
Mariia Vd / Shutterstock.com
After reading this article, you should be able to:
- Distinguish a genotype-predicted CYP2D6 phenotype from current functional enzyme activity and define phenoconversion;
- Identify medicine changes and other factors that can cause CYP2D6 activity to differ from the genotype-predicted phenotype;
- Assess a patient safely before attributing symptoms or treatment failure to a pharmacogenomic mechanism;
- Use product information, national guidance and pharmacogenomic resources to support action, communication and follow up.
Introduction
Pharmacogenomics uses inherited genetic variation to explain differences in medicine response in individuals and can be used to inform medicine or dose selection. For pharmacists, a pharmacogenomic result is only one part of a medicines assessment: indication, dose, adherence, organ function, interacting medicines and current response all require consideration1–3.
Commissioning arrangements for pharmacogenomic testing differ across the UK. In England, the National Genomic Test Directory defines the genomic tests commissioned by NHS England and their eligibility criteria. Pharmacogenomic testing is already used for selected gene–medicine pairs, including DPYD testing before fluoropyrimidine treatment and MT-RNR1 testing in defined aminoglycoside-related indications, while other applications remain limited, locally available or under evaluation4,5.
NHS England’s pharmacy genomics workforce framework positions pharmacists and pharmacy technicians within genomics-informed medicines optimisation, including interpretation of genomic information and contribution to medicine and dose selection6.
CYP2D6 is a cytochrome P450 drug-metabolising enzyme, which is involved in the metabolism of many commonly used medicines (e.g. codeine and several selective serotonin reuptake inhibitors)3,7. Genetic variation in CYP2D6 can alter enzyme function, so genotyping can be used to predict how an individual is likely to metabolise medicines for which CYP2D6 activity is clinically relevant.
The genotype is used to assign a genotype-predicted metaboliser phenotype — poor, intermediate, normal or ultrarapid — which reflects increasing expected CYP2D6 activity, from little or no activity in poor metabolisers to increased activity in ultrarapid metabolisers3,7. These categories are predictions based on inherited genetic variation, not direct measurements of enzyme activity at the time a patient is reviewed.
Phenoconversion occurs when non-genetic factors cause current drug-metabolising activity to differ from that predicted by genotype8,9. In clinical practice, CYP2D6 inhibition phenoconversion can be caused by concomitant medicines, including paroxetine, fluoxetine, quinidine and terbinafine. By reducing CYP2D6 activity, these medicines can cause current functional activity to differ from the genotype-predicted phenotype8–10.
Phenoconversion is not limited to CYP2D6. Genotype–phenotype discordance has also been investigated for other CYP enzymes, including CYP2C19, while non-genetic influences on CYP3A-mediated metabolism illustrate further complexity8,11.
The direction of change is not always towards lower activity. CYP inhibitors, inflammation and some disease-related factors can reduce metabolic activity, whereas other influences can increase it. For example, CYP2D6 activity can increase during pregnancy in individuals with functional CYP2D6 alleles8,11.
The clinical consequences of CYP2D6 inhibition depend on how the enzyme contributes to a medicine’s metabolism. For an active parent drug, such as atomoxetine, inhibition can reduce metabolism and increase drug exposure. For a prodrug, such as codeine, the same inhibition can reduce conversion to its active metabolite, morphine, potentially increasing toxicity risk10,12. CYP2D6 inhibition therefore does not change therapeutic effects in the same direction for every medicine. Conversely, increased CYP2D6 activity may have the opposite metabolic effects, depending on the medicine8,9.
CYP2D6 genotyping is technically complex because of allelic variation, deletions, duplications or multiplications and pseudogenes3,7,13. It does not measure current CYP2D6 activity or account for medicines introduced later. In 2025, NHS Genomics Education noted that a robust assay and appropriate interpretation is important when CYP2D6 results are used clinically13.
The NHS England-supported PROGRESS programme is piloting panel-based pharmacogenetic-guided prescribing in general practice and included common pharmacogenes such as CYP2D6. The programme has tested end-to-end pathways for test ordering, result return and electronic clinical decision support across sites in England to assess the feasibility and utility of pharmacogenetic-guided prescribing; however, this is an implementation study rather than routine CYP2D6 testing14,15.
