identification of alternate options that may be more appropriate based on the patient’s genetic results.
29
Reports may also contain medications that are not known to be affected by pharmacogenomics, which can be helpful in drug selection. One consideration for reports that categorize medications (eg, major vs moderate drug-gene interaction; use with great vs moderate caution; red/yellow/green binning)51
,52
is that the criteria for how a medication is categorized in the report is not always readily apparent.29
,53
In addition, as mentioned previously herein, a proprietary algorithm may have been applied. Furthermore, it may not be clear whether the drug must be avoided or whether a dose adjustment may be reasonable for that individual. In addition, if a drug is identified as a potential problem for the patient, an alternative drug may not be provided in a similar class for the desired indication. In the absence of clear recommendations, the clinician needs to make the adjustment with their best judgment. In this situation, returning to the basics of pharmacology and considering a drug’s therapeutic index and potential toxicity is important in making a wise clinical decision.In addition to general-purpose multigene panels, panels that are tailored for a specific practice or indication (eg, psychiatry panel) are also available. These reports are often divided in sections for drug recommendations or cautions more specific to the practitioner (eg, antidepressants and antipsychotics). Similar to general panels, recommendations may or may not be based on proprietary algorithms that combine weighted scores of different pharmacogenomic genes/variants together.
53
In addition to the drug recommendations provided in these reports, genotype and phenotype information is also provided, which may be applied by more advanced users to medications not included in the report. Also, the report will include information on the methods used and potential limitations. Depending on the laboratory, sometimes the phenotypes are reported in ranges that may span categories (ie, poor to normal metabolizer). If a patient has a phenotype that falls between the broad phenotypic categories, applying the CPIC guidelines to make a therapeutic decision may be challenging to the clinician. When this occurs, consideration might be given to the phenotype that provides the greatest safety for the patient. For example, when a poor to intermediate metabolizer phenotype is reported, the practitioner might consider the patient to be a poor metabolizer for therapeutic considerations.
Finally, the phenotypic report and medication recommendations may be confusing for individuals not familiar with pharmacogenomics. Tacrolimus and CYP3A5 may provide the best example. In general, a poor metabolizer or a rapid/ultrarapid metabolizer is most likely to require avoidance or dosage adjustment of a medication.
25
In the case of the tacrolimus and CYP3A5 drug-gene pair, it is actually the intermediate and normal metabolizers that require an increased dose of tacrolimus.54
This is due to the fact that most white individuals are poor metabolizers of CYP3A5, and original studies to determine the appropriate tacrolimus starting dose were performed on predominantly white populations.Application to Practice
What Should I Consider Along With Pharmacogenomics?
In general, unless you are ordering a test for a specific purpose, it is best to continue current approaches to drug therapy before applying pharmacogenomics. This is challenging if the patient presents with a large amount of pharmacogenomic test results ordered by another clinician that require immediate interpretation.
If a provider will be applying pharmacogenomic test results to individuals taking multiple medications, then drug-drug interactions should be reviewed in addition to the pharmacogenomic results to account for phenoconversion.
55
For example, a patient taking several medications known to prolong the QT interval remains at risk for medication-induced torsade de pointes despite having normal pharmacogenomic results. In the case of phenoconversion, a drug inhibitor or inducer may alter the function of a drug-metabolizing enzyme.55
For example, a patient taking fluoxetine for an extended period may have a phenotype similar to a poor metabolizer of CYP2D6 due to fluoxetine’s inhibition of the enzyme, but the patient’s pharmacogenomic report may indicate the phenotype of a normal metabolizer of CYP2D6. In general, using the pharmacogenomic results independently without consideration of other factors (eg, age, comorbidities, medications) is likely to result in a poor patient outcome.What Are Some of the Additional Clinical Considerations?
