Progress in Pharmacogenomics Implementation in the United States: Barrier Erosion and Remaining Challenges

Abstract

Barriers to incorporating pharmacogenetics into routine clinical practice in the United States are well documented. Initial surveys by the Clinical Pharmacogenetics Implementation Consortium (CPIC) in 2009 and 2010 identified barriers across four key domains that have hindered the widespread adoption of clinical pharmacogenetic testing. These are presented verbatim as: (i) absence of a definition of the processes required to interpret genotype information and to translate genetic information into clinical actions; (ii) need for recommended drug/gene pairs to implement clinically now; (iii) clinician resistance to consider pharmacogenetic information at the bedside; and (iv) concerns about test costs and reimbursement. Over time, many of these challenges have been overcome, and clinical pharmacogenetic testing has subsequently reached broader implementation. Despite this progress, several barriers remain that block further adoption. This narrative review used authors' expertise and experience to identify and describe current barriers to pharmacogenetic implementation across seven domains in the United States: equity and inclusion; guidelines and supporting evidence; regulatory agency oversight; payer coverage and insurance; availability of quality pharmacogenetic tests; electronic health records; and provider and patient education. Within each domain, it revisits past successes and challenges and explores remaining barriers. We also propose solutions to address ongoing challenges across these domains, including further expansion of recommendations beyond pharmacogenetic‐specific guidelines, standards for designing clinical decision support tools, and broader pharmacogenetics education. Addressing these remaining obstacles directs work to enable broader adoption of clinical pharmacogenetic implementation to ultimately improve patient outcomes.

The mission of the Pharmacogenomics Global Research Network's (PGRN) is to catalyze and lead research in precision medicine for the discovery and translation of genomic variation influencing therapeutic and adverse drug effects. ¹ The original Pharmacogenomics Research Network (PGRN) began in 2000 as an NIH‐funded network and transitioned to the current global independent scientific society in 2020. The PGRN and the Pharmacogenomics Knowledgebase (PharmGKB) collaborated to create the Clinical Pharmacogenetics Implementation Consortium (CPIC) in 2009 to create freely available, evidence‐based pharmacogenetic prescribing guidelines. ² , ³ Near its formation, CPIC conducted two surveys that identified the primary barriers to the implementation of pharmacogenetics. These were reported verbatim as: (1) the absence of a definition of the processes required to interpret genotype information and to translate genetic information into clinical actions, (2) the need for recommended drug/gene pairs to implement clinically, (3) clinician resistance to utilizing pharmacogenetic information, and (4) concerns about test costs and reimbursement. ²

Survey responses stressed the importance of genotype test interpretation based on supporting scientific evidence, as at the time, many believed that pharmacogenetic testing would become widely available due to increasingly lower genotyping costs, effectively removing that barrier. ² Hence, CPIC guidelines were intentionally designed to focus on how to interpret pharmacogenetic test results rather than when to order a test, an approach that continues to today. ⁴ CPIC guidelines have made substantial progress in addressing barriers #1 and #2 through their standardized format, evidence‐based rating scheme, mode of dissemination, and strategy for prioritization of drug/gene pairs. It is evident that CPIC has helped to advance clinical pharmacogenetics toward broader implementation. Despite this progress, previous barriers remain, and new barriers have arisen blocking further adoption. This review aims to describe current barriers to broad pharmacogenetic test implementation (Table 1 ). ⁵ , ⁶ , ⁷ This narrative review will describe the barriers, recent progress, and potential solutions (Table 2 ) for seven domains of pharmacogenetics implementation selected per author expertise: equity and inclusion; standardization, quality of evidence, and guidelines; regulatory agency oversight; payer coverage and insurance; availability of quality pharmacogenetic tests; integration into electronic health records (EHR) and clinical decision support (CDS); provider and patient education.

Table 1.

Barriers to Implementation of Pharmacogenetics into Care

Domain	Cited barrier
Equity and Inclusion	Continued underrepresentation of certain populations in pharmacogenetics research and implementation Cost of testing, particularly for patients from marginalized communities and/or those of lower socioeconomic status
Evidence Availability and Guidelines	Insufficient evidence Genetic exceptionalism that calls for randomized controlled trials for each use case Incorporation of pharmacogenetic testing into additional medical society guidelines
Regulatory Agency Oversight	Inconsistent guidance on test availability Inconsistent guidance on testing and dosing recommendations
Payer Coverage and Insurance	Sparse or inconsistent insurance coverage Variable evidentiary thresholds by payors
Test Availability and Utilization	Testing is not widely ordered Detection of rare variants in standard assays
EHR	EHR data structure and portability Comparative effectiveness and cost‐effectiveness data of competing CDS strategies
Education	Clinician knowledge Available post‐graduate training No gold‐standard resource for pharmacogenetic information in lay language for patients

CDS, clinical decision support; EHR, electronic health record.

Table 2.