This article will use a fictional case — inspired by practice — to demonstrate how pharmacists should respond to suspected phenoconversion, in the context of a previous pharmacogenomic test result. Numerical observations and chronology are constructed for educational purposes, while evidence-based clinical actions are referenced.
Case presentation
A 36-year-old man with attention deficit hyperactivity disorder (ADHD) is reviewed in general practice. He has no known cardiovascular disease or arrhythmia. Atomoxetine was initiated by a specialist ADHD service after appropriate stimulant treatment options were not tolerated and has been stable at 80mg once daily for eight months. NICE recommends shared care with primary care after titration and dose stabilisation, with ongoing monitoring of effectiveness, adverse effects, heart rate and blood pressure16.
When atomoxetine was initially considered, the patient used a direct-to-consumer pharmacogenetic testing service to obtain CYP2D6 genotyping. The patient made this decision after doing their own research and the experience of trialling other ADHD medicines unsuccessfully17. The patient was advised that CYP2D6 testing is not part of routine care and is not a required prerequisite for atomoxetine prescribing. The report recorded CYP2D6 *1/*1, two gene copies and a genotype-predicted normal-metaboliser phenotype. It also identified the testing laboratory and assay used, which helps establish the provenance and analytical scope of the result when it is reviewed later3,7,13. It was used to inform treatment individualisation, not as a stand-alone prescribing instruction13,18.
A total of 18 days before review, the patient’s GP started terbinafine 250mg once daily for laboratory confirmed toenail onychomycosis, with a planned 12-week course. Before prescribing terbinafine, the GP assessed for pre-existing liver disease and reviewed baseline liver-function test results obtained before treatment, in accordance with product information19. Terbinafine inhibits CYP2D6 activity19. Around day 10 of terbinafine treatment, the patient started to experience new insomnia and jitteriness, while by day 16, intermittent palpitations had developed. On day 18, the patient contacted his GP practice because the symptoms followed a recent medicine change, and an in-person medicines review was arranged with a clinical pharmacist in the GP practice.
Immediate assessment
The patient reported intermittent rapid palpitations but no chest pain, unexplained syncope or breathlessness. His pulse was 106 beats/min and regular, while his blood pressure was 146/88mmHg. At the most recent stable review, the patient’s pulse was 72 beats/min and blood pressure was 124/78mmHg. Medicines reconciliation confirms the list shown in Table 110,19. The patient reported no increase in caffeine intake, no use of stimulant-containing non-prescription products and no change in atomoxetine adherence. Unchanged intake does not entirely exclude caffeine as a contributor: the terbinafine summary of product characteristics (SmPC) reports reduced clearance of intravenously administered caffeine; although, its clinical significance for this patient is uncertain19.
Table 1: Medicines and products
The SmPC for atomoxetine states that patients who develop palpitations, exertional chest pain, unexplained syncope, dyspnoea or other symptoms suggestive of cardiac disease should undergo prompt specialist cardiac evaluation10. The pharmacist contacted the responsible GP during the review because the patient is symptomatic and cardiovascular observations have changed. The GP then arranged prompt specialist cardiac evaluation in line with the SmPC. The National Institute for Health and Care Excellence (NICE) separately advises dose reduction and referral for sustained resting tachycardia above 120 beats/min, arrhythmia or a clinically significant systolic blood-pressure increase measured on two occasions for individuals taking ADHD medication16.
What has changed?
Once immediate safety is addressed, the pharmacist reconstructed the timeline. Atomoxetine had been stable for months; terbinafine was the major recent medicine change; and symptoms appeared after its introduction. The historical CYP2D6 result required consideration: does “normal metaboliser” status still describe current functional activity?
The results of a 1999 prospective study of nine healthy volunteers — 6 genotypically extensive metabolisers and 3 poor metabolizers for CYP2D6 — who were given terbinafine 250 mg daily for 14 days revealed that the drug markedly altered CYP2D6 activity measured with dextromethorphan20. A total of 4 out of 6 genotypic extensive metabolisers were phenotypically classified as poor metabolisers after treatment, the study results showed20. This demonstrates CYP2D6 inhibition and genotype–phenotype discordance. The study did not measure atomoxetine exposure and does not inform the cause of this patient’s symptoms.