When working with metabolic enzymes and drugs, it is important to understand the metabolic consequences of the drug in question. It is also important to understand not only that the drug undergoes metabolism by a given enzyme but also whether it is a prodrug that requires activation or a drug that is active on administration and requires the enzyme for elimination. If we consider a poor metabolizer of CYP2D6, administration of paroxetine
56
would require a dosage reduction to prevent adverse effects due to the inability to properly clear the drug. Conversely, this same poor metabolizer of CYP2D6 would likely find codeine inefficacious because the codeine prodrug was not sufficiently activated or converted to morphine, which is responsible for codeine’s analgesic effect.46
It is clear that a poor metabolizer of CYP2D6 is likely to have pharmacogenomic issues because CYP2D6 metabolizes approximately 25% of the drugs that undergo CYP450 biotransformation.57
What is less clear is the degree of issues in the case of an intermediate metabolizer of CYP2D6. Although there is some reduction in metabolic capability, depending on the drug, the reduction may not be clinically significant enough to warrant a change in therapy.Pharmacogenomic results may become available after an individual has been administered a medication long-term rather than before initiation. An individual taking warfarin with an INR in the therapeutic range may receive pharmacogenomic results indicating warfarin sensitivity. In this case, although the individual may be sensitive to warfarin, the dose was already titrated to achieve this adequate INR. The potential value of warfarin pharmacogenomic testing is before warfarin initiation, when a lower starting dose could be selected and quicker adjustment to a therapeutic INR might be achieved.
58
,59
On comprehensive reports that identify individual drugs that may present a risk for the patient, it is the clinician’s responsibility to consider the risk vs benefit of prescribing that medication. An alternative drug may not have a pharmacogenomic interaction, but it might be an inappropriate drug choice for the clinical situation. Consider a pharmacogenomic report that flags codeine, tramadol, hydrocodone, and oxycodone due to CYP2D6 polymorphism resulting in poor metabolism. On review, many clinicians might conclude that due to the drug-gene interaction, none of these drugs are ever appropriate to use. Although CYP2D6-guided therapy may provide potential value in pain management,
60
the recommended alternative’s overall toxicity remains a consideration.One of the issues of working with some comprehensive reports is that they explain neither why the medication may be inappropriate nor the absolute degree of risk. In this case, codeine and tramadol are well documented to be inefficacious in poor metabolizers of CYP2D6.
46
However, the same level of evidence is not available for hydrocodone and oxycodone because they are not completely dependent on CYP2D6 metabolism to confer analgesic activity.61
Although some studies suggest some loss of activity of hydrocodone in poor metabolizers of CYP2D6 and with drug interactions that inhibit CYP2D6, this has not been clearly demonstrated with oxycodone.60
,62
Of note, several genes that are commonly tested for pharmacogenomic purposes are also associated with disease states. For example, G6PD genetic variants have implications for many drugs, and variants in this gene are known to cause glucose-6-phosphate dehydrogenase deficiency,
63
and UGT1A1 variation affects irinotecan22
and atazanvir64
but also can result in Gilbert syndrome or Criger-Najjar syndrome.65
For these and other genes associated with hereditary disorders, it is important for providers to be aware of the possibility of a secondary finding in addition to gaining information about drug metabolism so that the patient is appropriately counseled before testing.Discussion
Pharmacogenomics is increasingly being implemented in patient care. This is greatly assisted by consortiums of experts who provide drug-gene information. At some larger institutions, many resources have been allocated to implementation of pharmacogenomic guideline recommendations.
66
In some health care systems, the electronic health record (EHR) is used to assist clinicians with integration of pharmacogenomics into practice.67
,68
If genetic information is available in the EHR, computerized clinical decision support (CDS) rules are used to alert the prescriber of a potential drug-gene interaction for certain medications.69
These drug-gene alerts are analogous to drug-drug interaction alerts in place at many institutions and pharmacies.69
, 70
, 71
A computerized resource with additional pharmacogenomic information to supplement the abbreviated information provided in an alert may also be available to assist practitioners.69
Although some institutions may have the ability to provide these types of resources, currently many smaller institutions, pharmacies, and solo medical practices do not. However, with increased availability of consumer-directed pharmacogenomic tests72
it is increasingly likely that these clinicians will be presented with pharmacogenomic information directly from their patients.Because many clinicians have had limited education in the past, the first step in implementation in a practice should be education. Although education remains a challenge for implementation,
73
educational endeavors in pharmacogenomics are currently ongoing for practitioners, with courses, seminars, online modules, and other educational modalities.74
In addition, some providers of this education include a combination of self-study and live programs with simulated patients.75
For self-education, one of the best resources immediately available is the PharmGKB website (https://www.pharmgkb.org), which can be searched by medication or by gene.