Proposed solutions to the implementation of pharmacogenetics into care

Domain	Proposed solution
Equity and Inclusion	Increase genetic diversity among study populations in pharmacogenetics research. Implement pan‐ethnic pharmacogenetic testing. Inclusion of diverse patient groups will advance PGx evidence, improving care for all patients. Establish financing mechanisms for supporting testing in patients from marginalized communities and/or those of lower socioeconomic status. Improve integration of patient views and values regarding pharmacogenetic testing into medical care.
Standardization, Quality of Evidence, and Guidelines	Create consistent thresholds for clinical utility to guide appropriate testing. Incorporate different forms of evidence (e.g., mechanistic, pharmacokinetic) into gene validity and variant classification frameworks. Disseminate new research findings and clinical guidelines to frontline clinicians, regulatory agencies, payers, and society guideline committees. Include appropriate expertise on relevant society guidelines and statements.
Regulatory Agency Oversight	Ensure regulatory requirements for laboratories are clear and that resources to navigate requirements are available to support laboratories implementing PGx testing. Describe the evidence needed to modify the FDA's Table of Pharmacogenetic Associations and to recommend testing and/or genetics‐informed treatment in drug labeling.
Payer Coverage and Insurance	Develop a uniform set of health technology assessment guidelines for pharmacogenetic tests. Establish evidence thresholds and study designs to inform coverage decisions through collaboration among payers, clinicians, and researchers. Apply a transparent evidence review process to determine the clinical utility of pharmacogenetic testing.
Availability of Quality Pharmacogenetic Tests	Use multigene tests efficiently. Apply AMP recommendations for variant coverage in all clinical pharmacogenetic tests. Increase use of mechanisms to send specimen collection kits to patient homes for those engaging in telemedicine
EHRs and CDS	Increased adoption of data and application standards for genomic information to aid in data portability and EHR utilization. Conduct studies evaluating the comparative effectiveness and cost‐effectiveness data of competing CDS strategies
Education	Expand training pharmacogenetic opportunities at the university and post‐graduate levels Disseminate competency‐based pharmacogenetic education that meets the needs of busy healthcare professionals (continuing and workforce education). Develop standardized patient‐oriented educational materials

AMP, Association for Molecular Pathology; CDS, clinical decision support; EHR, electronic health record; FDA, Food and Drug Administration; LDT, laboratory‐developed test.

EQUITY AND INCLUSION

Barriers and recent progress

The field of pharmacogenetics is no exception to the lack of diversity, equity, and inclusion that plague biomedical research. Study participants from diverse backgrounds are underrepresented in pharmacogenetics research, spanning scientific discovery (of new drug‐gene targets) to clinical implementation studies. This inequity serves as a barrier to widespread implementation of pharmacogenetics because non‐inclusive clinical test design can introduce healthcare disparities, and it weakens the evidence for clinical validity and utility that is necessary for progress.

This inequity impacts everyone, not just the patients underrepresented in pharmacogenetic studies. For example, discrepant results between self‐identified Black and non‐Black participants in the Clarification of Optimal Anticoagulation through Genetics (COAG) trial amplified prior uncertainty surrounding the use of genotype‐guided warfarin dosing in routine care for everyone. ⁸ This discrepancy was due in part to the use of a genotype‐guided dosing algorithm that was not inclusive of alleles more common in individuals with predominantly sub‐Saharan African genetic ancestry. ⁹ This lack of favorable results for the genotype‐guided dosing arm in COAG has ultimately given insurance companies rationale to not cover routine pharmacogenetic testing of warfarin, thereby hindering pharmacogenetics uptake into practice. ¹⁰ Progress over the past decade to address this barrier has focused on the discovery and inclusion of pharmacogenetic alleles more common in underrepresented study populations. ¹¹ For instance, the most recent CPIC warfarin guideline now includes the alleles that were missing in COAG. ¹² Appropriately tailoring pharmacogenetics implementation to diverse populations is vital to promoting health equity. ¹³

Potential solutions

There are numerous potential solutions spanning the full spectrum of bench‐to‐bedside translation within the domain of diversity, equity, and inclusion that may promote pharmacogenetic implementation. ¹⁴ However, increasing genetic diversity among study populations in pharmacogenetics research may be the most impactful activity we can perform for addressing remaining barriers of clinical pharmacogenetic implementation related to diversity, equity, and inclusion. ¹⁵ This action will enhance the discovery of previously unknown pharmacogenetic alleles more common in understudied patient groups, improve the precision for clinical pharmacogenetic algorithms, and increase the likelihood of clinical trial success with pharmacogenetic‐guided treatment arms. The All of Us Research Program represents a monumental step forward as it has enrolled nearly a million participants, with the majority belonging to a group underrepresented in biomedical research. ¹⁶ , ¹⁷

Evidence demonstrating more universal benefits of pharmacogenetics across diverse patient groups will in turn compel multiple stakeholders to advance this therapeutic strategy. It will also facilitate the implementation of pan‐ethnic pharmacogenetic testing, which must expand beyond academic medical centers and large health systems into other community‐based and resource‐limited clinical settings such as those serving underserved communities. ¹⁸ Second, we need to establish financing mechanisms for supporting testing in patients from marginalized communities and/or those of lower socioeconomic status. Third, patient views and values regarding pharmacogenetic testing must be better integrated into medical care; innovative approaches like the use of CDS tools will promote such efforts. ¹⁹ Efforts to increase genetic diversity in research have expanded and these endeavors will undoubtedly benefit pharmacogenetics, among other specialties. ¹⁶ , ²⁰

STANDARDIZATION, QUALITY OF EVIDENCE, AND GUIDELINES

Barriers and recent progress

Historical barriers have included the lack of standardization to facilitate common nomenclature, the translation from genotype to phenotype to guidance for pharmacotherapy, and reaching an evidentiary threshold for implementation. ² Today, the “implementation” of pharmacogenetics in clinical practice can possess a different meaning depending on the audience. Some, such as many CPIC members, define implementation to denote clinical use of existing pharmacogenetic results. Others use a broader definition, encompassing the clinical use of existing and ordering of new pharmacogenetic tests. Progress has been more substantial under the narrower definition of implementation that focuses on using existing results and will be discussed first.

