Tablet to order
How and why do they create individual medicines
Dmitry Sychev, Post-science
A situation in which a patient is prescribed one drug after another and by trial and error they find out which one will work, in the scientific literature is called a reactive model of medicine: a person reacts to the drug in some way, and doctors already react to the patient's reaction. Clinical pharmacologist Dmitry Sychev told post–science why it is important to gradually move to a different way of interacting with the patient, how the foundations of personalized medicine were laid and on the analysis of which indicators – biomarkers - it is based.
How did horse beans lead to new medicine and why is PCR needed?
According to legend, the ancient Greek philosopher Pythagoras noticed that people reacted differently to horse beans, which were often eaten in the ancient world. He noticed that the reaction in the form of anemia was hereditary: if there was a person in the family who was poorly digesting horse beans, then his relatives also had intolerance. At the beginning of the XX century, scientists found out that such a reaction is associated with an innate (genetic) deficiency of a certain enzyme – glucose-6-phosphate dehydrogenase (G6FD), which leads to the development of hemolytic anemia, that is, anemia associated with increased destruction of red blood cells.
This hypothesis was confirmed in the 1940s, when during the military campaigns of European states and the United States in the southern countries, soldiers were given preventive antimalarial drugs, for example "Primakhin". Representatives of certain races and their relatives developed hemolytic anemia associated with G6FD deficiency. This hereditary deficiency is spread in different ways among ethnic groups, but there is a pattern: its frequency increases from north to south. In the USSR, where this problem was actively studied, most people with congenital G6FD deficiency were found in the republics of Transcaucasia, especially in Azerbaijan. Today there is a list of drugs that are not recommended to be given to people with such a genetic deficiency and that cause hemolytic anemia.
In 1952, the drug isoniazid began to be widely used, which revolutionized the treatment of tuberculosis and made it possible to avoid surgical intervention (previously, this disease was dealt with only by removing a part or a whole lung). However, doctors noticed that isoniazid acted on people with different effectiveness, and its concentration in the urine of patients also differed. It turned out that this is due to the unequal rate of acetylation, that is, the chemical transformation of isoniazid, and depending on the ethnicity of all people can be divided into slow and fast acetylators of isoniazid. By the way, the frequency of rapid acetylation also increases from north to south, and in European ethnic groups of fast and slow acetylators is approximately equal. At the same time, slow acetylators of isoniazid, as a rule, are blonde with light skin and blue eyes, that is, pharmacogenetic signs may be linked to phenotypic signs. Now doctors are trying to vary the dosage of isoniazid depending on the genetic characteristics of each patient.
Gradually, the peculiarities of the effect of drugs on different people began to be associated with genetic factors. As a result of active research of genes regulating the action of certain drugs, pharmacogenetics was born, which in the 1950s became a separate discipline.
Already in the 1960s and 1970s, scientists began to use the twin method to understand whether there are hereditary (genetic) factors that can influence the effect of a particular drug. To do this, they compared what chemical and biological processes occurred with the drug in the twins' body, and based on the "proximity" of the parameters of these processes, they concluded about the influence of heredity.
In the USSR, Academician Vladimir Kukes and his students have been actively engaged in the problem of identifying pharmacogenetic phenomena since the 1980s. They were able to identify special enzymes in the liver that are involved in the processing of active components of drugs. The presence of these enzymes influenced the choice of a drug suitable for a person and its dosage.
In 1983, American biochemist Cary Mullis invented the polymerase chain reaction (PCR), which in the 1990s began to be used to identify human genetic characteristics, which means that it was now possible to predict the response to drug therapy. For pharmacogenetic purposes, PCR was one of the first to be used by Professor Evgeny Schwartz, who studied gene mutations associated with sensitivity to anticoagulants – blood thinning drugs, in particular to warfarin.
How does PCR work?
The PCR method is based on multiple selective copying of the desired DNA site. To perform PCR, a blood sample is taken from patients and amplification is carried out using in vitro enzymes ("in vitro"), that is, a certain section of DNA is repeatedly doubled, making it visually accessible. Then this section of the gene is examined to understand whether a person is a carrier of mutations responsible for certain reactions to the drug.
Drawings by Katya Zolotareva
As an alternative to PCR in Russia used phenotyping methods: the concentration of drugs in urine and blood was determined and the rate of drug processing was determined by them, by which it was possible to judge how well it was absorbed. Today, these methods are called therapeutic drug monitoring and are used mainly in scientific research.
What's wrong with traditional medicines?
Traditional medicines are created to treat a disease, but from 38 to 75% of new drugs are ineffective and have any consequences for the patient. Before being introduced into clinical practice, any medicine must go through a difficult path of preclinical tests on tissues and experimental animals, as well as clinical studies, the results of which are eventually averaged. In fact, during drug trials there will always be patients who respond well to it, and for others this medicine will be ineffective or cause them undesirable reactions.