While the patient’s current CYP2D6 activity has not been measured, given the patient’s medicine history and recent changes, the appropriate interpretation is possible phenoconversion8,9,20.
Repeat CYP2D6 genotyping would not confirm phenoconversion because the inherited genotype is unchanged. Where in vivo CYP2D6 phenotyping is available, it could demonstrate whether current enzyme activity has diverged from the genotype-predicted phenotype — it is not required for routine management of this interaction9,20.
Figure: Overview of patients CYP2D6 genotype, medicines change and symptoms leading to suspected phenoconversion
What does the evidence say about atomoxetine?
Atomoxetine exposure is strongly influenced by CYP2D618,21,22. Paroxetine, which is a CYP2D6 inhibitor, increased atomoxetine maximum concentration around three-and-a-half-fold and area under the concentration–time curve around six-and-a-half-fold in CYP2D6 extensive metabolisers. The results of a 2020 study revealed that genotype-related pharmacokinetic differences were also attenuated during CYP2D6 inhibition23,24.
The UK atomoxetine SmPC does not provide a terbinafine-specific effect size. It states more generally that, with CYP2D6 inhibitors — listing fluoxetine, paroxetine, quinidine and terbinafine — atomoxetine exposure may increase six- to eight-fold and steady state maximum concentration three- to four-fold. Slower titration or a lower final dose may be necessary, while response and tolerability should be re-evaluated when a CYP2D6 inhibitor is started or stopped after titration10.
From evidence to a medicines management decision
The pharmacist and GP considered both medicines rather than assuming one must automatically change. They reviewed the antifungal indication, the treatment completed and alternatives. In addition, the ADHD service was contacted to reassess atomoxetine dose and tolerability. The interruption of terbinafine was judged acceptable. After discussion with the patient, the GP stopped terbinafine and, in consultation with the ADHD clinician, agreed to a temporary atomoxetine dose reduction under the local shared-care arrangement, during cardiac assessment and closer monitoring. The decision is based on symptoms, chronology, the recognised interaction and product information. PGx adds mechanistic context but does not dictate the intervention10,16,18,19.
If terbinafine could not reasonably be stopped, atomoxetine response, tolerability and dose would require reassessment with the ADHD prescriber, with closer cardiovascular and symptom monitoring10.
In the follow-up plan, the GP was responsible for the initial cardiovascular assessment and arranged prompt specialist cardiac evaluation, as well as the fungal treatment plan. Atomoxetine dose changes were agreed with the ADHD service. The GP initiated the specialist referral on the day of presentation, while cardiac assessment was completed before the seven-day follow-up. The follow up covered pulse, blood pressure, palpitations, sleep and ADHD control. Response and tolerability were then reassessed after either medicine changes, as advised by the atomoxetine SmPC10.
Seven days later, the patient reported no further palpitations. His pulse and blood pressure were moving towards baseline, sleep was improving, and the specialist cardiac assessment did not identify any arrhythmia or other findings requiring further intervention. Because both terbinafine exposure and atomoxetine management changed, improvement was not proof that phenoconversion caused the presentation.
Evidence from a 2005 case report involving amitriptyline indicated that terbinafine-related CYP2D6 interactions may persist after treatment discontinuation; although, the duration of any residual effect on atomoxetine was unknown25. Continued reassessment of atomoxetine tolerability and cardiovascular observations was therefore appropriate rather than assuming that the interaction has fully resolved.
Onychomycosis remained under active review. Systemic antifungal treatment was not immediately replaced, while the GP reassessed the confirmed diagnosis, extent of nail involvement and need for systemic treatment. If systemic therapy remains indicated, the next antifungal is chosen using relevant antifungal guidance and product information, taking account of the causative organism, interaction profile and patient-specific contraindications26.
Additional examples in pharmacy practice
UK studies have explored how pharmacogenomic services could be incorporated into community pharmacy practice27,28. CYP2D6-related phenoconversion can have different clinical consequences, depending on the medicine and care setting. Table 212,29–37 provides illustrative examples rather than an exhaustive list.