Persistent, incremental progress has led to significant improvements in standardization, quality of evidence, and guideline availability. CPIC has produced 28 clinical practice guidelines for over 30 genes and 150 drugs, with additional guidelines available from the Dutch Pharmacogenetics Working Group (DPWG), Canadian Pharmacogenomics Network for Drug Safety (CPNDS), and the French National Network of Pharmacogenetics (RNPGx). ²¹ Additional standardization efforts enable more efficient evidence generation and clinical application, including standardizing terminology (CPIC), ²² allelic variant designations (PharmVar ²³ ), guidance on recommended variant alleles to test clinically (Association for Molecular Pathology [AMP] Pharmacogenomics Working Group), genotype to phenotype translations (CPIC), and clinical recommendations (CPIC). The quality of evidence has also improved, as evidenced by recent landmark studies demonstrating that pharmacogenetics‐guided therapy improves clinical outcomes. ²⁴ , ²⁵ , ²⁶ , ²⁷ This has led to a large volume of standardized guidance and clinical workflows to ease implementation for a large number of drug‐gene pairs.

Remaining barriers and potential solutions

Key remaining barriers still limit broader implementation, including the clinical use of existing and ordering of new pharmacogenetic tests. These barriers include (1) genetic exceptionalism that calls for randomized controlled trials for each use case; and (2) dissemination and incorporation of current evidence into medical society guidelines. Criticism of pharmacogenetic implementation often comes in the form of genetic exceptionalism, requirements of additional considerations, or higher evidentiary thresholds prior to clinical use compared to similar non‐genetic data. ²⁸ Mechanistic and pharmacokinetic data are routinely extrapolated to inform care (e.g., drug–drug interactions, renal dose adjustments) despite the lack of confirmatory clinical trials comparing the intervention (e.g., drug interaction management, creatine clearance‐guided dosing) to a control arm. ²⁹ In addition, for some situations it may be considered unethical to conduct a traditional clinical trial (e.g., prescribe standard dose fluoropyrimidine to a DPYD intermediate or poor metabolizer). ³⁰

Although there has been considerable progress to support the use of existing pharmacogenetic results, one challenging barrier is deciding whether to order a test. ³¹ Elements of the genetic exceptionalism argument are still relevant here; however, further costs must be justified for ordering a new test compared to repurposing an existing result. Unfortunately, inconsistent thresholds for clinical utility have clouded guidance on when tests should be ordered. Inconsistent recommendations for pharmacogenetic testing between and within organizations are further described in a review by the Standardizing Laboratory Practices in Pharmacogenomics (STRIPE) Collaborative Community Study Designs Task Force. ³²

Some clinical practice guidelines recommend ordering or considering pharmacogenetic testing to guide pharmacotherapy (e.g., National Comprehensive Cancer Network [NCCN] guidelines for adult cancer pain, colon cancer, and acute lymphoblastic leukemia; American College of Rheumatology Guideline for gout; Department of Health and Human Services guidelines for HIV). ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ However, many guidelines do not address or provide recommendations regarding pharmacogenetic testing (e.g., NCCN breast cancer American Psychological Association [APA]), or recommend against testing (e.g., American College of Cardiology/American Heart Association [ACC/AHA], ). ³⁸ , ³⁹ There is a need to disseminate new research findings and guidelines to frontline clinicians, regulatory agencies, payers, and society guideline committees. Additionally, it is necessary that appropriate expertise is present on relevant society guidelines and statements. ⁴⁰

Lastly, continued research efforts to expand evidence are necessary to continue progress. Each phase of the research continuum is responsible for advancing the standardization, quality of evidence, and guideline incorporation of pharmacogenetics. As progress continues, more emphasis will be placed on implementation studies to better understand effective implementation approaches and optimization. Progress may also lead to a need to expand standardization (e.g., expand set of genes that have been standardized, substrate‐specific phenotypes). ⁴¹

REGULATORY AGENCY OVERSIGHT

Barriers and recent progress

Another important barrier to progress in implementing pharmacogenetics is an inconsistent regulatory landscape, both as it pertains to the US Food and Drug Administration's (FDA's) role in overseeing pharmacogenetic test availability and recommending pharmacogenetic testing and pharmacogenetics‐informed treatment within drug labels. In both areas, there has been progress, but more clarity and consistency from the FDA would be instrumental in advancing pharmacogenetics implementation.

Laboratory‐developed tests (LDTs), including many pharmacogenetic tests, are by definition not FDA‐approved. LDTs are developed and performed in a single laboratory and are under the purview of the Centers for Medicare and Medicaid Services (CMS) Clinical Laboratory Improvement Amendments of 1988 (CLIA). Although the FDA has a regulatory role over testing in the United States through the Food, Drug, and Cosmetic Act, the FDA has traditionally used its enforcement discretion and not applied this authority to LDTs. While there are a few FDA‐cleared pharmacogenetic tests, most pharmacogenetic tests fall into the category of LDTs. However, in late 2018, the FDA asserted its regulatory authority over pharmacogenetic LDTs against several clinical laboratories, including the Inova Healthcare system, stating that it was marketing clinical utility claims not reviewed by the FDA. ⁴² In 2024, the FDA finalized the rule to ensure the safety and effectiveness of LDTs by adding language to 21CFR§809.3 (in italics below).

“In vitro diagnostic products are those reagents, instruments, and systems intended for use in the diagnosis of disease or other conditions, including a determination of the state of health, in order to cure, mitigate, treat, or prevent disease or its sequelae. Such products are intended for use in the collection, preparation, and examination of specimens taken from the human body. These products are devices as defined in section 201(h) ¹ of the Federal Food, Drug, and Cosmetic Act (the act) and may also be biological products subject to section 351 of the Public Health Service Act, including when the manufacturer of these products is a laboratory.” ⁴³ There are at least two lawsuits challenging the FDA's authority, and these will be determined by the courts of law. At this time, laboratories offering LDTs must continue to comply with dual federal oversight, CMS/CLIA, and FDA.