14
Clinicians might start by considering medications applicable to their current practice and their pharmacogenomic implications. Full guidelines are also available from the CPIC and other organizations that can be reviewed in more detail.The next consideration is the types of pharmacogenomic testing available to your practice. Ideally, results from single-gene or panel tests will populate discrete fields in the EHR, which is essential if the institution plans to use a CDS system. Not having these resources does not preclude the solo practitioner from ordering a single-gene or panel test to guide therapy; however, there should be an archiving mechanism in place to ensure that the report is available for future use. Owing to the current variability in testing and reporting, it is ideal to limit the number of laboratories that provide testing for your patient population. Because many reports are laboratory specific, report format continuity will provide a greater ease of use for practitioners. In addition, limiting sources will facilitate migration of results into the medical record and the design of CDS systems. In large institutions, the testing available is often determined by a committee rather than by individual providers. If multiple testing options are available, clinicians may still find it helpful to order tests from a single source.
Finally, the individual practitioner should determine the resources available to support integration into his or her practice. In addition to physicians, other health care providers, including nursing,
76
physician assistants,77
and genetic counselors, have identified the importance of pharmacogenomics to their practice.78
This science may present a special challenge due to the combination of clinical pharmacology and genomic science. The successful application of pharmacogenomics and patient counseling is likely to arise from a multidisciplinary collaboration.79
, 80
, 81
When considering resources, the laboratory is usually the first point of support providing testing and interpretive information. In larger institutions or practices, integration can also be facilitated by working with the pharmacy department. Pharmacogenomics is a natural addition to the drug reviews currently being performed by some pharmacists daily. With pharmacogenomic training, pharmacists can be an institutional asset for ordering, interpreting, and applying pharmacogenomic results to medication therapy. Although this may seem easier to accomplish in larger institutions, many small practices usually have some access to a pharmacist in the community. These individuals can also likely provide assistance because many pharmacists are adding pharmacogenomics to their hospital, outpatient, and community pharmacy practices. However, for providers who do not have the ability to interact with a pharmacist on a regular basis, laboratories that perform testing typically also offer support for pharmacogenomic questions and test result interpretations, potentially including pharmacist consultations.
The field of pharmacogenomics is rapidly evolving from a scientific, economic, and regulatory perspective. On the regulatory front, shortly after approving 23andMe to perform pharmacogenomic tests and report genotypes and phenotypes—but not medication recommendations—directly to consumers in 2018,
36
the FDA issued a safety communication warning patients and providers about genetic tests to predict medication response.82
The FDA typically has practiced enforcement discretion for laboratory-developed tests, which is the category that encompasses most commercially available pharmacogenomic tests. In recent years, some laboratories have begun to offer consumer-initiated laboratory-developed tests where the consumer chooses to have testing performed, which is then ordered by a provider who may not be otherwise involved in that individual’s care. This process with limited provider involvement may represent a higher risk to the patient than traditional testing routes if the patient should choose to act on the results and stop taking medication without the guidance of his or her provider.Jeffrey Shuren, M.D., J.D., director of the FDA’s Center for Devices and Radiological Health and Janet Woodcock, M.D., director of the FDA’s Center for Drug Evaluation and Research on agency’s warning to consumers about genetic tests that claim to predict patients’ responses to specific medications. News release. U.S. Food & Drug Administration. November 1, 2018.
The FDA opted to regulate a subset of PGx laboratory-developed tests and issued a warning letter in early 2019 to Inova Genomics for illegally marketing PGx tests that have not been reviewed by the FDA.
83
In addition, the FDA contacted several other companies, which addressed the concerns by removing specific medication names from promotional materials and patient test reports. In early 2020, groups concerned about the FDA’s recent actions have filed a citizen’s petition.84
More recently, the FDA published a table of pharmacogenetic associations.