In such a study, there is a statistical indicator – NNT (number needed to treat), showing how many patients on average need to be "treated" with a new drug compared to the old one in order to avoid one heart attack, stroke or death. For example, if NNT is equal to 50 for some drug, it means that 50 patients need to be transferred from the old drug to the new one in order to avoid one death, but whose exactly is unclear. The essence of medical personalization is to identify exactly those patients for whom this drug will really help and who will not cause side effects.
The pharmacological response of different patients to the same medications depends on genetic and non-genetic factors. The latter include, for example, gender and age. Adverse reactions, especially allergic ones, are more common in women than in men, and with age, many people have impaired liver and kidney functions, drugs are worse excreted from the body and more often cause undesirable effects. Medications can act differently depending on whether a person smokes, whether he drinks alcohol, what diet he adheres to and what other medications he takes.
The approach to patient management, based on taking into account a set of individual characteristics, formed the concept of personalized medicine, that is, prevention, diagnosis and treatment based on indicators that can be measured and that reflect the individual characteristics of the person himself or the pathological process. Such indicators are commonly called biomarkers. They can be clinical (for example, heart rate, pulse), biochemical (blood clotting index, markers of protein, carbohydrate and lipid metabolism, indicators of the functions of various organs) and genetic (DNA markers). Some biomarkers predict the effectiveness of the response to medications, while others, on the contrary, predict the ineffectiveness of treatment, resistance to certain drugs or techniques, or the development of adverse reactions and complications.
What can genes say about the reaction to drugs and why do we need a genetic passport?
Genetic factors of the response to drugs are features inherited from our parents, grandparents, as well as intolerance to horse beans. Modern genetic tests, namely pharmacogenetic ones, make it possible to identify specific genotypes of patients associated with a pharmacological response to a particular drug. Blood, buccal epithelium (inside of the cheek), hair, saliva are suitable as genetic material.
Genetic markers can determine the response even to some vaccines. A separate promising area – vaccinomics - is devoted to the study of this phenomenon. It has been proven, for example, that there are genetic markers that determine the best immune response to rabies vaccination. Today, based on genetic research, dietary recommendations and tips on exercise are also being developed. This is done by nutrigenomics, a science that studies the effect of nutrition on gene expression. There are also many speculative commercial stories today. For example, there is a company in Europe that will select the most suitable red wine for you based on the analysis of your genotype and even send you a bottle. This direction is called vinomika.
Pharmacogenetic testing today in the vast majority of cases is carried out using PCR and is available to almost everyone. Molecular genetic tests predicting sensitivity to a particular drug are done once for a lifetime. By accumulating this data, it is possible to form a personalized pharmacogenetic map of the effectiveness of drugs, which in the future will become a genetic passport.
In addition to PCR, there are a number of new methods of molecular analysis today. The basis of modern personalized medicine, primarily pharmacogenomics, which studies the complete DNA sequence in genetic material, is the use of genomics, transcriptomics, proteomics, metabolomics and microbiomics. These technologies are called omics, since in English their names end in -omics (genomics, transcriptomics, and so on). Their other name is postgenomic technologies.
Transcriptomic technologies are aimed at studying the work of certain genes, on the basis of which it is possible to judge the effect on the body of a particular drug. Proteomics deals with the protein level: identification of new proteins, analysis of their content in plasma, urine, and so on. With the help of metabolomics, a prognosis of the response to a drug is given based on the analysis of a set of metabolites, that is, intermediate and final metabolic products of various compounds: carbohydrates, proteins, lipids, nucleic acids, as well as small organic molecules – in urine, blood, and so on. Microbiomics develops approaches to personalization depending on the microbiota of the intestine, mucous membranes or skin.
Compared to the PCR used in pharmacogenetic testing, pharmacogenomics uses more expensive next generation sequencing methods (NGS), chromatographic methods, mass spectrometric detection. So far, these technologies are used only for research purposes, but perhaps in the future they will find practical application and will complement each other, increasing the accuracy of predicting the response to drug therapy and thereby bringing us closer to the precision model of medicine.
If in 2018 the Russian Ministry of Health talked about the concept of medicine "three P" – preventive, personalized and predictive – today we are talking about the medicine of many "P": it is believed that it should be personalized, predictive, preventive, participatory and so on, and most importantly – precision, that is, medicine, in which a biomarker is an accurate diagnostic criterion predicting the choice of appropriate treatment.
How to create individual drugs
The main goals that scientists pursue when creating personalized drugs are to increase the effectiveness of medicines and reduce the number of adverse reactions. The drugs are created immediately together with biomarkers that will indicate the achievement of both goals. The drug–biomarker bundle should be tested in large, well-planned studies.
In addition, pharmacogenetic biomarkers are being developed not only to personalize new drugs, but also to form a similar approach to already known drugs. It is thanks to the concept of personalization that many old medicines have been "rehabilitated". The search for adequate biomarkers has shown that not everyone needs to prescribe a new generation of drugs, because there are people who respond well to old drugs without undesirable side reactions.
Theoretically, any drug can be personalized. But in practice, before taking painkillers, if you have, for example, a headache, you do not always need to go to a genetic laboratory. Most experts believe that this kind of testing is necessary only for highly effective, but problematic drugs with a wide range of side effects. Also, pharmacogenetic tests can increase the safety of treatment in patients with high risks of adverse reactions: elderly people, patients with concomitant diseases, polymorbidity (i.e. synchronously occurring diseases) of the liver and kidneys, people taking many medications at once.
