In this lecture, Professor Julia Stingl outlines the evolution of pharmacogenetics from early observations of genetic variation (e.g., Neanderthal factor V Leiden) to its current role in drug labeling and regulatory practice. She explains the multiple sources of genomic variability that affect drug response—germline, tumor, and microbiome—and why precision‑medicine regulation must shift from population‑wide efficacy/safety assessments to stratified, patient‑specific evaluations. Stingl describes how pharmacogenetic information appears in drug labels (indication, safety, dosing) and what makes it actionable. She presents the limited set of ~21 pharmacogenes that cover most clinically relevant variability, the exponential rise of drugs requiring mandatory genetic testing (especially in oncology), and the differences among regulatory agencies (FDA, EMA, PMDA, Health Canada). The lecture reviews the Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines, the ongoing EMA referral for pre‑emptive DPYD testing before fluoropyrimidine therapy, and the European UPGx project aimed at harmonizing testing across countries. Stingl contrasts US and EU approaches to companion diagnostics, highlighting the upcoming EU Medical Device Regulation that will demand more clinical evidence for diagnostics. Finally, she discusses the impact of pharmacogenetics on drug development—basket and umbrella trial designs, the need to address biomarker‑negative patients, and the continuous post‑marketing obligations to update labels and diagnostics. The talk concludes with a call for ongoing regulatory adaptation to keep pace with genomic medicine.
Chapters
Hello and welcome to this lecture on pharmacogenetics from bench to drug label, a kind of translation of pharmacogenetics. My name is Julia Stingl, I'm a professor of clinical pharmacology at the University Hospital Aachen in Germany. Well, actually pharmacogenetics is not really from the bench. It's the observation of genetic variability with regard to xenobiotics and to xenobiotic defense. And this has been observed already in times dating back to the Neanderthal. Actually, the Neanderthal was analyzed in genome It was found out that the Neanderthal had a pharmacogenetic polymorphism, the factor V Leiden variant, leading to thrombophilia. Indeed, the Neanderthal obviously has used plants for healing, for treatments. Coumarins that are derived from plants have been identified in the dental calculus of the Neanderthal. Therefore, the factor V Leiden in this case may have a protective role. Otherwise, the Neanderthal would have died from bleeding complications of the coumarins. Pharmacogenetic variability is observed between humans all over the world, and not only between humans but also in animals and plants. And in all living organisms that have a need for xenobiotic defense systems and barriers. Pharmacogenomic variability normally refers to germline genome variability, but taken together to the huge variability that we see in outcome of treatments, All genome variability contributes: the germline, host genome, but also the tumor genome predicting individual reactions to tumor therapy, and the microbiome affecting gut metabolism and the interaction between liver and gut drug metabolism and elimination also contributes with a lot of SNPs and variants to the huge variability between individuals leading to adverse drug effects, therapeutic failure, good response, bad response, to remission or whatever, up to therapy resistance. The regulatory challenges of precision medicine are that we not only have to evaluate efficacy, safety of a given drug, but to evaluate the efficacy-safety ratio in individual patients, or at least in stratified subgroups of patients with similar features leading to benefit or toxicity. So, it's a kind of stratified regulation that we need for precision medicine, And the way, the weight is shifting more and more looking on patient variability within a given treatment, and determining the efficacy-safety ratio in stratified individualized patient population in regulation. Pharmacogenetic information is given in drug labels, and there are different parts in the drug label where pharmacogenetic information can be found. Either it is important for the indication of a given drug treatment, then it is probably obligatory information, making a pharmacogenetic test obligatory before the onset of treatment. Or in other parts of the drug label giving safety information, pharmacogenetics may be just with an informative character that there is a polymorphism and some patients may have a certain risk for side effects. And then on, at certain parts of the drug label, we have dosing information. And dosing information, information of the selection of a treatment, treatment duration, and so on. And affecting these parts of the drug label would result in so-called actionable pharmacogenetic information. Actionable means that if we have a patient in front of us who knows that the patient has a certain genotype affecting pharmacotherapy, we can do something, we can adjust treatment. Either we can, for example, reduce the dosage or take a different choice for combination drug or even decide for a different substance in order to treat the same indication. So actionable means that with the genetic information available, we can do something to individualize drug treatment. On the website of the Pharmacogenomics Knowledge Base, you see a comparison between the drug labels issued by the different regulatory authorities: the FDA, the European, the EMA, the PMDA in Japan, and the Canadian regulatory authority. And they differ a little bit in the amount of information of pharmacogenetics, with the FDA having the longest tradition of providing pharmacogenetic information in drug labels. But in principle, it is the same worldwide. We have this information available in the drug label, and we can look if we individually adjust drug therapy, or if we have to perform a test before onset of drug therapy, or if it's just an information where we can look