Clinical pharmacogenomics contributes to optimal pharmacotherapy for a variety of conditions, including heart disease,1 venous thromboembolism,2 cancers,3 and epilepsy,4 by avoiding harmful adverse effects, drug–gene interactions, and ineffective treatment. Selected patients have already benefited from pharmacogenomic testing, as genetic markers are increasingly used to guide warfarin dosing,5 improve responses to antiplatelet therapy,6 and reduce toxicity from thiopurine therapy.7,8
There is growing consensus that pharmacogenomics is an essential component of pharmacy practice and, correspondingly, that pharmacists should lead implementation efforts in this area. Consensus recommendations developed as part of ASHP’s Pharmacy Practice Model Initiative (PPMI) underscore this point, as 74% of participants in the 2010 PPMI Summit agreed that “adjustment of medication regimens based on individual genetic characteristics is an essential pharmacist-provided drug therapy management service.”9 Notably, ASHP recently adopted a policy statement on the Pharmacist’s Role in Clinical Pharmacogenomics,10 supplementing policies, PPMI statements, textbooks, and an online resource center developed by the ASHP Section of Clinical Specialists and Scientists’ Advisory Group on Emerging Sciences.11
The ASHP statement on pharmacogenomics is a landmark for several reasons. First, clinical pharmacogenomic testing is slowly emerging as a routine practice. According to the 2013 ASHP national survey of pharmacy practice in hospital settings, pharmacogenomic testing was provided at 7% of respondent hospitals, as compared with just 2.7% of surveyed hospitals in 2009.12 Second, clinical implementation of pharmacogenomics is expected to grow––initially in academic medical centers, with steady expansion to other settings. In fact, according to the most recent ASHP Foundation Pharmacy Forecast report, 79% of pharmacy leaders participating in a recent survey expected that at least one academic medical center in their area will have a pharmacy-based pharmacogenomics service within the next five years.11 Third, the statement reinforces signals from non-pharmacist healthcare providers that pharmacy practice should include pharmacogenomics. One nationally recognized geneticist and researcher recently asserted that “pharmacogenomics may reside more comfortably in the purview of pharmacists rather than physicians, at least as far as programmatic development and leadership are concerned.”13
Value of pharmacogenomics
Despite advances in medication-use technology and practices over the past decades, many opportunities to optimize patient safety remain, and pharmacogenomics holds promise as part of a proactive medication safety strategy. To illustrate, in a recent cohort study of medication-use patterns in over 52,000 adult primary care patients, it was estimated that 64.8% of patients received a drug with a known pharmacogenomic association over a five-year period14; after pharmacogenomic panel testing for five genes was instituted, 91% of the patients tested had clinically actionable results.15 In a similar study in which testing for four genes was performed in a pediatric population, 78% of patients had clinically actionable results.16 Additional estimates suggest that preemptively screening patients for 12 pharmacogenes would result in a clinically actionable result in 97% of U.S. patients.17 Moreover, when considering the 30 most commonly used drugs associated with potentially actionable pharmacogenomic information, there were 738 million prescriptions dispensed in 2013—in essence, millions of opportunities to apply pharmacogenomics to comprehensive medication management.