The second area of regulatory inconsistency is in the FDA's function in describing how pharmacogenomics should be used to ensure medication safety and effectiveness. The FDA can recommend or require pharmacogenetic testing and provide dosing or drug selection recommendations in drug labeling for patients with available genetic information. The FDA maintains a Table of Pharmacogenomic Biomarkers in Drug Labeling, which aggregates the type and location of all pharmacogenetic information included in drug labels. ⁴⁴ They also maintain a Table of Pharmacogenetic Associations for which they believe there is sufficient scientific evidence that patients with certain variants are likely to have differential therapeutic effects. ⁴⁵ The cause of differences between the tables and drug labels is not clear, but they are presumably related to regulatory processes needed to update each of them. ⁴⁶ Similarly, in some cases, there are differences with CPIC guidelines that may cause confusion among users seeking to implement pharmacogenetics. Differences between the FDA and CPIC may occur due to the aforementioned reasons, underlying data sources (e.g., FDA access to unpublished data), timelines of updates, or different approaches, perspectives, and purposes for reviewing pharmacogenetic data.

Similarly, the FDA has not provided full transparency and consistency in the testing and dosing guidance recommendations in labeling, as evidenced by requests for further clarity in response to FDA decisions around pharmacogenetic testing for antidepressants and fluoropyrimidines. ³⁰ , ⁴⁷ , ⁴⁸ , ⁴⁹ , ⁵⁰ Specifically, there is a lack of transparency and consistency on the evidence necessary to demonstrate that (1) genetics is associated with a clinical outcome (i.e., clinical validity); (2) genetics should be used to inform dosing (i.e., actionability); (3) when pharmacogenetic testing should be conducted (i.e., clinical utility). The last evidentiary standard is particularly important as a decision to include a testing recommendation in drug labeling is often used as the basis for clinical guidelines to recommend testing and insurance companies to reimburse for testing.

Remaining barriers and potential solutions

FDA's recent decision to claim enforcement over LDTs may have, in the short term, made the regulatory environment more uncertain but could eventually provide greater clarity. The FDA's final rule describes its intended approach to overseeing LDTs, and the results of legal challenges to that authority could eventually provide clarity regarding the role (or lack thereof) of the FDA in regulating pharmacogenetic tests. The FDA could also provide clear guidance on the evidence needed to recommend specific dose adjustments in patients with known genetic results for updates to the Table of Pharmacogenetic Associations, which should be consistent with the principles of exposure matching the FDA uses to recommend dose adjustment for organ insufficiency and other intrinsic factors. ⁴⁹ In the past, the FDA reported a Pharmacogenetic Pyramid as a framework to support regulatory decisions related to pharmacogenetics, but recent decisions have appeared to not rely on this framework. ³⁰ In regard to recommending pharmacogenetic testing within drug labels, consensus may be achieved through the FDA's involvement in the Standardizing Laboratory Practices in Pharmacogenetics (STRIPE) collaborative community, which is developing evidentiary standards for pharmacogenetic testing recommendations. ⁵¹ , ⁵²

PAYER COVERAGE AND INSURANCE

Barriers and recent progress

One of the main barriers to the widespread adoption of pharmacogenetic testing has consistently been limited and variable payer coverage. However, given the numerous studies conducted to address clinical utility evidence gaps (see the section entitled Standardization, quality of evidence, and guidelines), payers are starting to recognize the utility of pharmacogenetic testing, with a measurable increase in coverage over time. ⁵³ , ⁵⁴ , ⁵⁵ However, major gaps in coverage remain, and there is significant inconsistency across insurers, tests, and regions.

Medicare coverage of pharmacogenetic testing is now available across 40 states for more than 100 medications, and coverage is now being considered for the remaining 10 states (Box 1). ⁵³ , ⁵⁴ , ⁵⁶ , ⁵⁷ , ⁵⁸ , ⁵⁹ , ⁶⁰ CMS coverage decisions can influence private payers' coverage decisions. ⁶¹ A recent analysis through STRIPE identified a widespread coverage by private payers for pharmacogenomic multigene panels, with at least one laboratory benefit manager (Avalon) also recommending coverage of many pharmacogenomic tests. ⁶² Laboratory benefit managers are third parties that support insurers with recommended coverage policies, among other services, and therefore can affect many insurance policies. Many policies with coverage of pharmacogenetic tests cite CPIC guidelines and the FDA Table of Pharmacogenetic Associations as evidence for coverage. A recent analysis through STRIPE identified a median of 10 drug‐gene pairs (range 4–65) covered across 12 publicly available medical policies from top insurers and laboratory benefit managers. Drug‐gene pairs with CPIC guidelines and those mentioned in the FDA Table of Pharmacogenetic Associations tended to have wider coverage across policies on average but were still covered in less than half of all policies. The most common drug‐gene pairs covered were abacavir‐HLA‐B (n = 12), carbamazepine‐HLA‐B (n = 12), clopidogrel‐CYP2C19 (n = 11), and thiopurines‐TPMT (n = 11). Several commonly prescribed medication classes had lower rates of coverage (e.g., citalopram‐CYP2C19 [n = 4], tramadol‐CYP2D6 [n = 4], pantoprazole‐CYP2C19 [n = 4], simvastatin‐SLCO1B1 [n = 2]). That study was restricted to specific drug‐gene pairs based on determinations by the CPIC and FDA. Moving beyond specific drug‐gene pairs (e.g., polygenic risk scores, preemptive testing) attracts more skepticism from payers. ⁶¹

Box 1. Medicare LCDs for coverage of pharmacogenetic tests.

