A study involving 100,000 people shows that they all carry genetic variants that may influence their response to medicines

The article is of great interest as it confirms, based on the largest number of patients to date, what all of us working in pharmacogenetics already knew: we are all carriers of some variant that may be useful in improving the efficacy and/or safety of medicines we are already taking or may take throughout our lives.

High-quality studies such as this one, carried out under a reputable initiative such as All of Us, can provide part of the impetus needed for both doctors and patients to understand the importance of pharmacogenetics and to capitalise on the enormous advantage we have in Spain, as we are pioneers in Europe in including pharmacogenetic tests in the catalogue of our public health system.

The article by Haddad et al., a high-quality retrospective observational study, analyses how genetic differences between individuals may influence their response to medicines. To this end, the authors worked with whole-genome sequencing data from 98,590 subjects and electronic clinical records from over 372,000 participants in the US All of Us programme, one of the most diverse and extensive cohorts currently available for pharmacogenomic research.

From a methodological perspective, the study’s main strength—in addition to the use of validated tools for haplotype assignment, the application of appropriate quality controls and systematic comparisons with reference genomic databases—lies in its enormous sample size and the researchers’ success in combining whole-genome sequencing with real-world data on prescription and medication use derived from electronic health records. This approach allows us to move beyond a simple genetic description and get closer to the reality of clinical care.

The results presented in this study are largely consistent with the scientific evidence currently available, in that they confirm that pharmacogenetic variability is extremely common and has clinical relevance in the use of numerous medicines. In fact, almost all participants in the study (99.5 per cent) carried a pharmacogenomic variant which, according to current recommendations, could justify an adjustment to their drug treatment. Furthermore, approximately one in four individuals had already been prescribed a medicine for which pharmacogenomic recommendations applicable to their genetic profile existed. These findings are fully consistent with the scientific evidence accumulated in recent years and position pharmacogenomics as a tool increasingly supported by clinical trials and international guidelines, particularly regarding the use of certain therapeutic agents, such as antiplatelet agents and anticoagulants, proton pump inhibitors or certain cancer treatments.

Perhaps the most novel contribution of the study is its ability to highlight population groups traditionally under-represented in genetic research. The authors have identified more than 17,000 previously undescribed variants and demonstrated that some genetic frequencies (allelic and phenotypic) used as international benchmarks may not adequately reflect the reality for people of African or Native American descent. This observation is of great scientific and ethical significance, as it highlights the risks of developing personalised medicine based on databases that are overly focused on European populations and fail to adequately reflect the genetic diversity of other populations.

The clinical implications of the study are also significant. The results reinforce the advisability of carrying out pharmacogenomic testing before or during treatment with widely used drugs, such as omeprazole and other proton pump inhibitors, atorvastatin and other statins, warfarin, codeine, tramadol or other drugs metabolised by enzymes such as the CYP2D6 and CYP2C19 isoforms. In practical terms, wider use of these tests could reduce adverse reactions, improve therapeutic efficacy and promote more personalised prescribing.

However, the study also has significant limitations which mean the results should be interpreted with some caution. The study demonstrates the potential of pharmacogenomics, but does not directly prove that the use of this information has improved participants’ health, as it is a retrospective analysis evaluating previous drug exposures. Furthermore, some complex genes were excluded from the analysis and certain population groups remain under-represented. Finally, electronic records may not capture the full extent of participants’ actual drug exposure.

In conclusion, this article represents a significant contribution to the field of pharmacotherapy, with a very robust methodological framework that confirms current trends in personalised and precision medicine, the main contribution of which is to improve pharmacogenomic knowledge of diverse populations and to strengthen the scientific rationale for the clinical implementation of pharmacogenomics in healthcare systems.

The fact that around 25 per cent of participants in the All of Us cohort have actionable pharmacogenetic variants for a particular drug is significant and supports the need for cost-effectiveness studies to determine whether pharmacogenetic screening strategies and alerts in medical records might prove useful in the future.