The emergence of each drug on the market is preceded by a complex and lengthy process. The newly synthesized molecule will become a medical preparation no earlier than official authorities will allow its use for the treatment of a specific disease in a certain group of patients. Such a decision can be made only on the basis of an analysis of sufficient information about the effectiveness and safety of the new drug. The rules of the Food and Drug Administration (USA) require that at least two basic clinical studies (pivotal studies) of its effectiveness and safety are conducted and successfully completed for each new drug. In addition to the results of these studies, other information should also be provided – for example, chemical properties, production features, toxicological data, etc. But the main emphasis is on the analysis of efficiency and safety. Why, then, does each new application for drug registration in the US (New Drug Application) include not 2, but on average 8 to 12 basic studies of efficacy and safety? And why, despite the huge work done, a significant number of applications for registration still rejected? The answer is simple: inadequately planned clinical studies at the earliest stages led to misdirection of tasks for subsequent baseline tests, inadequate selection of patient groups, dosage regimens, etc. Perhaps the pharmaceutical company had to repeat some studies. Therefore, it is extremely important to plan the whole clinical research program of a new molecule competently. Errors in the development of research schemes in the early phases will lead to incorrect conclusions and will multiply many times during clinical trials of later phases, which will adversely affect the fate of even a potentially good drug. Here it is necessary to make two digressions. First, we should not forget that the use of the drug in humans is preceded by a huge number of toxicological, pharmacokinetic and pharmacodynamic studies in animals. The quality and completeness of these studies undoubtedly determine the further fate of the molecule. Therefore, all of the above about the need for careful planning of the clinical trial program fully applies to preclinical research. Secondly, in this article we are only talking about the proper planning of research. No less, and perhaps more important, is the competent performance of the clinical trial itself, adherence to protocol requirements, correctness and honesty in the collection and analysis of data.
Phases of clinical trials
Typically, the drug passes through four phases of clinical trials; the second phase is divided into phases IIa and IIb, and the phase IIIb is separated within the third phase.
Phase I. The first experience of using a new active substance in humans. Most often, studies begin with volunteers (adult healthy men). The main goal of the research is to decide whether it is worth continuing work on a new drug and, if possible, to establish doses that will be subsequently used during the Phase II trials. During Phase I, researchers receive preliminary data on the safety of the drug and make the first description of its pharmacokinetics and pharmacodynamics in humans. The first phase studies are a wide range of medical experiments. They usually continue when the II, and sometimes the III, phase of the test is started (usually all pharmacokinetic studies are referred to as phase I).
During Phase I trials, the following is investigated:
- Safety, tolerability, pharmacokinetics (PK) and pharmacodynamics (PD) of a single dose (including the maximum tolerated dose). 2. Safety, tolerability, FC and PD multiple doses. 3. Bioavailability. 4. Proportional PK and PD of one dose and multiple doses for different routes of administration. 5. Metabolism of the drug and its relationship with body weight. 6. The influence of age, sex, food, liver function and kidney on the FC and PD of one dose and multiple doses. 7. Drug interactions.
Phase I studies have common features:
- They are attended by a small number of volunteers; an average of 4 to 24 people (up to 80 people throughout the I phase). 2. Each of the studies is conducted in one center. 3. Each study lasts a few days, a maximum of several weeks. 4. Carefully monitored by medical personnel; usually volunteers are monitored 24 hours a day.
Sometimes the increased toxicity of the drug (for example, for the treatment of cancer or AIDS) makes conducting such studies in healthy volunteers unethical. Then they are conducted with the participation of patients suffering from the corresponding disease. Usually, these non-therapeutic studies are conducted in specialized institutions.
Phase IIa. This is usually the first experience in patients with a disease for which a drug is intended to be used. Sometimes such studies are called pilot, since the results provide optimal planning for more extensive and expensive pivotal IIb phase studies. During phase IIa, it is necessary to verify the activity of the test substance, to assess short-term safety, to establish the patient population, the dosage regimen, to determine the dependence of the effect on the dose, to determine the criteria for assessing efficacy, etc. Tests are carried out on a limited number of patients (100-300), followed closely by observation, sometimes in a hospital.
Phase IIb. More extensive studies in patients with the disease, which is the main presumed indication for the purpose of the drug (for treatment, diagnosis or prevention). The main goal is to prove the effectiveness and safety of the new drug. The results of basic research serve as a basis for planning phase III studies and significantly influence the decision to register the drug. Many consider Phase II studies to be the most important moment in creating a new medicine.