Apply to CPIC gene‐drug pairs with a final assignment of Level A or B Includes both single gene and multigene tests (if testing for more than one gene is clinically indicated) LCDs are active for 40/50 states A proposed LCD is being considered for the remaining 10 states

CPIC, Clinical Pharmacogenetics Implementation Consortium; LCD, Local Coverage Determination.

Remaining barriers and potential solutions

Several barriers to increased and more consistent payer coverage remain. The lack of coverage consistency is particularly notable for patients with Medicare and Medicaid, as coverage varies for patients based on the state in which they reside. The inconsistencies in CMS policy can further factor the ability to effectively implement pharmacogenomics and may exacerbate inequities. These coverage barriers may relate to payers' perceptions of the required evidence. Keeling et al. interviewed payer representatives and identified heterogeneity in the required evidence for coverage (e.g., randomized controlled trial vs. other study designs). ⁶¹ The recent STRIPE analysis also demonstrated significant variability in the evidence cited to support coverage or lack of coverage within medical policies. ⁶² Society guidelines were the most cited type of evidence, followed by RCTs and prospective trials. Cost‐effectiveness studies were infrequently cited, despite the fact that a recent systematic review found the majority of cost‐effectiveness studies have found pharmacogenetics to be cost‐effective or cost‐saving. ⁶³ In some situations, the same references were cited in different medical policies with different coverage determinations (e.g., antidepressant RCTs cited to support coverage in one policy versus lack of coverage in another policy). This suggests that insurers have variable interpretations of the adequacy of available evidence to support coverage of a pharmacogenetic test. The variability in payer coverage may be partially explained by the previously described lack of recommendations on whether to order testing from clinical practice guidelines and FDA guidance. ⁶¹

Lack of consistent coverage across medical policies likely results in unequal care and dissuades providers from ordering pharmacogenetic tests if they experience insurance denials or burdensome appeals processes when ordering a test. As such, there is a critical need to develop a uniform set of health technology assessment guidelines for pharmacogenetic tests. Payers should collaborate with the clinical and research community to establish evidence thresholds and study designs to inform coverage decisions (e.g., RCTs, observational studies, real‐world data, cost analyses, etc.). Coverage of pharmacogenetic testing should occur at a minimum when it is supported by medical and scientific evidence, such as labeled indications for an FDA‐approved or ‐cleared test; indicated tests for an FDA‐approved drug; warnings and precautions on many FDA‐approved drug labels; CMS national coverage determinations or Medicare and Medicaid Administrative Contractor local coverage determinations (LCDs); or nationally recognized clinical practice guidelines and consensus statements. Existing guidelines like CPIC and information from the FDA, like the Table of Pharmacogenetic Associations, are also critical resources that can be used to support coverage decisions. Lastly, there must be transparency in the evidence review process to determine the clinical utility of pharmacogenetic testing, which can inform the research community on the types of studies that would contribute most meaningfully to coverage decisions.

AVAILABILITY OF QUALITY PHARMACOGENETIC TESTS

Barriers and recent progress

For pharmacogenetics to become a standard component of medical care for all patients, clinical laboratory tests must be widely available, affordable, have appropriate turnaround times, and prescribers must be aware of the existence, benefits, and limitations of pharmacogenetic tests. Historically, pharmacogenetic testing was offered by a small number of specialized laboratories; however, the voluntary Genetic Testing Registry (GTR), which is a database hosted by the US National Institutes of Health that lists laboratories and their tests based on voluntary self‐report (https://www.ncbi.nlm.nih.gov/gtr; last accessed 12/20/2024), currently lists 73 available clinical pharmacogenetic tests (with varying combinations of genes available) from 23 CLIA‐certified clinical laboratories in the United States, including academic and commercial laboratories, as well as specialized pharmacogenetics‐only laboratories. While many smaller clinics and hospitals throughout the United States are unlikely to have their own pharmacogenetics laboratory, they may order directly from a specialized commercial laboratory, or their standard reference laboratory, likely either performing pharmacogenetic testing or outsourcing this testing to another laboratory. ⁶⁴ As such, in 2025, pharmacogenetic tests may be readily available for nearly any patient nationwide. Several considerations are important when deciding on a laboratory for pharmacogenetics and are discussed in more detail elsewhere. ⁶⁵ Select laboratories offer additional services to improve access to testing. In the telehealth setting, specimen collection challenges may be mitigated by the shipment of collection kits (e.g., buccal, saliva) to patient homes. Some laboratories may alleviate financial barriers by offering billing services to patients (e.g., cash billing, insurer billing, financial assistance programs). Importantly, the ability to order a test does not guarantee that the patient has access to it. The cost may be prohibitive, or their healthcare provider may not be aware of the available pharmacogenetic tests or when to use them.

Historically, when pharmacogenetic tests were available, they were often expensive and not covered by insurers, leading to a financial barrier for many patients. With increases in multiplexing capabilities, the costs of tests have decreased such that a multigene panel may now cost hundreds of dollars rather than thousands. ⁶⁶ In addition, insurers are beginning to cover testing for some applications (see above). Some direct‐to‐consumer or consumer‐initiated pharmacogenetic tests have emerged that allow consumers to order their own test and typically feature a low price point. However, these tests may be more limited in the variants included and, in some situations, can require retesting or may not represent an optimal test for medical decision‐making. ⁶⁷