85
Although not advocating routine testing, the FDA recognizes that the knowledge of a patient’s genotype may assist in guiding therapy and preventing toxicity. The information initially curated in this table by the FDA will serve to provide an additional resource to practitioners and allows a mechanism for updates as new evidence-based pharmacogenomic information is discovered.Table of pharmacogenetic associations. U.S. Food & Drug Administration.
https://www.fda.gov/medical-devices/precision-medicine/table-pharmacogenetic-associations
Date accessed: February 21, 2020
On the reimbursement front, pharmacogenomic tests may not be covered by insurance plans and remain a practice issue for many providers. Recently, commercial insurers, led by the nation’s largest insurer, UnitedHealthcare, have taken first steps to cover pharmacogenetic testing ahead of following medical service coverage decisions by the Centers for Medicare & Medicaid Services.
86
,- England J.
3 things UnitedHealth’s pharmacogenetic testing coverage decision means for providers. OneOme. Updated August 23, 2019.
https://oneome.com/blog/3-things-uhg-pharmacogenetic-coverage-means-providers/
Date accessed: January 21, 2020
87
In October 2019, UnitedHealthcare announced its decision to cover testing for patients with major depressive disorder and/or anxiety who have had at least 1 treatment failure. This step provides testing access to more than 27 million individual and group plans.88
Although it is encouraging that testing coverage continues to gain traction, more cost-effectiveness and utility studies89
are needed to support further expansion of test coverage to limit out-of-pocket expenses by patients.Conclusion
Although many current clinicians have had little training in pharmacogenomics in the past, this new science is increasingly being used in practice. In addition, patients who have had previous pharmacogenomic testing are presenting their test results to their health care providers with the expectation that these results will be applied to their current and future medication therapies. Many considerations with pharmacogenomics are unique but can be integrated into regular pharmacologic practice because many principles are similar. Knowledge of the differences and resources available will assist clinicians with integration of pharmacogenomics into clinical practice.
Acknowledgments
We gratefully acknowledge the ongoing work provided by the CPIC, the DPWG, the Canadian Pharmacogenomics Network for Drug Safety, and Mayo Clinic Center for Individualized Medicine.
Supplemental Online Material
References
- Pharmacogenomics education in medical and pharmacy schools: conclusions of a global survey.Pharmacogenomics. 2019; 20: 643-657
- Pharmacy students' attitudes and perceptions toward pharmacogenomics education.Am J Health Syst Pharm. 2019; 76: 836-845
- Pharmacogenomics instruction in US and Canadian medical schools: implications for personalized medicine.Pharmacogenomics. 2010; 11: 1331-1340
- On the readiness of physicians for pharmacogenomics testing: an empirical assessment.Pharmacogenomics J. 2018; 18: 308-318
- Physician-reported benefits and barriers to clinical implementation of genomic medicine: a multi-site IGNITE-Network Survey.J Pers Med. 2018; 8: 24
- Pharmacogenomics in the curricula of colleges and schools of pharmacy in the United States.Am J Pharm Educ. 2010; 74: 7
- An effective approach to teaching pharmacogenomics in the first year of pharmacy curriculum.Am J Pharm Educ. 2018; 82: 6345
- Pharmacogenomics competencies in pharmacy practice: a blueprint for change.J Am Pharm Assoc (2003). 2017; 57: 120-125
- Clinical pharmacology education: the decade ahead.Clin Pharmacol Ther. 2020; 107: 37-39
- Assessment of the pharmacogenomics educational needs of pharmacists.Am J Pharm Educ. 2011; 75: 51
- Incorporation of pharmacogenomics into routine clinical practice: the Clinical Pharmacogenetics Implementation Consortium (CPIC) guideline development process.Curr Drug Metab. 2014; 15: 209-217