up if a patient doesn't react as, as expected. We have now pharmacogenetic information in labels of about 200 different drugs, but compared to the information in drugs, it's quite a few, or a low number of genes that matter. And here you see it's about a two handful genes that should be looked at in order to adjust drug therapy according to pharmacogenetic information. And you see that the same genes being in drug metabolism or drug transporters impact different substances, different drugs, pharmacokinetic metabolism and elimination. So it's enough to genotype for those 21 genes in a panel genotype test, and then we have actually overall pharmacogenomic information available in a given patient. And this may hold true for all different kinds of drug therapy. As you see here, all fields of drug therapy are covered by these pharmacogenes. There are also more and more drugs and therapies on the market that have to be given in combination with an obligatory test, with a required genetic test, and you see here that the number of these drugs is increasing in an exponential way. And also here the FDA is a little bit advanced. Most of the drugs are in the field of cancer treatment, or immunomodulation. And you see that the EMA as well has relabeled more and more drugs for required genetic testing, most of it in the cancer field, but some of the indications also hereditary disease treatment or others. And of course, most of the therapeutic reason is to target or determine the target for a targeted treatment to get patient stratification, to get the indication, but also some of them are obligatory for avoiding side effects or even for dose adjustment. For clinical guidance, there are international guidelines available. For example, the Clinical Pharmacogenetics Implementation Consortium of the Pharmacogenomics Research Network is an international consortium that now has a quite long tradition issuing guidelines that provide clinical recommendations based on the available evidence for a given drug treatment on pharmacogenetic data. Here you see the members of the Clinical Pharmacogenetic Implementation Consortium. It's mostly a US-based or driven consortium, but with members from all over the world. And most members are from academia, from the hospitals, and university hospitals. This is how these guidelines look like. They evaluate the clinical evidence, the evidence from clinical studies on the pharmacogenetic polymorphism, and they classify the recommendation according to evidence. And the recommendation can be consider a dose reduction in a metabolizer genotype, for example, here consider 25% reduction of the dose in an intermediate metabolizer of CYP2D6, or to have a different choice for the drug treatment, consider a drug that is not metabolized, not affected by the genetic polymorphism. And the The available evidence for this recommendation is usually always rated beside, so that you can directly see the precision in the recommendation that lies within the recommendation. And the recommendations are formulated in that way that they can really be used in clinical practice. So it's not a pure academic comment, but it is thought to be used in clinical practice where the dosage can easily be halved or lowered, or where decisions on treatment strategies are directly drawn at the patient's bedside. Currently in Europe, we have an interesting referral ongoing which is almost decided, and this is in oncology on the treatment with fluoropyrimidines, which is a very common treatment in cancer therapy. For example, in colon cancer, it is standard therapy. So it is given to many patients and quite common as cytostatic treatment of cancers. And the EMA has started a referral within Europe for the first time to decide on pre-therapeutic testing in a large field of drug therapy including a pharmacogenetic test. And this is to test for DPYD activity, the enzyme that detoxifies fluoropyrimidines like fluorouracil or capecitabine, and thereby eliminates the active part of the drug. And about 10% in the population lack or have largely decreased activity of the DPYD enzyme. And the idea is to genotype patients or phenotype patients for DPYD activity before onset of treatment. And now at the end of April, this referral was decided by the EMA, by the CHMP, and is now open to the European Commission for final decision. And it will be a Europe-wide recommendation for testing DPYD activity prior to treatment with fluoropyrimidines. And this is really an innovation that within Europe we have now the same obligation to perform pharmacogenetic testing prior to treatment. The test is meant to test for activity or inactivity, decreased activity for DPYD. And what happens in patients who are deficient for the enzyme is that the toxic metabolites increase and the elimination of, um, fluoropyrimidines is largely decreased. And this metabolizing step is necessary for all of the fluoropyrimidines. As I said, they are commonly used in cancer treatment, but some of them are also used for skin treatment, but mainly for colorectal cancer, Capicitabine is widely used, and here are the names of the drugs that are affected by this decision. And this is just to give you a picture of the pleiotropic landscape within Europe. We face very different healthcare systems in Europe, And the countries differ partially very much with regard to availability of pharmacogenetic testing. Here are some countries where, which are quite advanced, like the Netherlands, in France, and in Italy, testing is already common use. For example, in Germany, this is not the case, and I'm excited to see that it has really to be a change in the landscape to implement this pharmacogenetic testing in everyday clinical practice here in Germany. But this will be a Europe-wide decision and it will be valid in all countries within Europe. And in order to harmonize the situation of pharmacogenetic testing within Europe, a European-wide project was installed some years ago, the Ubiquitous Pharmacogenomics, UPGx, project, and it is dedicated to