Pharmacogenomics (PGx) is becoming a clinical reality. As testing costs decline and trial evidence accumulates, hospitals must address key questions regarding governance, system integration and equitable access. In his latest commentary, Dr João Gonçalves PhD proposes a pharmacy-led PGx model centred on a targeted set of high-impact drug-gene pairs, providing practical suggestions on service design, decision support, equity and outcome measurement.
For years, pharmacogenomics (PGx) has remained confined to pilot studies and promising case studies having limited generalisability. That phase is coming to an end as we now have international, real-world evidence that panel-based PGx reduces harm in everyday care.
In the PREPARE trial,1 spanning 18 hospitals across seven nations, genotype-guided prescribing with a 12-gene panel lowered clinically relevant adverse drug events by 30%.
Regulators have also moved in step: since 2020, the European Medicines Agency (EMA) has mandated the assessment of dihydropyrimidine dehydrogenase (DPD) function before fluoropyrimidine therapy.2
Meanwhile, the Clinical Pharmacogenetics Implementation Consortium (CPIC) and the Dutch Pharmacogenetics Working Group (DPWG) maintain continuously updated, drug-specific guidance.3–5 Collectively, these advances suggest a cultural shift from ‘could consider testing’ to ‘should have a plan’.
Why pharmacists should lead on PGx initiatives
For PGx to move from the genomics silo to the prescribing interface, pharmacy should lead the process. Pharmacists already manage formulary strategy, order sets, dosing calculators, alert governance and medication safety policies.
Experience across health systems shows that static laboratory reports are giving way to discrete, computable results embedded within electronic health records (EHRs). Instead of hunting through attachments and paperwork, clinicians now receive timely, actionable prescribing advice.6
The practical imperative is clear: do not build a generic genomics programme. The service should include a defined list of high-impact drug–gene pairs; clinical decision support and order set content; concise, usable consultation documentation and notes; and metrics that demonstrate value. This will ensure the service is reliable, embedded and predictably consistent.
Embed genotype data into prescribing workflows
To make PGx work in clinical practice, treat genetic data like any other lab result: make the data transparent, computable and available from the start when prescribing.
Utilise structured formats, such as HL7 FHIR Genomics, to store both genotype and phenotype data. This allows decision support systems to recognise relevant genetic traits (e.g. a CYP2C19 intermediate metaboliser) even years later when a related drug (such as clopidogrel) is prescribed.7
Use tiered clinical decision support, including high-risk alerts for serious or first-time issues (e.g. abacavir hypersensitivity with HLA-B*57 01) and inline suggestions for dose adjustments (e.g. tacrolimus for CYP3A5 expressers, thiopurines based on TPMT/NUDT15).8
Some EHRs can show genomic indicators directly in prescribing workflows, making decision support seamless.6 If genetic results are hidden away in scanned PDFs, the system is likely to fail as it will not be timely, followed or effective.
Governance underpins credibility in PGx
A PGx subcommittee established under a hospital’s Medicines or Drug and Therapeutics Committee should be chaired by a pharmacy representative and include multidisciplinary representatives from pathology or genomics, IT, clinical specialties, ethics and nursing.
The core responsibilities should include:
- Maintaining a PGx formulary defining in-scope drug-gene pairs and required actions per regulatory-aligned guidance
- Approving test menus and laboratories in accordance with EU IVDR requirements, maintaining ISO 15189 accreditation, participating in recognised external quality assessment schemes, incorporating ancestry-relevant alleles, and ensuring lifetime discrete phenotype reporting
- Owning and maintaining clinical decision support content and guardrails in the EHR
- Monitoring safety, adoption and equity regularly, with dashboards shared across directorates.
This structure aligns clinical policy with European regulatory and funding frameworks, including EMA safety communications, national health technology assessments and reimbursement positions, as well as hospital procurement requirements.
Governance must also satisfy GDPR obligations from the outset, providing a lawful basis for processing genetic data, implementing data minimisation, ensuring role-based access and providing multilingual patient materials.
Designing for interoperability with EU infrastructures such as EEHRxF and MyHealth@EU and the upcoming European Health Data Space will ensure long-term resilience.9 Done correctly, this ensures alerts are credible, testing is auditable and services are sustainable across diverse payer and regulatory contexts.
Funding and justification
Executive support could be best secured by adhering to two core principles.
First, funding follows clinical actionability. When testing links directly to specific drug-gene pairs and codified indications that align with guidance and regulatory standards, reimbursement routes become clearer. Embed PGx within recognised care pathways across oncology, neurology, transplantation and cardiology.
A strong business case references national health technology assessment or commissioning bodies and shows how testing operates within established reimbursement structures. Aligning the proposal with their terminology on clinical utility and patient safety is far more persuasive than broad ‘genomics innovation’ arguments.
Second, panel testing is economically justified when results are stored once and reused. Most patients will face multiple PGx-relevant prescribing decisions over their lifetime, and the discrete storage of phenotypes within the EHR substantially reduces the marginal cost per decision.
European commissioning pilots increasingly demonstrate favourable cost-effectiveness, particularly when accounting for avoided hospital admissions, fewer severe adverse drug reactions and faster achievement of therapeutic ranges.