Scenarios that require rapid decision‐making, such as whether to use clopidogrel in the setting of acute myocardial infarction, may result in an additional barrier to the adoption of pharmacogenetics. While the ultimate goal is for pharmacogenetic testing to be performed preemptively so that the result will already be available in the EHR at the time of prescribing, today, the majority of testing is performed ‘reactively’ when there is an indication for a medication that may be guided by pharmacogenetics. ⁶⁸ , ⁶⁹ As a result, the prescriber may have already decided on medication and dosing prior to receiving results, and, in some cases, the patient may have already experienced an adverse reaction. Until preemptive pharmacogenetic testing becomes more widely adopted, another potential solution involves tests with a rapid turnaround time. For example, a CYP2C19 testing platform manufactured by Genomadix was recently granted FDA clearance as a moderate‐complexity test. ⁷⁰ This test or similar tests have been used in multiple clinical trials and have a turnaround time of <1 hour. Limitations of this approach include allelic coverage limited to the current Association for Molecular Pathology (AMP) Tier 1 alleles, clinical workflows to integrate testing by clinical staff, and assay‐specific buccal swabs that require assay‐specific training. Another example is the rapid turnaround time MT‐RNR1 m.1555A>G point‐of‐care genotyping device that was recently used prospectively in neonatal intensive care units in the United Kingdom to inform antibiotic prescribing. ⁷¹

Remaining barriers and potential solutions

Currently, most available pharmacogenetic tests target specific variants within specific genes. This enables anticipated clinical interpretation, faster turnaround times, and more cost‐efficient testing. However, these tests are inherently limited in their ability to detect rare variants, determine the phase when multiple clinically relevant variants are detected, and may not capture pan‐ethnic germline variation. For example, CPIC guidelines recommend reduced fluoropyrimidine dosing in those with reduced or no dihydropyrimidine dehydrogenase (DPD) function to prevent fluoropyrimidine toxicity. However, there are uncommonly tested and rare DPYD variants that have been associated with severe or fatal toxicity. ⁷² An analysis of the RIGHT10K study run by the Mayo Clinic identified that 28% of participants possessed a potentially relevant rare variant captured on sequencing that would have been missed by standard targeted genotyping. ⁷³ The AMP consensus recommendations for genotyping allele selection provide a foundation by recommending a minimum panel of variants (Tier 1) and an extended panel of variants (Tier 2) for targeted genotyping tests. Further adoption of next‐generation sequencing (NGS) to go beyond testing common variants may offer a solution, but challenges in cost and complexity remain. ⁷⁴

Truly preemptive clinical pharmacogenetic testing in the US healthcare system as standard of care is currently not feasible. Instead, preemptive testing may be available through an IRB‐approved supportive research protocol with associated research funding. One incremental method to address the difficulty of obtaining preemptive results is the efficient use of panel tests. A single gene or multigene pharmacogenetic test may be ordered for an initial indication but repurposed to support additional drug‐gene pairs over time (e.g., 14% had ≥ additional 1 drug‐gene pair in 12 weeks). ²⁴ Indeed, this is inherent to the design of CPIC guidelines to help clinicians understand how available pharmacogenetic results can be applied to guide medication therapy. ² , ⁴ Tools intrinsic to the reuse of pharmacogenetic results include the EHR and CDS systems.

EHRs AND CDS

Barriers and recent progress

Ideally, a CDS system provides targeted pharmacogenetics information at the optimal time to improve patient care. The National Human Genome Research Institute (NHGRI) advocates for CDS as a critical method for integrating genomic knowledge into medical practice. ⁷⁵ CDS systems must have the necessary components to function, a sustainable cost‐value equation, and a design that meets the end user's needs. Yet, numerous barriers have impeded the advancement of pharmacogenetic CDS in each of these domains.

The greatest progress has been made in creating the necessary components for pharmacogenetic CDS. These include standardized terms and tools for interoperability. CPIC created a set of standardized phenotype terms for drug‐metabolizing enzymes and other pharmacogenes. ²² These terms are supported with Logical Observation Identifiers, Names, and Codes (LOINC) to help communication across systems. Notably, EHR vendors such as Epic have also increased focus on pharmacogenetic CDS with active expert working groups and now provide CPIC pharmacogenetic CDS content and rules as core offerings in their EHR systems. Fast Health Interoperability Resources (FHIR), Substitutable Medical Applications, and Reusable Technologies (SMART) on FHIR, CDS Hooks, and Infobuttons are all standards that have become common ways for third‐party applications to interface with data within EHRs and trigger real‐time CDS securely. ⁷⁶

Remaining barriers and potential solutions

Several barriers remain, including availability of results as structured data, result portability, comparative effectiveness data of competing CDS strategies, and cost‐effectiveness of CDS strategies. Although the developments discussed above have created an environment where CDS can be shared across EHRs, standard architectures—such as the lack of routinely available, structured, discrete data—to manage pharmacogenetics information is lacking. The lack of portability is particularly challenging between pharmacogenetic laboratories and health systems where data are often exchanged as unstructured data (e.g., PDFs) rather than structured data through automated interfaces. The use of star allele nomenclature creates a unique challenge to portability of PGx data when coupled with the lack of testing standardization. Incorporating non‐genetic clinical data alongside pharmacogenetic data may also improve accuracy and usability in patient care. ⁷⁷ Developing, integrating, and maintaining pharmacogenetics CDS systems also require non‐trivial investments in information technology, which may further limit uptake in resource‐limited settings. ⁷⁸ To balance out this cost, the value of pharmacogenetics must be well‐defined through clinical use cases. However, this literature is limited for this type of analysis. ⁷⁹ Absent a sustainable cost‐value equation with clearly defined clinical use cases, pharmacogenetic CDS will not achieve widespread adoption.