harmonize the procedures and implementation of pharmacogenetic testing, not only in different countries in Europe, but also in very different fields of medicine and of drug therapy within Europe. So in principle, pharmacogenetics can be regarded as a pioneer in precision medicine, and it has been a long time since first genetic tests came out that were used in drug therapy in clinical practice until now where we have large sequencing platforms and especially in cancer treatment whole genome sequencing or at least deep sequencing of many tumor driver genes is done in a routine way. And we already have these platform units where we generate sequencing data that also can be used for data research, for basic research to learn new pathways, new molecular pathways in drug therapy. But it is also provided to patients to select therapy based upon the genetic sequence, and genome medicine is now a kind of companion diagnostics for personalized treatment, for personalized therapy. However, we still have differences in the regulatory systems between the United States and Europe.. And while in the United States a companion diagnostic can come onto the market together with a new drug treatment, together with a new therapy, this is not the case in the European Union. Here we have still drug approval by the regulatory authorities, which is independent of the development of pharmacogenetic testing or genome sequencing or whatever is used as companion diagnostic. And the companion diagnostic is regarded as a medical device, and we have a completely different regulatory framework for medical devices, so also for companion diagnostics. But this will change in the coming time because we have a new medical device regulation that has been— it should become effective this year in May, but due to the corona crisis, it was delayed for one year, so it will become effective next year, 2021. And here you see the main changes in this new regulation. Medical devices will need to have, or to provide more clinical data, more clinical evidence for benefit in general. And this is also true for the companion diagnostic. Clinical data that allow a clinical validation of the benefit of the test will be necessary in order to get a market authorization for the medical devices. And the regulatory authorities work together, the notified bodies being responsible for the medical devices and the EMA or the national competent authorities being responsible for drug authorization. They will have common consultations and they will issue a common opinion on a combination, on a companion diagnostic that is necessary to be performed in a certain drug therapy. So this will become more similar to the US American system, but it will still be independent. So it will be possible to market a companion diagnostic without a drug, or as in the example of 5-FU or capecitabine, in old drugs, in drugs that are already on of patent, but to develop only the companion diagnostic. However, this will need clinical data, clinical and patient outcome data in order to provide the evidence necessary for authorization. So in general, we will have the need for more clinical data in the future for all medical devices. Because the medical device regulation that will be effective then next year in 2021 is also valid for whole Europe in every country, and it means that in general for medical devices more clinical validity data will be needed compared to now where you can get a certificate with analytical, showing analytical validity, showing the proof of principles of your medical device. In the future, it will be necessary, more necessary, to perform clinical studies. And the collaboration between the regulatory authorities for medical devices and drug therapies will continue for the whole field of medical device. So coming to the end, what does pharmacogenetics mean for regulation? It means quite a lot, and a lot of challenges have been imposed on regulation, and the regulatory framework has developed in the same time that genome medicine and pharmacogenetics has developed. For drug development, it is important to have valid biomarkers that allow a stratified benefit-risk evaluation in subgroups characterized by biomarker profiles. And already in clinical drug development, specific study designs have emerged like the basket trial Basket Study or the Umbrella Studies, and they are now state of the art in personalized medicine drug development. And we, in the future, we will face the problem that we have to develop also therapeutic options for biomarker-negative patients. If you imagine that for all indications more and more biomarkers come into use, then we have, we will have patients falling out of the frame because they are biomarker negative. And drug development should also include those patients who are biomarker negative. So drug development should not only jump on the train for biomarker stratified therapies, but also think about the patients falling apart otherwise. So this will be a challenge for regulation to adjust for, yeah, having enough options, therapeutic options available for biomarker-negative patients. And after market authorization, it's not over. There's a lot of work that has emerged With the emergence of pharmacogenetics and genome testing in therapeutics, the drug labels have to be updated. We have to generate new information on safety, especially on side effect profiles that should be avoided. And the companion diagnostic, once it is obligatory, has also to be controlled, to be developed. There will be new methods. So even after market authorization, the challenges that pharmacogenetics created in drug development, they will hold on, and we will always have regulatory activities to adjust the framework to the development of more and more testing methods and stratification methods and coming to individualized medicine. Yeah, I hope that I could cover some of the aspects of pharmacogenetics affecting regulation and how the new findings from the bench find their entrance in regulation. Thank you for your attention.
