Pharmacogenomics (PGx) is an evolving field that studies how genetic variations affect individual drug responses and represents a crucial intersection of genomics and pharmacology. PGx can be used to personalize medicine by tailoring drug treatments based on individual genetic profiles. This approach promises to enhance drug efficacy, reduce adverse drug reactions, and optimize therapeutic regimens. With significant advancements in genomic technologies and data analytics, the field of PGx is poised for transformative impact in healthcare.

This white paper focuses on advances in pharmacogenomic research, from technological advances to implementation science and the broader role of the inclusion of genetic data in clinical trials. Underpinning this is the importance of collaborative efforts among researchers, clinicians, and industry partners to drive innovation and improve patient outcomes.

Introduction 

PGx involves the investigation of genes that affect the absorption, distribution, metabolism, or excretion (ADME) of a drug, with the majority involved in drug metabolism. While pharmacogenetics was a term first coined in 1959, there has been a relatively slow adoption into mainstream healthcare. 

PGx was originally implemented in secondary care for a small group of high-risk medications (eg. Abacavir), today, PGx has applications in multiple clinical disciplines; cardiology, gastroenterology, psychiatry, transplant, neurology, and oncology. In the UK in primary care, over 80% of patients were found to be exposed to at least one PGx drug and 58% to two or more PGx drugs over a 20 year period (1). 

With advances in genomic technology now translated into the clinical pathway for rare diseases and oncology, the opportunity for PGx to be implemented into everyday use is starting to become a reality. Resources such as PharmGKB, (Pharmacogenomics Knowledgebase), in parallel with consensus guidelines from organizations such as Clinical Pharmacogenetics Implementation Consortium (2) (CPIC), variant level data standardization from PharmVar (3), recommended minimum requirements for testing from American College of Pathology (4) (AMP) and development of clinical decision support and electronic health records are enabling the scaling of PGx testing.

Pharmacogenomic analysis relies on the detection of a specific combination of genetic variants,(the combination of which is known as a haplotype. Haplotypes are labeled as star alleles (or *allele) and these are in turn translated to a metabolizer status.  Genetic testing methods to date have relied on only testing these specific variants and assuming the phase (whether the variants are inherited from one parent or another) based on prior knowledge. However, these data sets were derived from largely European populations and thus there is an element of bias (5)^. Advanced laboratory testing now means that there is an opportunity to investigate the genome in full, resolve typically hard-to-sequence regions, and provide a more complete understanding of the impact of genetic variation on the metabolism of medications.

Innovation in Pharmacogenomics (PGx) Research

Whilst reactive PGx testing has been introduced to healthcare systems for specific gene-drug pairs such as HLA-B*5701 for abacavir, preemptive testing has been slow in widespread adoption. Barriers to implementation include concerns over information technology infrastructure, workforce understanding and education, perceived complexity of results, timeliness of result availability, and health economics, in addition to lack of standardization and regional guidance (6). Recent advances in the application of PGx have been driven by technological advancements, comprehensive genomic and pharmacogenomic databases, and implementation research demonstrating real-world benefits and solutions of integrating PGx in the clinical pathway.

Technological Advancements – High Throughput Sequencing

Until recently, PGx was limited to testing a relatively small number of common genetic variants. Advances in microarray and subsequently next-generation sequencing (NGS) technologies have significantly impacted pharmacogenomic research and opportunity for delivery in the clinic. 

The majority of PGx markers currently tested clinically are biased to those that are most prevalent in the populations on which research has been historically conducted, primarily European ancestry. These therefore do not capture all the genetic variation that may contribute to drug metabolism, both within genes and across populations. This ‘missing information’ partially accounts for the varied response to medication within a particular metabolizer status. For example, Kennedy et al (7)compared whole genome sequencing to commercially available targeted genotyping and found that in 1% of cases, additional rare variants were identified which altered the interpretation of metabolizer status. 