In the past decade, multiple groups have published examples of successful implementation of pharmacogenetic CDS in clinical practice. ⁸⁰ A systematic review found interruptive alerts were the most commonly reported CDS tool. These interruptive alerts possessed a wide range of effectiveness, as 12–73% of alerts had a clinical action that corresponded with the alert's recommendation. There is no gold standard for the set or design of pharmacogenetic CDS tools, and there is a notable gap in literature in this space. This includes what tools need to be created, where in the care journey the CDS should support decision‐making, and what non‐genomic features are relevant at each decision point.

To reach the full potential of pharmacogenetic CDS for use by clinicians and patients will require studies from the fields of implementation science, human factors engineering, artificial intelligence, and more. ⁸¹ Human factors engineering uses knowledge about human capabilities and limitations to design safe, effective, and efficient work systems, which may be applied to develop and evaluate CDS for pharmacogenomics. Rapid advancements in natural language processing and artificial intelligence mean the gold standard is constantly evolving. Like other areas of medicine, artificial intelligence is poised to radically change current practices in pharmacogenetics. ⁸²

EDUCATION

Barriers and recent Progress

Knowledge gaps among healthcare professionals are a well‐known barrier that hinders broader pharmacogenetics implementation. ⁸³ Compounding these knowledge gaps, pharmacogenetics clinical experts are largely concentrated at large academic centers where the focus has been on educating local staff, and little broad‐based education has occurred in community healthcare settings. Surveys conducted over the past decade have shown that providers have historically had poor confidence in their ability to integrate pharmacogenetics into practice despite their enthusiasm and professional responsibility to do so. ⁸⁴ , ⁸⁵ , ⁸⁶

However, significant advancements have eroded many of the traditional barriers to pharmacogenetics education so that dissemination of solutions may now be the key challenge. Genomics competency standards have been developed for all healthcare professionals and are promulgated by NHGRI's Genomic Education Center (GenomeEd, https://www.genome.gov/GenomeEd). For pharmacists, these competencies and Doctor of Pharmacy curricular standards have bolstered pharmacogenetics education significantly in pharmacy schools. ⁸⁷ , ⁸⁸ For existing practitioners, continuing education is also available. ⁸⁹ For example, the American Society for Health‐System Pharmacists—American Medical Association (ASHP‐AMA) pharmacogenetics produced Pharmacogenomics Virtual Summit Series in 2021. Recently, the Inter‐Society Coordinating Committee for Practitioner Education in Genomics released the free, peer‐reviewed Pharmacogenomics Learning Series (https://genome.gov/PGx‐Learning‐Series). Pharmacogenetics certification programs and graduate degrees are now offered by several universities and professional societies in virtual formats. Within institutions offering clinical testing, pharmacist consultative services have also been established, with EHR‐based clinical decision support increasingly available to support prescribing at the point of care.

A lack of understanding is also a barrier for patients. Patients may not be aware that pharmacogenetics testing and services are available, lack access to lay language explanations of test results, or have inadequate access to knowledgeable healthcare professionals. Fortunately, solutions are emerging. Pharmacogenetic testing is increasingly available and an FDA‐authorized, direct‐to‐consumer product (23andMe) is now available in the US. ⁶⁷ Institutions such as the University of Pittsburgh Medical Center (UPMC), Endeavor Health (formerly Northshore University Health System), University of Florida Health, University of Colorado Health (UCHealth), Denver Health, MedStar Health, Mayo Clinic, and Cleveland Clinic, among others, have established outpatient and telehealth clinics or consultative services that offer pharmacogenetic testing with interpretations by pharmacists. Further, when a test is performed, the results are increasingly available in EHR patient portals following new 21st Century Cures Act mandates that require immediate electronic release of medical results. ⁹⁰ National research initiatives such as the All of Us Research Program and large institutional biobanking programs also increase awareness of, and access to, pharmacogenetic research, further increasing awareness. Lastly, tools to support patient awareness of results (mobile applications, quick response [QR] code cards) have grown. For example, a safety‐code card was developed for the PREPARE trial that found pharmacogenetic testing was associated with a 30% reduction in adverse events. ²⁴ , ⁹¹

Remaining barriers and potential solutions

The educational community now has an ongoing focus to increase the accessibility, fidelity, and broader dissemination of training and education. While pharmacogenomics is in pharmacy education standards, education is not required in medical school, and few plan to increase instruction on pharmacogenomics in the near future. ⁹² The limited post‐graduate training programs (e.g. advanced practice professional experiences, residency, fellowships) need to be expanded. ⁸³ , ⁹³ At the point of care, EHR clinical decision support tools need to further integrate clinical factors with genetics, as described in the previous section. ⁸⁰ Further, there is a push to make education for busy practitioners more accessible, higher fidelity, and more tailored to real‐world clinical roles and workflows. Online solutions have increased, which allow for a self‐paced, competency‐based approach with the use of real data/cases that award credentials for achieving clinical competency. Groups like NIH's Inter‐Society Coordinating Committee for Practitioner Education in Genomics and the PGRN disseminate educational materials in partnership with other professional societies to their audiences at conferences and online.

Persistent challenges also remain for patient education. Some concepts may be underappreciated by patients and frontline clinicians, including the identification of results that require confirmation with clinical‐grade testing prior to use (e.g., some research or direct‐to‐consumer tests ⁶⁷ ) and the role of phenoconversion due to drug–drug interactions. A key challenge is the lack of a gold‐standard resource for effective pharmacogenetic information in lay language. Emerging patient‐focused solutions include patient‐facing clinical decision support, chatbots, and other materials to increase the accessibility of pharmacogenetic information. ⁸¹ , ⁹⁴ The degree to which patients need to be informed of varying details is debated among experts. However, these tools should be paired with a clinical consultative service to maximize impact and prevent inappropriate prescribing and deprescribing.