Phase III. Multicenter trials involving large (and, if possible, diverse) groups of patients (average, 1000-3000 people). Recently, the term “megatrials” has appeared, in which more than 10,000 patients can participate. Phase III studies are conducted to obtain additional data on the safety and efficacy of various forms of the drug. Phase III studies the nature of the most frequent adverse reactions, clinically significant drug interactions, the effect of age, concomitant conditions, etc. Typically, clinical trials of this phase are double-blind, controlled, randomized trials. The research conditions are as close as possible to the normal conditions of application of the preparation. The data obtained in the clinical trials of the III phase are the basis for creating instructions for the use of the drug and an important factor for the decision by the official authorities to register the drug and the possibility of its medical use. The IIIb phase of clinical trials is singled out, which includes studies that take place during the period from the submission of materials for registration of the drug to official authorities until the moment of registration and obtaining permission for medical use. They are conducted in order to obtain some additional information about the drug, to assess the quality of life, the position of the future drug on the market, etc.
Phase IV. Studies are conducted after the start of the drug sale in order to obtain more detailed information on safety and efficacy, various dosage forms and dosages, long-term use in various patient groups and at various risk factors, etc., and thus more fully evaluate the strategy of application medicinal product. A large number of patients take part in the research, which allows revealing previously unknown and rare undesirable phenomena. There is the concept of postmarketing surveillance; these non-experimental studies-observations are sometimes referred to the V phase of clinical trials. After the registration of the drug, clinical trials, the purpose of which is to study new, unregistered indications, methods of application or combinations, are considered as trials of a new drug, i.e. are considered studies of early phases.
Study design
Design is a scheme, a template, a general outline of the research, its organizational framework. If we imagine the study as a river flowing in the valley, then the design will describe where the flow is to flow, determines whether it’s going straight or winding, where a single channel breaks into sleeves and rejoins where the water will go underground, and then appears on the surface, and a point that can be reached. None of the design types a priori has advantages over others. Everything depends on the tasks of the specific study. The correct choice of design determines the success of the test.
Observation and experiment. In the research-observation, the researcher does not interfere in the events, as if from the side he analyzes their natural course. For example, two groups of people are selected, one of which has risk factors, and the other does not. For a certain time, without any intervention, the frequency of occurrence of cardiovascular diseases in both groups is estimated. In the experiment, the researcher actively intervenes in the events, for example, appoints a certain treatment to two groups of patients and analyzes the results. Most clinical trials are experimental.
Retrospective and prospective studies. In retrospective studies, the events that have already occurred are evaluated. For example, patients who have had a heart attack, patients who received and did not receive any type of drug are selected, and mortality in the two groups is analyzed. In prospective studies, a research plan and the order of data collection and processing are first compiled and then the upcoming events are analyzed and analyzed. For example, it is decided to select suitable patients with myocardial infarction and some of them to prescribe a new drug, and then compare the mortality between groups of patients who received and did not receive treatment. Today, almost all clinical trials are prospective. A retrospective study should be conducted only if prospective is not possible. This is due to the fact that too many factors can negatively affect the reliability of the results of a retrospective study: the absence of a systematized, pre-planned approach to the distribution of patients between groups; the possible dependence of the outcome in a particular case on additional factors, of which it is already impossible to know; it is very difficult to check whether the patient’s examinations were performed correctly at the time of the examination, etc. Therefore, despite the high cost, duration and complexity, the research should be prospective – it is a payment for the reliability and quality of the data received.
Cross-sectional studies and prolonged studies. In the “cross section” studies, each participant is examined once. For example, patients with a risk factor are selected and the number of them has an interesting disease. Typical examples of “cross-section” are various questionnaires. The Health and Nutrition Examination Survey, conducted in the United States in 1973, for many years was the basis for a variety of analyzes – from the prevalence of hypertension to daytime fat intake. In extended studies, participants are examined more than once, i.e. observed over a period of time. Most clinical trials relate to prolonged studies and sometimes last for many years. A classic example is the famous Framingham study.
Non-comparative and comparative studies. In non-comparative clinical trials, the study treatment is not compared with anything. In this case, either methods of descriptive statistics are used to establish observations (for example, “at the end of the treatment with the study drug, the normalization of BP occurred in X patients out of a total of Y included in the study, which is Z% of Y), or analyze the dynamics of any criterion in one group of patients (for example, “at the beginning of treatment with the study drug, the average value of diastolic blood pressure was X mm Hg, at the end of treatment – Y mm Hg, the decrease in blood pressure was significant with probability p.