- Pharmacogenetics: from bench to byte—an update of guidelines.Clin Pharmacol Ther. 2011; 89: 662-673
- The Canadian Pharmacogenomics Network for Drug Safety: a model for safety pharmacology.Thyroid. 2010; 20: 681-687
- PharmGKB: a worldwide resource for pharmacogenomic information.Wiley Interdiscip Rev Syst Biol Med. 2018; 10: e1417
- A retrospective investigation of HLA-B∗5801 in hyperuricemia patients in a Han population of China.Pharmacogenet Genomics. 2018; 28: 117-124
- Clinical Pharmacogenetics Implementation Consortium guidelines for human leukocyte antigen-B genotype and allopurinol dosing.Clin Pharmacol Ther. 2013; 93: 153-158
- Stevens-Johnson syndrome/toxic epidermal necrolysis: a multicenter retrospective study of 377 adult patients from the United States.J Invest Dermatol. 2018; 138: 2315-2321
- HLA-B∗57:01 allele prevalence in HIV-infected North American subjects and the impact of allele testing on the incidence of abacavir-associated hypersensitivity reaction in HLA-B∗57:01-negative subjects.BMC Infect Dis. 2017; 17: 256
- The mechanistic differences in HLA-associated carbamazepine hypersensitivity.Pharmaceutics. 2019; 11: 536
- Clinical Pharmacogenetics Implementation Consortium (CPIC) guideline for dihydropyrimidine dehydrogenase genotype and fluoropyrimidine dosing: 2017 update.Clin Pharmacol Ther. 2018; 103: 210-216
- Clinical Pharmacogenetics Implementation Consortium guideline for thiopurine dosing based on TPMT and NUDT15 genotypes: 2018 update.Clin Pharmacol Ther. 2019; 105: 1095-1105
- Study on the optimal dose of irinotecan for patients with heterozygous uridine diphosphate-glucuronosyltransferase 1A1 (UGT1A1).Biol Pharm Bull. 2019; 42: 1839-1845
- Using pharmacogenetics to understand adverse drug reactions in children.Paediatr Child Health. 2011; 16: 537-538
- Challenges in CYP2D6 phenotype assignment from genotype data: a critical assessment and call for standardization.Curr Drug Metab. 2014; 15: 218-232
- Polymorphism of human cytochrome P450 2D6 and its clinical significance: part I.Clin Pharmacokinet. 2009; 48: 689-723
- Trends in tramadol: pharmacology, metabolism, and misuse.Anesth Analg. 2017; 124: 44-51
- A meta-analysis of CYP2D6 metabolizer phenotype and metoprolol pharmacokinetics.Clin Pharmacol Ther. 2013; 94: 394-399
- Preemptive panel-based pharmacogenetic testing: the time is now.Pharm Res. 2017; 34: 1551-1555
- Pharmacogenomics implementation: considerations for selecting a reference laboratory.Pharmacotherapy. 2017; 37: 1014-1022
- Recommendations for clinical CYP2C19 genotyping allele selection: a report of the Association for Molecular Pathology.J Mol Diagn. 2018; 20: 269-276
- Recommendations for clinical CYP2C9 genotyping allele selection: a joint recommendation of the Association for Molecular Pathology and College of American Pathologists.J Mol Diagn. 2019; 21: 746-755
- Insurance coverage policies for personalized medicine.J Pers Med. 2012; 2: 201-216
- Genotype, phenotype, and medication recommendation agreement among commercial pharmacogenetic-based decision support tools.Pharmacogenomics J. 2018; 18: 613-622
- False-positive results released by direct-to-consumer genetic tests highlight the importance of clinical confirmation testing for appropriate patient care.Genet Med. 2018; 20: 1515-1521
- A call for clear and consistent communications regarding the role of pharmacogenetics in antidepressant pharmacotherapy.Clin Pharmacol Ther. 2020; 107: 50-52
- FDA authorizes first direct-to-consumer test for detecting genetic variants that may be associated with medication metabolism. News release. U.S. Food & Drug Administration. October 31, 2018.
23andMe. FDA grants 23andMe clearance to offer interpretive drug information for two medications. 23andMe website, FDA grants 23andMe clearance to offer interpretive drug information for two medications. Published August 18, 2020. Accessed October 25, 2020.