Two prevailing approaches to clinical implementation include reactive and preemptive pharmacogenomic testing. Reactive testing is typically ordered when initiating a drug or when a patient experiences unexplained—and potentially drug-related—adverse effects. Advantages to reactive testing include a reduced upfront financial outlay, reduced informatics infrastructure needs, and reliable access to laboratories that provide single-gene testing; however, this approach has several drawbacks. First, the correct test must be ordered, retrieved, and interpreted by the clinician, after which the patient may need to be recontacted if drug therapy changes are needed; this type of scenario is especially problematic when treatment cannot be delayed (e.g., clopidogrel is indicated for acute coronary syndrome). Second, a delay in obtaining genetic test results may render the information obsolete (interim decisions are required), and providers may not respond effectively to genetic results returned beyond the typical time frame of a clinical encounter. Finally, tests for single genes are expensive (relative to the potential benefit) when used to guide a single therapeutic decision.14
Preemptive testing involves assaying multiple pharmacogenes prospectively and storing the results electronically for future use, often before an indication for drug therapy exists.14 An appealing feature is that testing can be multiplexed, with hundreds to thousands of genetic variants assayed at once, which enables more affordable genotyping and application of advanced clinical decision support tools at the point of care. In this paradigm, prescription order entry automatically triggers a search for relevant drug–gene interactions for an individual patient; if clinically actionable variants are identified, the system guides the clinician toward appropriate therapy.18 Actionable genetic results can be reused over time as patients are prescribed new medications and as the knowledge base of pharmacogenomics-based prescribing grows. While the cost per genotype is lower, such testing requires a larger upfront financial investment—particularly for the development of clinical decision support tools. However, there are important ethical, legal, and social implications to consider with this model; for example, questions still remain about how to resolve clinical uncertainties posed by collecting genetic information of unknown significance and by potentially significant incidental findings. Ongoing and future implementation research efforts are targeting practical ways to translate genomic information into routine clinical practice.19
Several significant trends and enabling factors bolster the rationale for pharmacists to provide strategic leadership in initiatives to expand the clinical use of pharmacogenomics. First, widespread availability of affordable genotyping platforms, especially array-based panels, enhances pharmacists’ capacity to proactively use patients’ genotype information to guide medication selection and dosing in practice. Second, the establishment of the Clinical Pharmacogenetics Implementation Consortium, which has published 14 evidence-based clinical practice guidelines (including 3 endorsed by ASHP), assists clinicians with evidence-based recommendations to optimize drug therapy.20 Third, new Accreditation Council for Pharmacy Education requirements for pharmacy students21 slated to go into effect in 2016 will empower schools of pharmacy to require teaching of clinical pharmacogenomics, thereby increasing the competence of future pharmacists to practice in this area.
Beyond pharmacy, initiatives to develop genomics-oriented educational resources for genetics practitioners and healthcare providers without advanced genetics expertise (e.g., the National Institutes of Health–sponsored Genetics and Genomics Competency Center, or G2C2) recognize pharmacists as part of the genomics team; other resources are listed in the appendix. Several national organizations are focusing on the role of pharmacists in pharmacogenomics, underscoring the point that pharmacogenomics fits into a larger ecosystem (i.e., “genomic medicine”). A recent illustration came from an Institute of Medicine workshop that explored potential ways to enhance genetics training in graduate professional education (e.g., residencies and fellowships) for healthcare work forces of the present and the future.22 Although the workshop focused on the educational needs of healthcare providers without advanced genetics expertise (e.g., most pharmacists, physicians, nurses, and physician assistants), a persistent theme emerged: Education by itself may be insufficient to elicit meaningful behavioral change among providers or to inspire interest among health professions students and trainees; therefore, deliberate cultivation of interprofessional leadership and teamwork is needed to move the needle. Cobbling together clinical education, continuous professional development, and team-based care within the “learning health system” may be the best pathway for successful implementation across the spectrum of healthcare delivery.23
Successful clinical implementation of pharmacogenomics involves collaboration and coordination with third-party payers. Within the current care delivery landscape, the “reimbursability” of a particular diagnostic test significantly contributes to the extent of its implementation into routine use. While reimbursement for pharmacogenomic testing has been successful in selected situations,24 a lack of uniformity in approach among payers and across geographic regions hampers uptake.
A lack of education and awareness among physicians and pharmacists also limits broad application of pharmacogenomic information. Fortunately, there are emerging enablers that leverage existing clinical pathways and electronic health records (EHRs) to provide reliable and reproducible clinical decision support. This underscores the importance of clinical informatics in realizing the potential of pharmacogenomics.
