EMA/CHMP 发布药物基因组学优良实践指南修订概念文件并启动征求意见

Official address Domenico Scarlattilaan 6 ● 1083 HS Amsterdam ● The Netherlands

An agency of the European Union

Address for visits and deliveries Refer to www.ema.europa.eu/how-to-find-us

Send us a question Go to www.ema.europa.eu/contact Telephone +31 (0)88 781 6000

1 December 2025

EMA/282050/2025

Committee for Medicinal Products for Human Use (CHMP)

Concept paper on the guideline revision on good pharmacogenomic practice

Agreed by Methodology Working Party (MWP) August 2025

Adopted by CHMP for release for consultation 01 December 2025

Start of public consultation 01 December 2025

End of consultation (deadline for comments) 01 March 2026

Agreed by MWP September 2026

Adopted by CHMP 05 October 2026

The proposed revised guideline will replace the 'Guideline on good pharmacogenomic practice' (EMA/CHMP/718998/2016).

Keywords Pharmacogenomics, good practices, pharmacogenomic analyses, biomarkers, study design, pharmacokinetics, DNA sequencing Concept paper on the guideline revision on good pharmacogenomic practice

EMA/282050/2025 Page 2/6

Introduction

Since its initial publication in 2018, the guideline on good pharmacogenomic practice has played a crucial role in shaping best practices, ensuring consistency, and providing a clear framework for regulators and stakeholders. As anticipated at the time of its implementation, the field of genomics has continued to evolve, driven by scientific advancements, technological innovations, and emerging regulatory considerations.

To reflect these developments, it is essential to revise the guideline on good pharmacogenomic practice, incorporating new evidence, methodologies, and best practices. By doing so, this guideline revision seeks to provide regulators and industry stakeholders with an improved framework that supports the highest standards of practice while fostering innovation and progress.

This guideline revision will take into account feedback from experts, recent scientific discoveries, and lessons learnt from the application of the initial guideline. The revised document will aim to maintain its core objectives while integrating improvements that align with the latest advancements in the field.

Additionally, the revised guideline will merge and incorporate updated elements from the

‘EMA/CHMP/281371/2013 Guideline on key aspects for the use of pharmacogenomics in the pharmacovigilance of medicinal products ’.

1. Problem statement

Advancements in the field of pharmacogenomics necessitate a revision of the existing guideline to ensure its continued relevance and applicability. In addition to incorporating new scientific knowledge, this revision aims to provide greater clarity on topics already covered, facilitating more precise and actionable guidance for regulators and stakeholders.

The issues requiring major revisions or inclusion of new guidance are:

Pharmacogenomic methodology

a. Sequencing technologies

b. Specific analytical issues

i. Polymorphic genes & substrate specificity

ii. Deoxyribonucleic acid (DNA) variants in different ancestries

iii. Proficiency testing

Interpretations and recommendations

a. Phenotype and genotype correlations

b. General versus specific medicinal product use recommendations

Reporting and nomenclature
Pharmacogenomic study design

a. Interventional studies

b. Non-interventional studies

Pharmacogenomics in the pharmacovigilance of medicinal products

2. Discussion (on the problem statement)

By integrating the following critical revisions, the revised guideline aims to provide a more comprehensive and scientifically rigorous framework for pharmacogenomic applications in medicines development, regulatory decision-making, and clinical practice:

Pharmacogenomic methodology

a. Sequencing technologies

Improvements in sequencing technologies, particularly the ability to generate longer reads, now allow for a more accurate distinction between complex gene structures, such as hybrid genes, genes with high sequence homology and genomic phasing. These advancements have significant implications for the identification and interpretation of pharmacogenetic variants. The revised guideline will therefore include aspects on the usage of third generation long read sequencing and will include aspects on quality control measures that were not elaborated on previously. Direct ribonucleic acid (RNA) sequencing and epigenetic sequencing are not within the scope of the guideline. The quality control section of the current guideline will also be revised to reflect these aspects. Concept paper on the guideline revision on good pharmacogenomic practice

EMA/282050/2025 Page 3/6

b. Specific analytical considerations

i. Polymorphic genes & substrate specificity

Cytochrome P450 (CYP) enzymes, such as the enzyme CYP2D6, are highly polymorphic and play a crucial role in metabolising a wide range of xenobiotics, including medicinal products, environmental chemicals, dietary compounds and endogenous compounds (endobiotics). With over 170 haplotypes

identified, CYP2D6 polymorphisms pose significant analytical challenges, necessitating specialised approaches which will be addressed in the revised guideline.