- Standardization can accelerate the adoption of pharmacogenomics: current status and the path forward.Pharmacogenomics. 2018; 19: 847-860
- Standardizing terms for clinical pharmacogenetic test results: consensus terms from the Clinical Pharmacogenetics Implementation Consortium (CPIC).Genet Med. 2017; 19: 215-223
- Characterization of 107 genomic DNA reference materials for CYP2D6, CYP2C19, CYP2C9, VKORC1, and UGT1A1: a GeT-RM and Association for Molecular Pathology collaborative project.J Mol Diagn. 2010; 12: 835-846
- Methods to analyze big data in pharmacogenomics research.Pharmacogenomics. 2017; 18: 807-820
- Pharmacogenomic discovery to function and mechanism: breast cancer as a case study.Clin Pharmacol Ther. 2018; 103: 243-252
- Clinical Pharmacogenetics Implementation Consortium guideline for HLA genotype and use of carbamazepine and oxcarbazepine: 2017 update.Clin Pharmacol Ther. 2018; 103: 574-581
- Race influences warfarin dose changes associated with genetic factors.Blood. 2015; 126: 539-545
- Pharmacogenetic allele nomenclature: international workgroup recommendations for test result reporting.Clin Pharmacol Ther. 2016; 99: 172-185
- Clinical Pharmacogenetics Implementation Consortium guidelines for cytochrome P450 2D6 genotype and codeine therapy: 2014 update.Clin Pharmacol Ther. 2014; 95: 376-382
- Clinical Pharmacogenetics Implementation Consortium (CPIC) guideline for CYP2D6 and tamoxifen therapy.Clin Pharmacol Ther. 2018; 103: 770-777
- PharmVar GeneReview: CYP2D6.Clin Pharmacol Ther. 2020; 107: 154-170
- Standardizing CYP2D6 genotype to phenotype translation: consensus recommendations from the Clinical Pharmacogenetics Implementation Consortium and Dutch Pharmacogenetics Working Group.Clin Transl Sci. 2020; 13: 116-124
- Challenges in ordering and interpreting pharmacogenomic tests in clinical practice.Am J Med. 2017; 130: 1342-1344
- Utility of integrated pharmacogenomic testing to support the treatment of major depressive disorder in a psychiatric outpatient setting.Pharmacogenet Genomics. 2013; 23: 535-548
- The 1200 patients project: creating a new medical model system for clinical implementation of pharmacogenomics.Clin Pharmacol Ther. 2012; 92: 446-449
- Analytical validation of a psychiatric pharmacogenomic test.Per Med. 2018; 15: 189-197
- Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines for CYP3A5 genotype and tacrolimus dosing.Clin Pharmacol Ther. 2015; 98: 19-24
- Addressing phenoconversion: the Achilles' heel of personalized medicine.Br J Clin Pharmacol. 2015; 79: 222-240
- Clinical Pharmacogenetics Implementation Consortium (CPIC) guideline for CYP2D6 and CYP2C19 genotypes and dosing of selective serotonin reuptake inhibitors.Clin Pharmacol Ther. 2015; 98: 127-134
- Influence of cytochrome P450 polymorphisms on drug therapies: pharmacogenetic, pharmacoepigenetic and clinical aspects.Pharmacol Ther. 2007; 116: 496-526
- Warfarin loading dose guided by pharmacogenetics is effective and safe in cardioembolic stroke patients: a randomized, prospective study.Pharmacogenomics J. 2019; 19: 446-454
- Effect of genotype-guided warfarin dosing on clinical events and anticoagulation control among patients undergoing hip or knee arthroplasty: the GIFT randomized clinical trial.JAMA. 2017; 318: 1115-1124
- CYP2D6-guided opioid therapy improves pain control in CYP2D6 intermediate and poor metabolizers: a pragmatic clinical trial.Genet Med. 2019; 21: 1842-1850
- Clinical perspective on the Clinical Pharmacogenetics Implementation Consortium updated 2014 guidelines for CYP2D6 and codeine.Clin Chem. 2015; 61: 319-321
- Predicting inadequate postoperative pain management in depressed patients: a machine learning approach.PLoS One. 2019; 14: e0210575
- G6PD deficiency: a classic example of pharmacogenetics with on-going clinical implications.Br J Haematol. 2014; 164: 469-480
- Clinical Pharmacogenetics Implementation Consortium (CPIC) guideline for UGT1A1 and atazanavir prescribing.Clin Pharmacol Ther. 2016; 99: 363-369