Probably, in the future, individual drugs will be created for each person. There are already such examples, for example, therapy of rare genetic diseases. However, most likely, this personalization will be based on large studies that will take into account a variety of genetic and non-genetic factors. Based on them, a model will be built that predicts the drug response. Special programs – decision support systems – will predict for each patient which drugs will work best in his particular case. These programs will be based on big data – accumulating DNA biobanks containing all the information about a person: what is sick, what drugs cause adverse reactions and so on. Such registers already exist now and are created according to a certain methodology – accordingly, the information in them is updated and analyzed using machine learning. Of course, people consent to the use of their genetic material for this kind of research.
From the standpoint of pharmacoeconomics, which implies an assessment of the economic feasibility of a particular treatment method, pharmacogenetics is very beneficial, because it can tell who will benefit from cheap old medicine, and who needs more expensive modern drugs. There is a whole system of clinical and economic evaluation of healthcare technologies, tied to pharmacoeconomics and studying how a drug or test moves from development to clinical practice. There are also models that take into account regional peculiarities of the ethnic composition of the population and the prevalence of certain gene mutations.
Difficulties in implementing personalized medicine, oncological diseases and COVID-19
Unfortunately, personalization technologies, especially old drugs, and their study in well-planned studies are not very well funded. The fact is that pharmaceutical companies manufacture medicines and make sure that as many people as possible are prescribed them. It is unprofitable to invest in personalization, because this approach assumes that medicines are prescribed only to a narrow circle of patients. Therefore, research in the field of personalized medicine is mainly sponsored by various foundations, including state ones. Due to lack of funding, these studies do not have the necessary scale. But it is possible to strengthen the evidence of personalized methods with the help of systematic reviews and meta-analysis, and then genetic testing will be included in clinical recommendations, and therefore, in the future, in treatment standards.
The slow pace of the introduction of personalized medicine is not only due to economic factors. There is a problem of availability of pharmacogenetic tests and insufficient awareness of doctors and health care organizers about the need for such testing. Therefore, it is important that appropriate competencies are formed in educational organizations and professional communities that will allow doctors to understand when it is necessary to use personalized tests and drugs and how to use them correctly. It is proved that if during the training of doctors, in addition to traditional formats (lectures, seminars, etc.), active training is used, they will more often apply pharmacogenetic tests in their practice. To do this, it is possible to conduct molecular genetic testing by the students themselves – of course, with their consent – and analyze approaches to personalization using the example of their own results.
The personalized approach in practice is now used primarily in the treatment of oncological diseases, especially those associated with a poor prognosis and a rapid course. The appearance of targeted drugs is nothing more than the personalization of medicines aimed at a key target in carcinogenesis, that is, the process of tumor formation. Increasingly, a personalized approach is being used in cardiology to prevent primary and secondary cardiovascular complications, especially thrombotic ones. The appointment of psychotropic drugs also requires personalization.
During the COVID-19 pandemic, there was also a need for a personalized approach to treatment. At the very beginning of the pandemic, therapy for coronavirus infection was based primarily on an empirical, pathogenetic, symptomatic approach and literally hints at evidence that a particular drug can help. New recommendations are gradually emerging in all countries, approaches to drug therapy change depending on the severity of the disease and the characteristics of the course – these are also biomarkers that you can focus on.
Pathogenesis is the mechanism of the origin and development of diseases and their individual manifestations. It is considered at various levels – from molecular disorders to disorders in the body as a whole. By studying the pathogenesis, doctors identify how the disease develops.
Most likely, there will be no universal drug for coronavirus infection. Many doctors notice that some medications work only for a certain category of patients. As recent studies show, there are indeed gene mutations associated with a more severe course of the disease and a poor response to drug therapy. Therefore, today there is a search for molecular genetic biomarkers in the field of COVID-19 treatment. By the way, such biomarkers already exist for anticoagulant therapy, for example, D-dimer.
The transition from a reactive model of medicine, when doctors prescribe medicines by trial and error, to a precision, that is, an accurate model, allows you to reduce the selection time and immediately prescribe the drug that is most likely to work efficiently and safely. However, modern technologies of personalized medicine do not cancel taking into account the patient's clinical data. It's just that today, a whole set of other biomarker tools can now be added to the patient's medical history, gender, age, body weight and complaints.
In addition, pharmacogenetics and personalized medicine should in no case be opposed to evidence-based medicine based on randomized trials. Pharmacogenetic tests are also tested, all the "millstones" of evidence-based medicine are passed, and only after that they are integrated into the system of clinical recommendations and treatment standards.
About the author: Dmitry Sychev – Doctor of Medical Sciences, Professor, Professor of the Russian Academy of Sciences, Corresponding Member of the Russian Academy of Sciences, Head of the Department of Clinical Pharmacology and Therapy, Rector of the Russian Ministry of Health.
Portal "Eternal youth" http://vechnayamolodost.ru