Implementation of Pharmacogenomics to Prevent Adverse Drug Reactions

What is pharmacogenomics?

Pharmacogenomics is defined by the European Medicines Agency (EMA) as ‘the study of variations of DNA and RNA characteristics as related to drug response’.1 Simply stated, it is the study of how genes affect a person’s response to medicines. It is estimated that more than 97% of people carry at least one variant in a gene that can influence drug response.2 Genomic variations may relate to the pharmacokinetics or pharmacodynamics of a drug, and therefore affect clinical outcome. These variations are also known as genetic polymorphisms.

Pharmacokinetic processes like absorption, distribution, metabolism and excretion are affected because of variations in genes that code for enzymes and transporters involved in these processes. An example of this is the enzyme CYP2D6 which is responsible for the metabolism of a variety of  antidepressants. Patients carrying multiple copies of the active gene for CYP2D6 have a higher rate of drug metabolism resulting in a reduction in the response and failure of the treatment.3

Pharmacodynamic properties are affected due to variations in genes that affect drug targets, for example the VKORC1 gene which codes for active sites of the enzyme, epoxide reductase. This enzyme is responsible for activating vitamin K and coagulation factors. Warfarin, an inhibitor of epoxide reductase, has a reduced efficacy in individuals with a genetic polymorphism of the VKORC1 gene. As a result of this polymorphism, more vitamin K will be activated that requires a higher dose of warfarin to be inhibited.3

One of the first links between genetics and adverse drug reactions (ADRs) was related to glucose-6-phosphate dehydrogenase (G6PD) deficiency. G6PD deficiency is an X-linked deficiency which results in an increased sensitivity to drug-induced haemolysis when exposed to certain medicines that increase oxidative stress. Some individuals with G6PD deficiency require strict avoidance of drug triggers, as haemolytic anaemia may be life-threatening.4

The application of genetic knowledge to personalised medicine could help to tailor treatment to each patient’s unique characteristics as opposed to a one-size-fits-all. Identifying these subpopulations who are likely to respond differently to medicines due to genomic factors, could provide important information to help mitigate the risk of ADRs and lack of efficacy.


Integrating pharmacogenomics into pharmacovigilance

Pharmacogenomic considerations should be applied as part of the benefit-risk evaluation throughout a product’s life cycle. The EMA’s guidance document (EMA/CHMP/718998/2016) provides a framework on how to evaluate pharmacovigilance related issues associated with pharmacogenomic biomarkers, with the aim of translating the results to appropriate treatment recommendations in the product labelling.5 Guidance is provided on considering pharmacogenomic data in the risk management plan, signal detection activities, periodic safety update reports and labelling updates. These should be considered together with the guidance provided by good pharmacovigilance practices (GVP Module V, VI, VII, IX and XVI). Below is a brief summary of the necessary recommendations.


Risk management plan (RMP)

Part II - Safety Specification

The safety specification summarises important identified risks, potential risks and missing information. Additionally, it addresses populations potentially at risk and any outstanding safety questions.

As part of the safety specification, any data regarding relevant genomic biomarkers, which may affect safety or efficacy available at the time of marketing authorisation, should be included. If a potentially clinically important genomic polymorphism has been identified but not fully studied, it should be reflected as missing information or a potential risk.

Polymorphisms which are clinically relevant and present an ethnic-dependent pattern of distribution may also be appropriate for discussion in this section for either preventing ADRs or improving benefits.

Module SIV “Populations not studied in clinical trials” allows for inclusion of data on the low exposure of populations with genetic polymorphisms, where available and appropriate, as well as the type of polymorphism.


Part III: Pharmacovigilance Plan

The pharmacovigilance plan describes pharmacovigilance activities (routine and additional) and action plans for each safety concern. Furthermore, any proposed actions to address identified safety concerns, together with the procedures in place to detect safety signals, are detailed. These may include additional post-authorisation safety or efficacy (PASS/PAES) studies.

Genomic aspects as part of pharmacovigilance planning should be considered early in the drug development program, and should continue post-marketing. PASS/PAES may be required to identify and/or characterise genomic biomarkers, the impact on patient selection, dose selection and choice of concomitant medications.


Part V: Risk minimisation measures

The types of risk minimisation measures relevant for genomic data are determined by the impact on the medicinal product’s effects, risks and clinical outcome. Where genomic biomarker information is included in the labelling or where genetic testing is indicated, these will be described as routine risk minimisation measures.

Additional risk minimisation measures may be necessary when routine measures are not sufficient. Such measures may include restricted access to medicinal products based on genotype or phenotype testing, patient registries or additional educational materials to the prescribers or patients.

Evaluation of the effectiveness of risk minimisation measures is mandatory. This may be done through specific studies to investigate whether genomic biomarker guided use of a medicinal product has been effective or not. It is also important to assess whether genetic testing may have had unintended consequences and the impact of including information in the labelling in terms of clinical actions. For example, if there are changes to how the medicine is used, if the recommendations are followed when not mandatory or what the influence is on clinical decision making.