Frame the case using meaningful metrics, including population at risk, actionable recommendations per drug-gene pair, absolute risk reduction, resource use avoided (e.g. bed days, ICU transfers, rescue therapies), and budget impact under local tariffs.
Delivering the PGx consultation service
A pharmacy-led PGx consultation should feel familiar, employing the same clinical reasoning pharmacists use daily. Triggers can be reactive, such as the ordering of a risky drug without a known phenotype, or testing can be pre-emptive, such as in oncology, transplant or percutaneous coronary intervention pathways. In either case, provide a concise one-page consult summarising genotype, phenotype, action and evidence, and implement order changes directly.
For algorithmic dosing, such as for thiopurines and tacrolimus, embed calculators or direct links within order sets. Above all, preserve the longevity of PGx data. Store phenotypes and actionable flags discretely so that future prescribers can see them instantly. This is where HL7 FHIR Genomics and EHR genomics modules deliver tangible value.
Conclusion
A turning point has arrived. The evidence is strong, the standards are clear, and the digital tools finally make PGx practical at the bedside. Pharmacy is uniquely placed to lead on these initiatives.
Start with a focused bundle of high-impact drug-gene pairs, integrate them cleanly into the EHR, embed equity into your allele tables and measure what matters. Make PGx reliable, routine and quietly transformative because that is how actual clinical change endures.
Take-home messages
Scale succeeds when event rates and actionability are both high. Launch with a focused bundle that delivers measurable safety gains and streamlined workflows.
- Fluoropyrimidines–DPYD: the EMA mandates pretreatment DPD assessment; CPIC and DPWG provide dosing algorithms. Include ancestry-relevant alleles to avoid blind spots2
- Clopidogrel–CYP2C19: the 2022 CPIC update emphasises the use of alternative P2Y12 inhibitors in patients with intermediate or poor metaboliser status, particularly after percutaneous coronary intervention or ischaemic stroke and transient ischaemic attack4
- Thiopurines–TPMT/NUDT15: genotype-guided dosing prevents myelosuppression in inflammatory bowel disease, oncology and transplant care8
- Tacrolimus–CYP3A5: expressers typically require 1.5–2-times higher starting doses; genotype-guided dosing shortens time to therapeutic troughs3
- Human leukocyte antigen (HLA) ‘hard stops’: avoid abacavir in HLA-B*57:01-positive patients and carbamazepine in HLA-B*15:02 or HLA-A*31:01 carriers to prevent severe cutaneous adverse reactions5
- Selective serotonin/noradrenaline reuptake inhibitors–CYP2D6/CYP2C19 (±CYP2B6): CPIC’s 2023 guidance clarifies dose adjustments or alternative agents10; DPWG updates offer practical nuances for polypharmacy.11
Select three to five drug-gene pairs, implement them flawlessly, demonstrate the benefit, and then expand.
Author
João Gonçalves PharmD PhD
Faculty of Pharmacy, University of Lisbon, Portugal
References
1 Swen J et al. A 12-gene pharmacogenetic panel to prevent adverse drug reactions: an open-label, multicentre, controlled, cluster-randomised crossover implementation study. Lancet 2023;401(10374):347–56.
2 European Medicines Agency. EMA recommendations on DPD testing before treatment with fluorouracil, capecitabine, tegafur and flucytosine. EMA/229267/2020.
3 Birdwell KA et al. Clinical Pharmacogenetics Implementation Consortium (CPIC) Guidelines for CYP3A5 Genotype and Tacrolimus Dosing. Clin Pharmacol Ther 2015;98(1):19–24.
4 Lee C et al. Clinical Pharmacogenetics Implementation Consortium Guideline for CYP2C19 Genotype and Clopidogrel Therapy: 2022 Update. Clin Pharmacol Ther 2022;112(5):959–67.
5 Phillips EJ et al. Clinical Pharmacogenetics Implementation Consortium Guideline for HLA Genotype and Use of Carbamazepine and Oxcarbazepine: 2017 Update. Clin Pharmacol Ther 2018;103(4):574–81.
6 Newsom KJ et al. Integration of pharmacogenetic data in epic genomic module drives clinical decision support alerts. Front Pharmacol 2024;15:1458095.
7 HL7 FHIR. Genomics reporting implementation guide. [Accessed October 2025].
8 Relling MV et al. Clinical Pharmacogenetics Implementation Consortium Guideline for Thiopurine Dosing Based on TPMT and NUDT15 Genotypes: 2018 Update. Clin Pharmacol Ther 2019;105(5):1095–105.
9 European Commission. European Health Data Space Regulation (EHDS)v. [Accessed October 2025].
10 Bousman CA et al. Clinical Pharmacogenetics Implementation Consortium (CPIC) Guideline for CYP2D6, CYP2C19, CYP2B6, SLC6A4, and HTR2A Genotypes and Serotonin Reuptake Inhibitor Antidepressants. Clin Pharmacol Ther 2023;114(1):51–68.
11 Brouwer JML et al. Dutch Pharmacogenetics Working Group (DPWG) guideline for the gene–drug interaction between CYP2C19 and CYP2D6 and SSRIs. Eur J Hum Genet 2022;30(10):1114–20.