Table 2: Four CYP2D6 situations pharmacists may recognise
A practical sequence before acting on a PGx result
- Assess the patient first — are there new symptoms, abnormal observations or warning signs requiring escalation? Examples include new palpitations, exertional chest pain, unexplained syncope or dyspnoea in patients receiving atomoxetine. Assess urgency using the relevant medicine-specific guidance;
- Establish any recent changes — new, stopped or dose-changed medicines, adherence, prescribed and non-prescription products, and relevant clinical factors (e.g. intercurrent inflammation, changes in liver function or pregnancy, where applicable);
- Identify the interaction independently of PGx — what effect does the new medicine have on CYP2D6, and what does reduced enzyme activity mean for a medicine metabolised by CYP2D6 such as atomoxetine? A first terbinafine prescription in a patient on atomoxetine should prompt an interaction review at the point of supply;
- Verify the PGx result — original report, genotype/diplotype, predicted phenotype, source, assay limitations and copy-number information. If provenance or analytical validity is uncertain, clarify this before using the result in clinical decision-making. Even a verified PGx result must be interpreted alongside the patient’s clinical status, medicines and relevant guidance, rather than used in isolation to direct prescribing;
- Separate evidence from inference — an inhibitor plus a plausible mechanism can support possible phenoconversion. They do not prove causality in an individual patient;
- Use the right resources — check relevant UK product information and NICE guidance. The Clinical Pharmacogenetics Implementation Consortium can support interpretation of a verified genotype but does not turn a historical result into an automatic NHS prescribing instruction;
- Agree ownership and follow-up — document the PGx result, interaction, reasoning, action and monitoring. At discharge, communicate medicine changes, rationale, relevant PGx information/provenance and follow-up responsibility7–10,13,16,18–20,38,39.
Best practice
- Treat genotype-predicted phenotype as durable genetic information, not a permanent measurement of current enzyme activity;
- Assess symptoms or loss of efficacy and the medicine interaction before attributing either to a pharmacogenomic mechanism;
- Verify the PGx report and use medicine-specific UK product information, national guidance and appropriate PGx resources before genotype-based recommendations;
- Document the decision, monitoring and follow-up responsibility, as well as communicate them across care settings7–10,13,16,18–20,38,39.
Conclusion
A historical CYP2D6 result can remain genetically correct, while current medicines may alter functional CYP2D6 activity, so the genotype-predicted phenotype may no longer reflect current metabolic activity. The response is not to ‘correct’ the genotype or automatically change a dose, but to identify what changed, assess the patient, verify the result, understand the medicine-specific consequence and coordinate a documented plan8,9.
Artificial intelligence disclosure
ChatGPT (OpenAI) was used as an assistive tool for literature scoping, source comparison and manuscript development. The author independently reviewed the cited sources, made all clinical and editorial decisions, and takes full responsibility for the submitted content.
Affiliations and financial disclosure
Dario Mlinarić is founder and scientific lead of MEDORA Intelligence and undertakes freelance medical-writing work. He has submitted or contributed work to other healthcare publications. No manufacturer, pharmacogenomic testing provider or other commercial organisation funded or commissioned this article, and no relevant financial conflict is declared.