From a research perspective, the ability to rapidly and affordably sequence entire genomes is empowering researchers to identify and analyze genetic variants associated with drug response beyond the scope of traditional testing. In the clinic, as more patients have their genome sequenced as part of their diagnostic pathway, there is clear utility in being able to ‘reuse’ this data for PGx reporting. For example, van der Lee et al (8) showed that 86% of individuals with exome data already available, had at least one actionable PGx phenotype. 

In addition to being able to clinically sequence the whole human genome, long-range sequencing (LRS) via platforms such as Oxford Nanopore Technologies (ONT) and Pacific Bioscience (PacBio), is revealing even deeper insights into pharmacologically relevant variation, enabling accurate phasing of variants and resolution of difficult to sequence genes such as CYP2D6. On a practical level, this can obviate the need to perform multiple laboratory tests on a single sample, in addition to yielding information on variant phasing which can more accurately determine metabolizer status.

With almost 100% of the population expected to have at least one genetic variation which would influence the metabolism of at least one drug, there will be a need for PGx testing to be both scalable and economic. In addition to the expanded scope of laboratory testing, an area which has seen progress in this regard is in bioinformatics and data analysis. 

Technological Advancements – Bioinformatics, Data Analytics and Information Technology

Advanced bioinformatics tools and machine learning algorithms are critical for analyzing large-scale genomic data. Not only do these tools help identify novel genetic markers, predict drug responses, and uncover complex gene-drug interactions, but they are also required for scalable, reproducible, and auditable delivery of PGx in the clinic. Integrating bioinformatics with clinical data enhances the predictive power of pharmacogenomic models, leading to more precise and personalized treatment strategies and in the long term can contribute to a ‘learning health system’.

Bioinformatics tools have had to evolve in parallel with advances in sequencing technology; for example, Illumina DRAGEN secondary analysis includes special callers for resolving variation in CYP2D6 and HLA, in addition to a number of other genes, whilst Oxford Nanopore Technologies and PacBio secondary analysis enables calling of phased variants thanks to the long reads. 

The benefit of directly phasing variants in the context of PGx is that to date PGx diplotypes relied on ‘assumed phase’, based on linkage disequilibrium. Whilst accurate most of the time, on an individual level it could lead to the inaccurate assignment of diplotype and therefore metaboliser status (9). 

A further consideration of data is the capacity for data reanalysis – star alleles status are updated with additional Single Nucleotide Polymorphism (SNP) contributions, medication recommendations can change in line with best practice guidelines and testing for additional gene-drug pairs may be recommended. All this requires capacity for data storage (in line with information governance) and reanalysis/reannotation. A step forward from this is integration and ability to query results alongside electronic health records (EHRs); healthcare providers frequently state that real-time ability to query PGx results in the context of a particular medication (eg. at time of prescribing) would be the desired method of integrating PGx into standard of care (10)

The above components rely on data standardization; PhamVar (Pharmacogene Variation Consortium) is a central repository for PGx variation haplotype structure and allelic variation whilst PharmGKB is a key resource for converting diplotype to metabolizer status and medication recommendations. When including this information in electronic health records, SNOMED CT (Systematized Nomenclature of Medicine Clinical Terms )(11)  can be used – this is an international dictionary of coded clinical terminology, and this includes metabolizer status of the more common PGx genes (for example CYP2D6 poor metabolizer is SCTID:738532000). In addition, HL7 FIHR have defined specifications for PGx reporting interoperability (12).

Implementation Research

An area which has seen considerable progress of late is the plethora of studies investigating the real world implementation of PGx in a range of healthcare systems and clinical settings. One of the seminal publications is that of the U-PGx Consoritum and the PREPARE study (13). This multicenter international study explored the implementation of a 12-genes, preemptive PGx testing on the incidence of clinically relevant adverse drug reactions and found a significant reduction (OR 0.70) in the PGx guided therapy arm. Furthermore, economic evaluations emerging from the study demonstrate cost-effectiveness, such as in the Spanish antiplatelet treatment group (14). 