- Genotype of UGT1A1 and phenotype correlation between Crigler-Najjar syndrome type II and Gilbert syndrome.J Gastroenterol Hepatol. 2016; 31: 403-408
- Preemptive clinical pharmacogenetics implementation: current programs in five US medical centers.Annu Rev Pharmacol Toxicol. 2015; 55: 89-106
- Integrating pharmacogenomics into the electronic health record by implementing genomic indicators.J Am Med Inform Assoc. 2020; 27: 154-158
- Integrating pharmacogenomics into electronic health records with clinical decision support.Am J Health Syst Pharm. 2016; 73: 1967-1976
- Electronic medical record-integrated pharmacogenomics and related clinical decision support concepts.Clin Pharmacol Ther. 2017; 102: 254-264
- Development and use of active clinical decision support for preemptive pharmacogenomics.J Am Med Inform Assoc. 2014; 21: e93-e99
- Multidisciplinary model to implement pharmacogenomics at the point of care.Genet Med. 2017; 19: 421-429
- User considerations in assessing pharmacogenomic tests and their clinical support tools.NPJ Genom Med. 2018; 3: 26
- Education and knowledge in pharmacogenomics: still a challenge?.Clin Pharmacol Ther. 2018; 103: 752-755
- Educational strategies to enable expansion of pharmacogenomics-based care.Am J Health Syst Pharm. 2016; 73: 1986-1998
- Academic and professional pharmacy education: a pharmacogenomics certificate training program.Per Med. 2015; 12: 563-573
- Pharmacogenomics and implications for nursing practice.J Nurs Scholarsh. 2015; 47: 496-504
- Pharmacogenetics: what PAs need to understand and why.JAAPA. 2009; 22: 61-62
- A role for genetic counsellors and clinical geneticists in pharmacogenetics?.Clin Genet. 2012; 82: 201-202
author reply 203.
- Collaborative counseling considerations for pharmacogenomic tests.Pharmacotherapy. 2017; 37: 990-999
- Implementation of a multidisciplinary pharmacogenomics clinic in a community health system.Am J Health Syst Pharm. 2016; 73: 1956-1966
- Clinical utilization of pharmacogenetics in psychiatry: perspectives of pharmacists, genetic counselors, implementation science, clinicians, and industry.Pharmacopsychiatry. 2020; 53: 162-173
- Jeffrey Shuren, M.D., J.D., director of the FDA’s Center for Devices and Radiological Health and Janet Woodcock, M.D., director of the FDA’s Center for Drug Evaluation and Research on agency’s warning to consumers about genetic tests that claim to predict patients’ responses to specific medications. News release. U.S. Food & Drug Administration. November 1, 2018.
- FDA issues warning letter to genomics lab for illegally marketing genetic test that claims to predict patients’ responses to specific medications. News release. U.S. Food & Drug Administration. April 4, 2019.
- Coalition files citizen petition to stop ‘unlawful' FDA actions against PGx testing. 360Dx.(Updated January 10, 2020)
- Table of pharmacogenetic associations. U.S. Food & Drug Administration.(Updated February 20, 2020)https://www.fda.gov/medical-devices/precision-medicine/table-pharmacogenetic-associationsDate accessed: February 21, 2020
- 3 things UnitedHealth’s pharmacogenetic testing coverage decision means for providers. OneOme. Updated August 23, 2019.https://oneome.com/blog/3-things-uhg-pharmacogenetic-coverage-means-providers/Date accessed: January 21, 2020
- Insurance coverage for genomic tests.Science. 2018; 360: 278-279
- DNA tests for psychiatric drugs are controversial but some insurers are covering them. NPR.(Updated October 17, 2019)
- Cost-effectiveness of multigene pharmacogenetic testing in patients with acute coronary syndrome after percutaneous coronary intervention.Value Health. 2020; 23: 61-73
Article info
Publication history
Published online: December 09, 2020
Footnotes
Potential Competing Interests: The authors report no competing interests. Mayo Clinic Laboratory provides pharmacogenetic testing. Mayo Clinic has stock ownership in and has licensed intellectual property to OneOme LLC. Mayo Clinic also receives royalties from Assurex Health.
Identification
Copyright
© 2020 Mayo Foundation for Medical Education and Research