- 1.Personalised prescribing: using pharmacogenomics to improve patient outcomes. Royal College of Physicians and British Pharmacological Society. 2022. https://www.rcplondon.ac.uk/resources/personalised-prescribing-using-pharmacogenomics-to-improve-patient-outcomes/
- 2.Introduction to pharmacogenomics. NHS England Genomics Education Programme. https://www.genomicseducation.hee.nhs.uk/genotes/knowledge-hub/introduction-to-pharmacogenomics/
- 3.Pharmacogenomics of CYP2D6. NHS England Genomics Education Programme. https://www.genomicseducation.hee.nhs.uk/genotes/knowledge-hub/cyp2d6/
- 4.National genomic test directory. NHS England. 2026. https://www.england.nhs.uk/publication/national-genomic-test-directories/
- 5.Genomics in Pharmacy. NHS England Genomics Education Programme. https://www.genomicseducation.hee.nhs.uk/genomics-in-healthcare/genomics-in-pharmacy/
- 6.Pharmacy genomics workforce, education and training strategic framework. NHS England. 2024. https://www.england.nhs.uk/long-read/pharmacy-genomics-workforce-education-and-training-strategic-framework/
- 7.Caudle KE, Sangkuhl K, Whirl‐Carrillo M, et al. Standardizing <scp>CYP</scp> 2D6 Genotype to Phenotype Translation: Consensus Recommendations from the Clinical Pharmacogenetics Implementation Consortium and Dutch Pharmacogenetics Working Group. Clinical Translational Sci. 2019;13(1):116-124. doi:10.1111/cts.12692
- 8.Klomp SD, Manson ML, Guchelaar HJ, Swen JJ. Phenoconversion of Cytochrome P450 Metabolism: A Systematic Review. JCM. 2020;9(9):2890. doi:10.3390/jcm9092890
- 9.Cicali EJ, Elchynski AL, Cook KJ, et al. How to Integrate CYP2D6 Phenoconversion Into Clinical Pharmacogenetics: A Tutorial. Clin Pharma and Therapeutics. 2021;110(3):677-687. doi:10.1002/cpt.2354
- 10.Atomoxetine 80 mg hard capsules: Summary of Product Characteristics. Electronic Medicines Compendium. https://www.medicines.org.uk/emc/product/10336/smpc
- 11.Scodellaro S, Triantafilou S, Cohn I. Factors Influencing Phenoconversion in CYP-Mediated Drug Metabolism: A Scoping Review. Pharmacy. 2026;14(4):108. doi:10.3390/pharmacy14040108
- 12.Crews KR, Monte AA, Huddart R, et al. Clinical Pharmacogenetics Implementation Consortium Guideline for CYP2D6 , OPRM1 , and COMT Genotypes and Select Opioid Therapy. Clin Pharma and Therapeutics. 2021;110(4):888-896. doi:10.1002/cpt.2149
- 13.Results: patient with a known CYP2D6 genotype requiring ondansetron. NHS England Genomics Education Programme. 2025. https://www.genomicseducation.hee.nhs.uk/genotes/in-the-clinic/results-patient-with-a-known-cyp2d6-genotype-requiring-ondansetron/
- 14.McDermott JH, Tsakiroglou M, Newman WG, Pirmohamed M. Pharmacogenomics in the UK National Health Service: Progress towards implementation. Br J Clin Pharmacol. 2025;91(8):2241-2250. doi:10.1002/bcp.70109
- 15.Brunton L, Newman W, McDermott J, Wilson P. Implementing pharmacogenetic-guided prescribing in general practice: a qualitative process evaluation. Implement Sci Commun. 2026;7(1). doi:10.1186/s43058-026-01005-x
- 16.Attention deficit hyperactivity disorder: diagnosis and management. National Institute for Health and Care Excellence. 2019. https://www.nice.org.uk/guidance/ng87
- 17.Direct-to-consumer constitutional (germline) genomic testing. NHS England Genomics Education Programme. 2023. https://www.genomicseducation.hee.nhs.uk/genotes/knowledge-hub/direct-to-consumer-constitutional-germline-genomic-testing/
- 18.Brown JT, Bishop JR, Sangkuhl K, et al. Clinical Pharmacogenetics Implementation Consortium Guideline for <scp>Cytochrome P450 (CYP)</scp>2D6 Genotype and Atomoxetine Therapy. Clin Pharma and Therapeutics. 2019;106(1):94-102. doi:10.1002/cpt.1409
- 19.Terbinafine 250 mg tablets: Summary of Product Characteristics. Electronic Medicines Compendium. https://www.medicines.org.uk/emc/product/102562/smpc
- 20.ABDELRAHMAN S, GOTSCHALL R, KAUFFMAN R, LEEDER J, KEARNS G. Investigation of terbinafine as a CYP2D6 inhibitor in vivo. Clinical Pharmacology & Therapeutics. 1999;65(5):465-472. doi:10.1016/s0009-9236(99)70065-2