Hayward et al. (15) reviewed 13 studies describing 8 implementation models in primary or community care settings and found that key themes to emerge were point-of-care clinical decision support, education and commissioning. It is clear that there is a global acknowledgement of the potential benefit of PGx, however implementation and obstacles are likely to have regional nuance, for example, the sceptism of healthcare providers to adopt may vary (16) .

Adverse Drug Reaction Monitoring for Pharmacogenomics

Pharmacogenomic testing helps identify patients at risk for adverse drug reactions, allowing for proactive monitoring and management. This approach enhances patient safety and can lead to the identification of biomarkers for adverse drug reactions, further informing drug development and regulatory approval processes. In the UK, the Yellow Card Scheme (run by the MHRA) collects and monitors information on suspected safety events involving healthcare products. In collaboration with Genomics England (GEL), the MHRA have launched the Yellow Card Biobank (17) – this aims to understand further how an individual genetics may increase their harm from side effects . Patients who have experienced an adverse drug reaction following one of a number of medications can consent to their DNA being tested via GEL to help understand the genomic contribution to this.

Role of Genetic Data in Clinical Trials, Drug Discovery & Targeted Therapy

With a high attrition rate in drug development (over 90% of drugs fail in development (18) methods are required to improve success and it has been shown that the probability of success of a drug in the clinic is over twice as likely with genetic evidence (19) . 

In 2013 the FDA published guidance for the industry on clinical PGx and early-phase clinical studies (20). Both the European Medicines Agency (2018) and Food and Drug Administration (2013) have recommended collection of DNA samples across all phases of clinical development. Genomic characterization of clinical trial subjects is now commonplace, with ~80% of I-PWG (Industry Pharmacogenomics Working Group) members reporting use of next-generation sequencing (21). Whilst relatively little of this is focused on traditional PGx markers of metabolism, there is increasing use for the other area of precision medicine, biomarkers (targeted therapy). The majority of genetic characterization of trial participants is in oncology studies, with other areas cardiovascular, neuroscience, immunology and rare diseases. 

A review of trends in USA FDA approval of new drugs with PGx guidance found that the annual proportion of new drug approvals with PGx labeling increased from 10.3% (n=3) in 2000 to 28.2% (n=11) in 2024 (22). Cancer therapies was the area which saw the biggest growth. 

Incorporating pharmacogenomic data into clinical trial design enables personalized approaches to drug testing. By identifying genetic subgroups that are more likely to respond to a particular treatment, researchers can design more efficient and targeted trials, reducing the time and cost of drug development.

Patient Stratification

Genetics can aid in stratifying patients, ensuring that clinical trials include participants who are most likely to benefit from the investigational drug. This stratification enhances the trial’s statistical power and increases the likelihood of observing significant treatment effects. For example, in oncology for non-small cell lung cancer, EGFR mutations were the strongest predictive biomarker for progression free survival and tumor response to first-line gefitinib versus carboplatin/paclitaxel (23) 

For patients with rare diseases, until the evolution of genetic testing over the last 10 years there was relatively little targeted therapy available. Now, about 10% of rare diseases have a targeted therapy available (24) . Some of these are specifically designed therapies and others are a result of drug repurposing (reviewed in (25)). Therapy can even be given in utero (26) . Identifying the molecular cause of disease is crucial for development and access to these therapies. As whole genome sequencing becomes more common, such patients will also benefit from the use of their genetic data to avoid side effects from more generic drugs such as pain relief which are used to manage their symptoms.

Conclusion

Genomics holds immense potential to transform healthcare, both by PGx and reduced adverse reactions and targeted therapies and enabling personalized medicine. Recent innovations, the integration of PGx in clinical trials, and collaborative efforts among researchers, clinicians, and industry partners are driving this transformation. Continued investment in pharmacogenomic research and development will pave the way for more effective and safer drug therapies, ultimately improving patient outcomes.