PhinC Development is today a key player in Europe when it comes to the use of modeling and simulation approaches during the drug development process. Thus, the uniqueness and strengths of the PhinC model are based on the following 3 axes:
- Deep Learning, which consists in developing the best models to exploit all the research data (in vitro, in vivo animal and human) and to carry out robust simulations and predictions. Then validate these models to obtain acceptance from drug agencies (EMA, FDA, etc.).
- Agility and Biotech orientation: PhinC adapts its work organization and directs its collaborations to meet the needs and specificities of Biotechs (funding method, attrition rate, interaction with many other research providers, advice upstream on the development strategy, justification of the benefit of modeling, etc.).
- Drug development expertise that allows PhinC to keep a pragmatic approach to its modeling work to serve the development goals of the drug candidate.
In this article, we will highlight the first axis and explain how PhinC uses the best innovative tools to deliver quality service to its customers as well as reliable and robust results while optimizing time and costs.
Preclinical, clinical phase, obtaining marketing authorization... 10 to 12 years of research are necessary for a molecule of therapeutic interest to enter the drug market (excluding medical devices). The regulatory milestones that accompany the development of the drug pose an even greater risk of attrition for drug candidates. Likewise, the costs for the development of a molecule amount to an average of 300 million euros and up to 1.5 billion euros including the probability of success.
Faced with these major obstacles and challenges, PhinC brings its expertise in pharmacometry, pharmacokinetics, pharmacology and biostatistics to develop innovative predictive models. Such predictive models are now extremely powerful tools, able to give Biotechs a decisive advantage in the race to develop drug candidates. Indeed, they make it possible to respond to a multitude of "challenges":
• the increasing complexity of test protocols: our models make it possible to evaluate several parameters simultaneously, to isolate particular effects, and to quantify their pharmacological effects alone or in interaction;
• the multiplication of data sources: our models allow the consideration of literature data, in vitro, in vivo on the species studied. In addition, these models will be fed and enriched progressively with the data that will be obtained during development;
• risk management: of cardiac, renal or hepatic toxicity, for example by modeling drug exposure versus toxicity biomarkers (troponin or QT wave for cardiotoxicity, for example), by helping to choose the first dose to be administered to humans thanks to inter-species predictive models;
• ethical requirements and cost reduction: our predictive models make it possible to predict the systemic or target exposure levels of the candidate under study, with a sufficiently well-defined margin of error to reduce the scope of the different doses of the drug to be tested , and thus optimize the number of individuals (animal, human) or samples to be included in the experimental designs;
• the need for "derisking": our models make it possible to anticipate the therapeutic margin (between toxicity and the desired pharmacological activity) and thus not to start a development with too narrow therapeutic margin. Likewise with the risks of drug interaction (with statins for the elderly, for example), or modification with food intake, etc.
The evolution of modeling techniques is constant as well as the framework for acceptance of their use by health authorities. PhinC's mission is therefore to develop proprietary modeling approaches that will be deployed in specific research areas (such as for the development of monoclonal antibodies) and for targeted therapeutic indications (oncology, autoimmune disease). These proprietary methods and our associated know-how will allow us to facilitate the development of predictive pharmacokinetic models specific to the drug candidate developed by a Biotech. Indeed, the use of these methods accelerates the development process and reduces technical and regulatory uncertainties (case law of other submissions using this methodology).