Finally, without direct engagement and support from health-system leaders and stakeholders, clinical implementation efforts are challenged. Additional issues may arise in systems lacking suitable infrastructure (i.e., EHR systems and adequate data storage facilities); however, as one noted pharmacy leader and pioneer in pharmacogenomics has stated, “Culture trumps strategy,” suggesting that many barriers can be overcome by cultivating a can-do leadership culture.25
Using pharmacogenomic information for patient care requires systematic review and critical appraisal of published research. The prevailing implementation models discussed above provide different lenses with which to evaluate such evidence. For example, collecting evidence to elucidate the optimal time to order a pharmacogenomic assay as part of a reactive testing approach is very challenging given complex research considerations, including adequate sampling, recruitment, and funding. In this scenario, pharmacogenomic tests are often held to excessively high evidentiary standards.26,27
Alternatively, if pharmacogenomic results are available for patients, there is well-developed evidence to support interventions based on known drug–gene relationships. These insights are driven by a deep understanding of pharmacokinetic and pharmacodynamic mechanisms and how they influence drug disposition, toxicity, and response. Therefore, it stands to reason that with the preemptive approach, there are scenarios in which it becomes questionable to not use available genetic information to inform drug therapy management.
Further, in consideration of clinical pharmacogenomics as part of a comprehensive medication safety strategy, the evidence standard is then similar to what is used to support the implementation of other proven and widely used medication safety interventions. For example, managing critical drug interactions is part of standard pharmacy practice despite the lack of randomized controlled trial evidence to support each individual intervention. In lieu of such evidence, rigorously collected and validated mechanism data support routine clinical interventions, with resultant clear evidence of improved clinical and safety outcomes.
At a broader level, certain changes in clinical practice can and should occur for various reasons—not just the publication of data from randomized controlled trials. A prominent current example is widespread implementation of EHRs per the meaningful-use criteria outlined in the Health Information Technology for Economic and Clinical Health Act, which has been implemented with little (and, in some cases, negative) randomized controlled trial data.28 While EHRs are not simple to use, maintain, and optimize to improve patient care, when effort is invested the benefits to patients are obvious. Similarly, implementing pharmacogenomics requires a strategic commitment but provides clear value in terms of improved pharmacotherapy.
The path forward
Fortunately, robust experience is emerging from organizations that have already invested in clinical implementation. Here are some critical success factors for relevant stakeholders to consider:
Identify and hire a pharmacist with expertise in pharmacogenomics and dedicate time to coordinate the implementation of pharmacogenomics; some organizations have a position devoted entirely to this area.
Engage other pharmacists with relevant expertise in the implementation process, including clinical specialists and those with informatics, medication safety, medication-use policy, and education experience.29
Determine the genes and drugs of interest to the targeted patient population and the organization.
Partner with a capable clinical laboratory for pharmacogenomic testing.
Collaborate closely with clinical informatics personnel to develop clinical decision support tools that facilitate pharmacogenomic testing and appropriate use of results at the point of care.
Develop an ongoing process to provide education to pharmacists and other healthcare professionals, leveraging the growing number of resources already available.
The profession faces a future in which drugs will increasingly be developed for discrete subpopulations, with pharmacogenomic applications contributing to more efficient and precise therapy. Additionally, the growing number of useful pharmacogenomic tests with the potential to enhance the safety and effectiveness of pharmacotherapy will be translated into measurable population health improvement.
To fully realize the promise and potential of precision medicine, pharmacogenomics must be seamlessly embedded into EHRs with clinical decision support tools, driven by tailored blood and tissue tests aimed at precise diagnosis, and (perhaps) a personal genomic sequence linked to every patient’s medical record.30 Because pharmacists are entrusted to lead safe and effective medication-use processes across care delivery settings, we must ensure the inclusion of clinical pharmacogenomics into comprehensive medication safety strategies across health systems and care delivery settings.
Appendix Pharmacogenomics resources relevant to pharmacists
James Hoffman, Pharm.D., M.S., and Cyrine Haidar, Pharm.D., are acknowledged for their support and contributions during manuscript development.
The author has declared no potential conflicts of interest.
- Copyright © 2015 by the American Society of Health-System Pharmacists, Inc. All rights reserved.