By Arnab Mukherjee
Consider the cost to the pharmaceutical industry to develop and get approval for a new drug. The Tufts Center for the Study of Drug Development estimates the cost to be $2.6 billion, mostly attributable to higher clinical trial costs and failure of late-stage clinical trials. While there is no silver bullet, informed rational decisions, made routinely for every drug candidate and at every stage of drug development, offer the potential to reduce the overall rate of attrition at later stages of clinical programs. Utilization and integration of relevant knowledge, e.g, study data, literature information, and disease information, among others, in a quantitative framework to inform decisions is core to the role of modeling in drug development.
Whether the decision is related to selection of the biological target, progression through each stage of drug development, regulatory approval, or reimbursement by payers, the appropriate modeling methodology and tools enable a risk-based assessment to inform the decision. For example, systems pharmacology models mathematically integrate available knowledge of a physiological system to predict clinical outcomes related to modulation of molecular targets within the system. At the other end of the spectrum, meta-analytic models that integrate clinical data in an indication across competing therapies enable comparison of safety and efficacy for approval and payer decisions.
A key objective of modeling is to identify factors that explain inter-individual variability in drug exposure and response. Such factors may include demographics, concomitant medications, disease severity, genetic markers, mechanism markers, and other relevant covariates that may allow further optimization of therapy. Therefore, modeling is a natural tool of precision medicine, an emerging approach for disease treatment and prevention that takes into account individual variability to target therapy.
Other common applications of modeling include optimization of trial design and doses, translation of efficacy and safety from preclinical models to humans, and bridging efficacy from one population to a new population based on predictive covariates. In most pediatric development programs, for example, model-based bridging and extrapolation from adults is a necessary component for successful implementation, given ethical and operational considerations that make it difficult to perform extensive clinical trials in children. While the role of modeling in drug development is highlighted here, there is a substantial opportunity and need for the application of modeling in clinical practice and healthcare utilization, where optimization of therapy and outcome for the individual patient is the ultimate goal. Modeling methods and tools used in drug development are equally applicable for integration of available drug-specific information with patient-specific information, in the framework of effective clinical decision support tools (e.g, dashboards), to achieve this goal.
The 2015 AAPS Annual Meeting and Exposition program highlights the role of modeling in drug development, with examples of utilization in decision making, preclinical to clinical translation, and bridging strategies. There is a special focus on systems pharmacology application and methodologies. The meeting promises to be of high educational value to those that want to learn about modeling and understand its role in drug development.