Metabolising enzymes of active substances can exhibit specificity, which describes an enzyme’s affinity

to bind to different substrates (i.e., medicines). Substrate specificity is crucial when a DNA variant leads to reduced enzymatic function. In these cases, the degree of functional impairment relative to the wild-type enzyme can vary depending on the specific medicinal product being metabolised. In this guideline revision, special attention will be paid to the CYP3A4/CYP3A5 interplay. Given that both enzymes share a substantial number of substrates and that a notable proportion of people of European ancestry carry a non-functional CYP3A5 allele, this should be carefully considered for relevant medicines during development and regulatory assessments (pre- and post-marketing).

Drug transporters (e.g. ABCB1 , SLCO1B1 ) will also be considered in the revised guideline.

ii. DNA variants in different ancestries

In recent years, increasing awareness of the differences in DNA variants across ancestries has necessitated a re-evaluation of how such data is interpreted and applied in pharmacogenomics. Genetic ancestry plays a crucial role in influencing interindividual variations in medicinal product exposure and response, as different genetic variants associated with ancestry can impact benefit-risk profiles. This variability challenges the conventional approach of testing for only the most common variants within broadly defined populations and calls for a more refined, population-specific strategy for variant selection. To ensure both safe medicines development and appropriate post-market pharmacovigilance strategies, it is essential to incorporate relevant pharmacogenomic data and genetic characteristics of different populations into the product information, thereby enabling optimal treatment regardless of genetic background. The revised guideline will address these complexities and provide recommendations on integrating pharmacogenomic differences across populations into medicines development practices. Relevant examples of known polymorphisms and the wording in the product information will also be discussed in the revised guideline.

iii. Proficiency testing

The guideline revision will include information on appropriate allelic testing, including quality requirements.

Interpretations and recommendations

a. Phenotype and genotype correlations

The revised guideline will expand on phenotype and genotype correlations and how they translate into clinical recommendations, such as clearly distinguishing between phenotypes with associated clinical recommendations and theoretical extrapolations based on enzymatic function, and discuss aspects related to phenoconversion as a result of the frequent concomitant use of inducers or inhibitors of specific pharmacogenes. This guideline will further inform the use of genetic information for medicinal products.

b. Medicinal product use recommendations

Greater clarity is needed regarding medicinal product use recommendations before and after market authorisation. When a medicinal product is predominantly metabolised by a single pharmacogene product, it is essential to ensure that non-functional variants of that gene are incorporated into safety protocols. If specific dosing recommendations are based on specific DNA variants or haplotypes, this must be explicitly stated. In such cases, a precise understanding of the *1 (wild type) designation in pharmacogenes is paramount to avoid misinterpretation and ensure accurate clinical guidance.

Reporting and nomenclature Concept paper on the guideline revision on good pharmacogenomic practice

EMA/282050/2025 Page 4/6

Unambiguous allele and genotype nomenclature is essential for regulatory assessments (pre- and post-marketing), as its absence significantly impairs the interpretation of pharmacogenomic results and hampers the determination of appropriate dose adjustments.

The revised guideline will:

Provide guidance on the proper use of internationally accepted allele and genotype nomenclature for pharmacogenes and their haplotypes.
Recommend internationally recognised sources for identifying and defining pharmacogenetically relevant genotypes and associated phenotypes.

The nomenclature used in the guideline must be corrected and refined, and above all used consistently. Overall, it should reflect an awareness that a large proportion of genetic variation remains unsampled and not all functionally relevant variants are known.

The revised guideline will further expand its definitions to include distinguishing between “allele ”,

“haplotype ” and “variant ”, and define functional categories (“loss -of-function”, “gain -of-function”) .Definitions will be provided for terms such as “variant,” “mutation,” “polymorphism,” “single nucleotide

polymorphism (SNP) ”, “structural variant” and “copy number variant (CNV)” . And while this is reflected in the current version of the guideline, the wording around “common ” and "rare” variants should be more carefully chosen to reflect our increasing understanding that variant frequencies differ across populations, and that the interpretation of frequency thresholds must therefore be applied cautiously and in relation to diverse populations.

Pharmacogenomic study design

a. Interventional studies

The revised guideline will provide recommendations on when and how to best utilise pharmacogenomics in clinical development from Phase 0 to Phase IV. It will be a comprehensive reference for designing interventional studies that takes the connections between genetic variability and medicines safety and efficacy into account. Special emphasis will be placed on referring to existing

guidelines, such as the ‘ICH M12 Guideline on the investigation of drug interactions’ , the

‘EMA/CHMP/37646/2009 Guideline on the use of pharmacogenetic methodologies in the pharmacokinetic evaluation of medicinal products’, and the ‘EMA/446337/2011 Reflection paper on

methodological issues associated with pharmacogenomic biomarkers in relation to clinical development

and patient selection’ in relevant sections and providing an overarching context for when they are applicable. Topics covered in detail will include:

