Slow and costly, clinical trials account for about 75% of the expenses related to the development of a molecule. Their design, implementation and patient recruitment protocols have hardly changed in decades. The Covid health crisis has shaken up this situation, demonstrating that it is more necessary than ever to deliver drugs and vaccines quickly.
The clinical research of the Covid-19 vaccine had to be set up quickly in the face of the emergency of the global health situation and the whole process had to be accelerated. In this unprecedented context, the organization of the trials was adapted, inevitably leading to many changes. The recent study conducted by IQVIA for Leem (The French Pharmaceutical Companies Association) reveals an acceleration in the digitization of clinical trials. Treatments have been sent to patients’ homes, teleconsultations have replaced on-site medical visits. Easy and quick to implement, digitization is now increasing hybrid clinical trials to 40% from 10% before the health crisis. This is a significant increase, but it leaves France far behind the United States, England and Spain. Another notable change is the review of evaluation procedures.
Usually carried out when the file is complete, this step has been carried out progressively while maintaining the expected level of requirements.
A beneficial change that has allowed the approval of innovative and effective vaccines in record time. In this field, AI can unquestionably optimize clinical trials. By analyzing population health data, patient medical records, and hospital data, AI is able to facilitate, but more importantly, better target patients to recruit for trials.
The in-silico clinical trial, an alternative model already validated
The clinical trial is the ultimate condition for obtaining a marketing authorization. The future treatment must go through the different phases of trials to measure and, above all, demonstrate its effectiveness. In addition to the in vitro and in vivo clinical trials used in clinical research, in-silico trials have been added to which AI gives a new dimension. The concept is based on the development of diseases, new therapies and virtual patient populations in the form of mathematical models and computer programs. The newly developed therapies are first tested by computer simulation on digital patient twins before being tested in conventional clinical trials. This method provides valuable information for evaluating and predicting the toxicity of a drug candidate before in vivo use!
A solution chosen by 4P-Pharma, a start-up specializing in the development of therapies that meet the unmet needs of patients. In early March, the biotech start-up announced a partnership with QuantHealth to initiate in silico simulation of its drug candidate 4P004 for osteoarthritis patients. With more than a trillion pieces of medical, clinical and pharmacological data, QuantHealth’s solution is capable of simulating thousands of variations of a clinical trial in just a few minutes. Its In-Silico platform evaluates each parameter one by one to highlight the appropriate protocol that will optimize the efficiency and safety of the future drug. In silico modeling will virtually test the 4P004 drug and at the same time identify the profile of patients likely to respond best to the therapy and increase the probability of success of the in vivo clinical trial.
These simulations therefore offer many advantages. They improve the chances of success of a new molecule, reduce development costs and accelerate its arrival on the market, thus giving patients faster access to innovative treatments. To date, virtual trials alone are not sufficient to validate a new drug. However, they are complementary and have been taken into account since 2017 by the European Medicines Agency (EMA) in the evaluation of the benefit-risk ratio of a candidate treatment.
Sylvie Ponlot