Signal detection

In order to capture previously unidentified reactions relating to specific genetic polymorphisms, genomic data should be collected during signal detection. Signals can be identified from a number of important data sources including spontaneous ADRs, clinical trials and published studies. In some instances, genomic information can be generated using data from non-clinical in vitro and in vivo studies.

A pharmacogenomic surveillance system should be implemented for those medicinal products with genomic data included in the labelling. This system should include information tools, processes and studies to ensure that genomic testing is performed as per the labelling. When patients experience serious ADRs or lack of efficacy, genomic sampling is indicated to further investigate a genomic biomarker. The use of biobanks should be considered.

As per GVP, any identified signals should be evaluated according to the general process of signal management and reporting to regulatory agencies is expected if the findings fulfil the criteria for signals or emerging safety issues.


Periodic safety update reports (PSURs)

In section 16 of PSURs, the signal and risk evaluation should include any pertinent discussions on pharmacogenomic information. Any product usage data and characterisation of benefits and risks in subpopulations with genomic biomarkers should be presented, where relevant. Evaluation of data may relate to the strength of an association between a genomic biomarker and a safety concern, to severity/magnitude of the effect, and to patient ethnicity.


Labelling updates

Where there is an identified impact on the benefit-risk balance in a specific genomic subpopulation, pharmacogenomics related information should be considered in the product labelling. The information should be sufficiently detailed and clear with guidance for prescribers and patients on risks. If relevant, pharmacogenomic testing can be classified as mandatory, recommended or for information.


Overcoming the challenges

Given the complexity of pharmacogenomics, it is not surprising that many reasons are cited as barriers to the implementation in clinical practice. There is a plausible argument that the lack of its application among healthcare professionals is largely due to a lack of understanding of the topic and lack of evidence-based guidelines. Moreover, ethical issues surrounding genetic information may be of concern to some patients and healthcare professionals. Another significant obstacle is the cost of genetic testing.6

Despite the challenges, there are ongoing efforts globally to collate evidence and information in this area. The Pharmacogenomics Knowledgebase, funded by the National Institute of General Medical Sciences, is an interactive tool for researchers investigating how genetic variation affects drug response. It provides a repository of pharmacogenomic-based drug dosing clinical guidelines from multiple sources and currently includes 784 drug label annotations with pharmacogenomic information.7

The Ubiquitous Pharmacogenomics Consortium initiated the PREPARE (Preemptive Pharmacogenomic Testing for Preventing Adverse Drug Reactions) project with the aim to determine the impact of pre-emptive testing on ADR frequency, severity and associated costs. Major challenges and obstacles for implementation of pharmacogenomic testing in patient care will be addressed.8

In the UK, the Medicines and Healthcare products Regulatory Agency (MHRA) has launched a new initiative to establish a Yellow Card Biobank for researchers. The aim is to investigate genetic factors behind ADRs and optimise the safe use of medicines and vaccines. It is hoped that the operating model for the Yellow Card biobank will be established towards the end of this year.9

Over the past decade an increasing number of genomic markers of efficacy and adverse events have been discovered. Currently, pharmacogenetic testing is already available in various conditions such as haemolytic anaemia, cancer, malignant hyperthermia, porphyria, severe cutaneous disorders, Brugada and long QT syndromes,10 thereby enabling clinicians to select therapy more appropriately, predict therapeutic responses and identify those patients at high risk for side effects.

Pharmacovigilance plays a vital part in ensuring that doctors and patients are provided with adequate safety information to make educated therapeutic decisions. Pharmacogenomics is an emerging field and a powerful tool in understanding and preventing ADRs. It should be applied as part of the benefit-risk evaluation throughout a product’s life cycle. Integrating pharmacogenomics into pharmacovigilance can provide valuable information to improve safety and efficacy of therapeutic management.


Liezl Schonken is a GMC-registered medical doctor and is Panacea’s in-house Safety Physician.



  1. Guideline on good pharmacogenomic practice. Committee for Medicinal Products for Human Use (CHMP). EMA/CHMP/718998/2016. 22 February 2018.
  2. Dunnenberger HM, et al. Preemptive clinical pharmacogenetics implementation: current programs in five US medical centers. Annu Rev Pharmacol Toxicol 2015;55:89–106.
  3. Rojgar Hamed Ali. Impact of genetic variations on pharmacokinetic and pharmacodynamic properties of medicines and the future of drug therapy. Zanco J. Med. Sci. 2020;24(1).
  4. Bubp J, et al. Caring for Glucose-6-Phosphate Dehydrogenase (G6PD)-Deficient Patients: Implications for Pharmacy. P T. 2015;40(9):572-574.
  6. Virelli CR, et al. Barriers to clinical adoption of pharmacogenomic testing in psychiatry: a critical analysis. Translational Psychiatry (2021) 11:509.
  10. Micaglio E, et al. Role of Pharmacogenetics in Adverse Drug Reactions: An Update towards Personalized Medicine. Front Pharmacol. 2021; 12: 65172