Medicinal products are an essential key to improving health outcomes. To ensure the safety, efficacy, and effectiveness of medicines, they are carefully investigated along the entire product life cycle. This comprises the steps of clinical development before receiving market authorisation and during pharmacovigilance, the post-authorisation system of surveillance. These investigations are some of the responsibilities of the regulatory and health technology agencies, to make decisions regarding medicines based on various reliable information.
Clinical trials, also named randomised controlled trials (RCTs), are a long-standing approach to evaluating medicines. They consist of scientific studies that test if an intervention (such as a new medicine) is safe and can produce the desired effect. To achieve this goal, the study randomly assigns the participants to a treatment group (receiving the new medicine) or to a control group (receiving either a placebo or a standard medicine).
This randomisation of patients is made under very controlled conditions, using a specified number of people (from large hundreds to thousands) that often don’t have other health problems than the one being studied. Unfortunately, some groups of patients that could potentially benefit from the new treatment are not always included in the clinical trial. For example, pregnant women and persons with multiple diseases are often excluded from clinical trials. This is highly problematic as these groups of patients will be part of the populations that will benefit from using the new medicine. Thus, this may leave some clinically vulnerable patients either without access to the drug or having to take the medicine without clear safety guidelines. In the case of rare diseases, it can be difficult to find enough persons with the exact condition for the study. Additionally, sometimes limited knowledge about the disease can be an obstacle to designing a proper study.
With many conceivably limiting factors, it can be difficult to generate evidence for every situation from clinical trials. Clinical trials test efficacy and safety under ideal conditions, but we can also obtain valuable information from real-world settings outside of clinical trials. And that’s why new data sources are being explored, aiming to produce information about effectiveness, i.e., the ability of an intervention to produce a desired result in the target population. The following figure displays exemplarily the opportunities of RWD to complement evidence from RCTs if heterogeneous data sources are collected and analysed with appropriate and user-friendly methods and tools. And this is exactly what Real4Reg is applying.
Real-world data (RWD) is collected from various sources outside a clinical trial or controlled setting. It can be electronic health records, insurance claims, patient registries, and other sources of health-related data collected in routine clinical practice. Thus, RWD includes data from patient groups who are often excluded in RCTs, e.g., pregnant women or very old people. RWD provide information about the usage of medicines in the real world, where, e.g., patients do not always correctly follow medical advice for various reasons, or drug replacements due to supply shortages are necessary. This helps us to know better how medicines work in real-life conditions and to ensure better decisions about them. RWD will not replace clinical trials but can add an additional layer of knowledge about medicines.
Artificial Intelligence (AI) refers to especially intelligent computer programs. Machine Learning is a type of AI dedicated to making programs that can “learn” from the data, gradually improving their results.
To use RWD and generate new evidence from it (Real-world evidence (RWE)) we may have to analyse massive amounts of data using advanced methods. Emerging methods such as Artificial Intelligence (AI) and Machine Learning (ML) can help to explore the data, find patterns and identify which are the best medicines for the different types of patients.