Utilising pharmacokinetic/pharmacodynamic data to support genotype-guided dosing (including discussions on sparse sampling, modelling approaches, and selection of relevant clinical endpoints)
Specifics of randomised controlled trials (RCTs) in establishing genotype-medicinal product response relationships (e.g., sample size planning, power for rare variant analysis, multiple testing correction)
Complex trial designs such as enrichment designs, adaptive designs (ICH E20 adaptive designs for clinical trials), and master protocols
Genome-wide association studies in an RCT setting
Incorporating pharmacogenomic findings into the product information (aligned with A GUIDELINE ON SUMMARY OF PRODUCT CHARACTERISTICS (SmPC))

b. Non-interventional studies

The revised guideline will include a section on how real-world data (RWD) related to pharmacogenomics can be generated and utilised, an addition to the current version of the guideline.

Non-interventional studies using RWD can provide insights on populations for whom information from interventional studies is limited or missing, including but not limited to different ethnic groups, age groups, patients with comorbidities such as renal or hepatic impairment, patients taking co-medication (long-term), and patients at risk for rare adverse drug reactions (EMA/99865/2025 Reflection paper on use of real-world data in non-interventional studies to generate real-world evidence for regulatory Concept paper on the guideline revision on good pharmacogenomic practice

EMA/282050/2025 Page 5/6

purposes). RWD may also be used to expand knowledge on real-world clinical outcomes of trial populations, or support identification of suitable trial populations for pharmacogenomic studies.

Pharmacogenomics in the pharmacovigilance of medicinal products

The revised guideline will incorporate and update relevant elements from the ‘EMA/CHMP/281371/2013 Guideline on key aspects for the use of pharmacogenomics in the pharmacovigilance of medicinal products ’. The aim is to consolidate pharmacogenomic aspects relevant to the lifecycle of medicinal products into a single guidance document.

3. Recommendation

Considering the points identified above, the Methodology Working Party (MWP) recommends revising the guideline on good pharmacogenomic practice. MWP and the Pharmacovigilance Risk Assessment Committee (PRAC) also recommend the merging and revision of the guideline on key aspects for the use of pharmacogenomics in the pharmacovigilance of medicinal products.

4. Proposed timetable

Release of draft concept paper for public consultation in Q4 2025, with a deadline for comments in Q1 2026. Finalisation of concept paper in Q3 2026, and adoption of concept paper expected in Q4 2026 by the Committee for Medicinal Products for Human Use (CHMP).

Release of draft guideline for public consultation in 2027-2028, and a workshop with external participants. Finalisation of guideline in 2028.

5. Resource requirements for preparation

A temporary Drafting Group for the guideline revision will be established and will be composed of both MWP members, Methodology European Specialised Expert Community (ESEC) members, and Pharmacovigilance experts from the European Medicines Regulatory Network (EMRN). Regular monthly meetings will be scheduled to advance the drafting of the guideline and monitor progress. Input from MWP, scientific Committees (CHMP, PRAC, Committee for Advanced Therapies (CAT)) and Working Parties (Scientific Advice Working Party (SAWP) and Oncology Working Party (ONCWP)) will be requested throughout. The need for a workshop is anticipated for 2028.

6. Impact assessment (anticipated)

The guideline will enhance the understanding of methodological concepts and challenges in the field of pharmacogenomics. It will outline best practices for pharmacogenomics, ensuring consistency, and providing a clear framework for regulators and stakeholders, which is anticipated to benefit public health.

7. Interested parties

Guidance on this topic is expected to be important to the following Committees and Working Parties who will be consulted prior to the release of the draft guideline: CHMP, PRAC, CAT, SAWP and ONCWP. Input from industry, academia, and healthcare professionals will be gathered through public consultation and further facilitated via a dedicated workshop with external stakeholders in 2028. Concept paper on the guideline revision on good pharmacogenomic practice

EMA/282050/2025 Page 6/6

8. References to literature, guidelines, etc.

See Guideline on good pharmacogenomic practice for the current version of the guideline

See Guideline on key aspects for the use of pharmacogenomics in the pharmacovigilance of medicinal

products for the current version of the guideline

See Guideline on the use of pharmacogenetic methodologies in the pharmacokinetic evaluation of

medicinal products

See ICH M12 Guideline on drug interaction studies_Step 5

See Reflection paper on methodological issues associated with pharmacogenomic biomarkers in

relation to clinical development and patient selection

See A GUIDELINE ON SUMMARY OF PRODUCT CHARACTERISTICS (SmPC )

See Reflection paper on use of real-world data in non-interventional studies to generate real-world

evidence for regulatory purposes