- 21.Sauer JM, Ponsler GD, Mattiuz EL, et al. Disposition and Metabolic Fate of Atomoxetine Hydrochloride: The Role of CYP2D6 in Human Disposition and Metabolism. Drug Metabolism and Disposition. 2003;31(1):98-107. doi:10.1124/dmd.31.1.98
- 22.Sauer JM, Ring BJ, Witcher JW. Clinical Pharmacokinetics of Atomoxetine. Clinical Pharmacokinetics. 2005;44(6):571-590. doi:10.2165/00003088-200544060-00002
- 23.Belle DJ, Ernest CS, Sauer J, Smith BP, Thomasson HR, Witcher JW. Effect of Potent CYP2D6 Inhibition by Paroxetine on Atomoxetine Pharmacokinetics. The Journal of Clinical Pharma. 2002;42(11):1219-1227. doi:10.1177/009127002762491307
- 24.Jung EH, Lee YJ, Kim DH, et al. Effects of paroxetine on the pharmacokinetics of atomoxetine and its metabolites in different CYP2D6 genotypes. Arch Pharm Res. 2020;43(12):1356-1363. doi:10.1007/s12272-020-01300-8
- 25.Castberg I, Helle J, Aamo TO. Prolonged Pharmacokinetic Drug Interaction Between Terbinafine and Amitriptyline. Therapeutic Drug Monitoring. 2005;27(5):680-682. doi:10.1097/01.ftd.0000175910.68539.33
- 26.Antimicrobial guidelines for primary care. Version 16. Greater Manchester Medicines Management Group. 2024. https://gmmmg.nhs.uk/wp-content/uploads/2025/01/GM-Antimicrobial-guidelines-October-2024-v16.pdf
- 27.Rendell T, Barnett J, Wright D. Co-designing a community pharmacy pharmacogenomics testing service in the UK. BMC Health Serv Res. 2022;22(1). doi:10.1186/s12913-022-07730-y
- 28.Rendell T, Barnett J, Scott S, Wright D. Designing an evidence based community pharmacy service specification for a pharmacogenomic testing service. Int J Clin Pharm. 2022;44(6):1406-1416. doi:10.1007/s11096-022-01483-8
- 29.Metoprolol tartrate 50 mg film-coated tablets: Summary of Product Characteristics. Electronic Medicines Compendium. https://www.medicines.org.uk/emc/product/5200/smpc
- 30.Bahar MuhA, Kamp J, Borgsteede SD, Hak E, Wilffert B. The impact of CYP2D6 mediated drug–drug interaction: a systematic review on a combination of metoprolol and paroxetine/fluoxetine. Brit J Clinical Pharma. 2018;84(12):2704-2715. doi:10.1111/bcp.13741
- 31.Duarte JD, Thomas CD, Lee CR, et al. Clinical Pharmacogenetics Implementation Consortium Guideline (CPIC) for CYP2D6, ADRB1, ADRB2, ADRA2C, GRK4, and GRK5 Genotypes and Beta‐Blocker Therapy. Clin Pharma and Therapeutics. 2024;116(4):939-947. doi:10.1002/cpt.3351
- 32.Nahid NA, McDonough CW, Wei YJJ, et al. CYP2D6 Phenotypes and Emergency Department Visits Among Patients Receiving Opioid Treatment. JAMA Netw Open. 2025;8(7):e2523543. doi:10.1001/jamanetworkopen.2025.23543
- 33.Aripiprazole Aspire 10 mg tablets: Summary of Product Characteristics. Electronic Medicines Compendium. https://www.medicines.org.uk/emc/product/6954/smpc
- 34.Kiss Á, Menus Á, Tóth K, et al. Phenoconversion of CYP2D6 by inhibitors modifies aripiprazole exposure. Eur Arch Psychiatry Clin Neurosci. 2019;270(1):71-82. doi:10.1007/s00406-018-0975-2
- 35.Tamoxifen 10 mg tablets: Summary of Product Characteristics. Electronic Medicines Compendium. https://www.medicines.org.uk/emc/product/11607/smpc
- 36.Paroxetine 30 mg tablets: Summary of Product Characteristics. Electronic Medicines Compendium. https://www.medicines.org.uk/emc/product/539/smpc
- 37.Goetz MP, Sangkuhl K, Guchelaar H, et al. Clinical Pharmacogenetics Implementation Consortium (CPIC) Guideline for CYP2D6 and Tamoxifen Therapy. Clin Pharma and Therapeutics. 2018;103(5):770-777. doi:10.1002/cpt.1007
- 38.Medicines optimisation: the safe and effective use of medicines to enable the best possible outcomes. National Institute for Health and Care Excellence. 2015. https://www.nice.org.uk/guidance/ng5
- 39.Pharmacogenomics: resources to support answering questions. Specialist Pharmacy Service. 2026. https://sps.nhs.uk/articles/pharmacogenomics-resources-to-support-answering-questions/