CURRENT INITIATIVES AND FUTURE WORK

The PGRN and its membership are conducting a wide array of ongoing clinical and research programs to advance the field of pharmacogenetics implementation. Highlighted are two examples of programs led by PGRN members, which use different approaches to advance clinical pharmacogenetics implementation. More broadly, the PGRN seeks to catalyze and lead research in precision medicine for the discovery and translation of genomic variation influencing therapeutic and adverse drug effects. This manuscript is limited by the US‐centric approach, and a subsequent manuscript from the PGRN Publications Committee will focus on more global endeavors in pharmacogenetics implementation.

Example 1 – Advancements in pharmacogenetics have led to large‐scale initiatives that incorporate novel research and clinical implementation strategies. Population‐scale biobanks are one avenue to advance pharmacogenetic research discoveries and, in some cases, return actionable findings back to participants for use in their health care. An example of this approach is the biobank at the Colorado Center for Personalized Medicine (CCPM Biobank), which was established at the University of Colorado Anschutz Medical Campus in partnership with UCHealth. ⁹⁵ , ⁹⁶ For this initiative, UCHealth adult patients are invited to enroll in the biobank via an electronic self‐consent model through their online patient portal. A single consent form authorizes consent for research, participant recontact, use in industry partnerships, and return of clinical genetic test results to the EHR. A key facilitator of this initiative is the CCPM Biobank Laboratory, which is College of American Pathologists (CAP)‐accredited and CLIA‐certified for high complexity testing. Clinical pharmacogenetic test results generated by the CCPM Biobank Laboratory are preemptively returned to the UCHealth EHR and used to trigger CDS tools for more than 50 medications. ⁹⁷ As of March 2025, over 250,000 participants have enrolled in the CCPM Biobank and more than 160,000 have provided a biospecimen for genotyping. Nearly 800,000 clinical pharmacogenetic results have been returned to the EHR for more than 74,000 participants, demonstrating how biobanking initiatives can be leveraged to implement precision medicine at scale across a health system. ⁹⁵ Other institutions have established processes to use non‐CLIA institutional genetic data repositories for efficient pharmacogenetics implementation; patients who carry actionable genotypes are sent for confirmatory testing when they are scheduled to receive the relevant drug. ⁹⁸

Example 2 – The Veterans Affairs (VA) began implementing pharmacogenetic testing in 2019 through a large donation to VA from Sanford Health (Sioux Falls, SD) through the Pharmacogenomics Testing for Veterans (PHASER) program. ⁹⁹ The initial goals of the program were to make preemptive, panel‐based pharmacogenetic testing available and integrate pharmacogenetic results at approximately 30 VA health care systems. Since that time, the scope and shape of VAs implementation have shifted over time as part of a learning healthcare system‐based approach. ¹⁰⁰ The VA has now formalized a National Pharmacogenomics Program (NPP) that will oversee the operational elements of expanding pharmacogenetics access, care, and policies across the nearly 150 VA healthcare facilities. In addition to providing access to high‐quality, panel‐based pharmacogenetic testing, the NPP is investing has hired over 100 clinical pharmacy practitioners at the local health system, regional network, and national levels through the Expanding Clinical Pharmacist Practitioners in Pharmacogenomics (EXCLAIM) program. EXCLAIM pharmacists are building upon the foundation put in place by the PHASER program in (1) transitioning from a preemptive to reactive strategy by focusing on implementation and evaluation in high‐impact areas (e.g., cardiac catheterization laboratory, specific chemotherapeutics, and depression) (2) educating the workforce of VA clinical pharmacists (N > 5,000 nationwide) on pharmacogenetics and incorporating it into their clinical practice and (3) establishing clinical pharmacogenetic services (consultative as well as population health management) for Veterans who have undergone pharmacogenetic testing. Over the coming years, the VA NPP will continue its transition toward reactive pharmacogenetic testing and identifying patient populations where there is a high level of evidence for clinical utility based on internal VA criteria; continue the learning process as they evaluate the effectiveness and return on investment of EXCLAIM pharmacists in their objectives around pharmacogenetics education, implementation, and clinical care; and begin to plan for secondary research regarding the clinical and economic outcomes of the nearly 100,000 Veterans expected to complete pharmacogenetic testing in the coming years.

CONCLUSIONS

The review has described barriers, improvements, remaining barriers, and potential solutions for seven domains of pharmacogenetics implementation: equity and inclusion; guidelines and supporting evidence; regulatory agency oversight; payer coverage and insurance; availability of quality pharmacogenetic tests; EHR considerations; and provider and patient education. Recent years have seen the erosion of barriers that previously seemed insurmountable (e.g., Medicare coverage for testing, standardized process for genotype to phenotype to therapeutic action). Progress has shined a light on remaining barriers that demand increased attention (e.g., incorporating diversity, equity, and inclusion principles into pharmacogenetic efforts; inclusion of pharmacogenetics in society guidelines; clinician and patient education). Patients will receive improved care as interdisciplinary champions continue to overcome barriers and increase utilization of pharmacogenetics in clinical care.

FUNDING

No funding was received for this work.

CONFLICT OF INTEREST

D.M.S., C.L.A., B.D., P.E.E., D.L.H., A.A.M., L.S., S.A.S., E.L.W., H.M.D, A.O.‐O., V.M.P., J.N.P., and M.W.‐C. are PGRN members. D.M.S. reports grants (to institution) from Kailos Genetics, Inc. V.M.P. is an employee of Agena Bioscience. P.D. reports receiving consulting fees for projects unrelated to pharmacogenomics. J.N.P. has served as a paid consultant for VieCure and Clarified Precision Medicine and speaker for Illumina, Inc. H.M.D. has served as a paid consultant for Veritas Intercontinental. All other authors declared no competing interests for this work.