Controlled studies. In a broad sense, these are studies conducted in strict accordance with a carefully planned protocol and monitored by the monitor, ethics committee and official authorities. In the context of this article, another, “narrow”, value, according to which the study is called controlled, is of greater importance if the test drug is compared with a control (treatment with already known efficacy and tolerability). For example, comparing two previously unexplored doses of a new drug in two parallel groups can not be attributed to controlled studies in this sense, as the efficacy and tolerability of both treatments is unknown, and comparing a higher dose with a dose already well studied in this group of patients, can. In addition, there are placebo-controlled studies (see below).
One, two or more patient groups. In a study with one group of patients, all participants receive the same therapy. If criteria are set for changing therapy under certain conditions, then at the end of the study, one group may be split into two or more. For studies involving two groups of patients, parallel and cross designs are most common. In a parallel study, one group of patients from the beginning to the end of the study receives one therapy, and the second group – another. In a cross-sectional study, each group receives both types of therapy at different times, for example, the first group of patients first receive drug A, then preparation B, and the second group first drug B, then preparation A.
Advantages of parallel design before cross: 1. Parallel research can be completed faster, as in each group there will be only one treatment period. 2. The quality of the data from the parallel study is more “stable” to protocol violations, for example, visits by patients, their dropping out of the test, etc. 3. Cross-sectional design can be used only in patients with stable long-term course of the disease, as their condition should be approximately the same before the beginning of both periods of treatment. 4. In a parallel study, there is no effect of the previous therapy (carryover effect), when treatment with the first drug affects the results of treatment with the second. Disadvantages of parallel design in comparison with cross: 1. For conducting a parallel study, more patients are required. 2. Parallel design assumes greater variability of data, as different patients receive different therapies.
To reduce the impact of previous therapy in cross-sectional studies between treatment periods, various drugs usually have a washing-out period. During this period, patients do not receive any therapy and their condition approaches the initial; In addition, it is possible to avoid possible drug interactions between the drugs being studied. Sometimes the washing period occurs at the beginning of the study in order to minimize the effects of the previous treatment, and, less often, at the end of the study. The washing period at the end of the study is introduced in order to assess the patient’s condition some time after the end of the drug intake or avoid the interaction between it and subsequent conventional treatment.
Control groups. The comparison group in comparative clinical trials is called a control group. Control can be:
- Placebo.
- Other active treatment.
- A group that does not receive any treatment.
- Another dose of the same drug.
- The group receiving “usual care” (usual care); this treatment is not strictly stipulated by the protocol; This is in contrast to the group “other active treatment”, where comparative therapy is clearly defined by the protocol.
- Comparison with the anamnestic data of the same patients.
- Comparison with the anamnestic data of other patients.
Conclusion
Depending on the latitude of the circle, which is in the dark about the therapy in a particular patient, the following studies are distinguished:
- Open (open) – the doctor and the patient know which therapy is prescribed.
- Simple blind (single-blind) – the patient does not know what treatment is assigned to him, and the researcher has such information.
- Double-blind – neither the doctor nor the patient knows the prescribed therapy.
- Triple blinds – neither the doctor, nor the patient, nor those who organize the research and process the data do not know the intended therapy in a particular patient.
There is the term “total blind”, when everyone who interacts directly with the patient, the researcher and the data is in ignorance of the treatment, for example, radiologists, pathologists, statistics, etc .; while all parties do not have information about the type of treatment until the analysis is completed. Codes controlling the distribution of patients between groups are prepared by individuals not associated with this study. During double and triple blind studies, closed codes are located in each research center, which, in case of medical necessity (for example, in the event of some serious adverse events) can be opened to determine what therapy the patient is receiving.
As many studies as possible should be double or triple blind. Exceptions include Phase I pharmacokinetic studies, early phase II studies, long-term Phase II and III studies, which are usually a continuation of double-blind studies, and studies that can not be double-blind or triple-blind for ethical reasons. This can be illustrated by three examples taken from B.Spilker. One of the antidepressants was tested in Parkinson’s disease in two studies, one of which was open, and the other was double-blind, placebo-controlled. In an open study, 67% of patients felt improved, while in a double blind, only 25%. In summarizing the experience of a significant number of studies in psychiatry, it was found that a positive response to therapy was 83% in open trials and only 25% in controlled trials. A group of scientists analyzed the effect of zinc on reducing taste sensations. In a simple blind study in patients who did not have a placebo effect, there was a significant improvement in the treatment with zinc. However, a double-blind, cross-over study showed no difference between zinc and placebo. In addition to the term “blind”, there is the term “masked” that has the same meaning. In accordance with tradition, it is used, for example, in clinical trials of ophthalmic drugs.
