• Siladitya Ray Chaudhuri Ph.D.

Proposal Description:

Inter-person variability can often lead to a failed drug candidate/product owing to lack of efficacy or adverse effects in individuals or specialized subpopulations of patients during clinical administration. Historically, this was assessed in a more ad hoc or subjective manner and was based solely on the experience and instinct of the clinician. However, the modern practice encompasses "personalized medicine" as an essential strategy to drive the most effective clinical decisions for optimal patient and population health. This mode of individualized treatment aims at delivering the right drug at the right dose by attempting to capture any patient variability upfront. While "right drug" emphasizes the need for therapies specifically designed for subpopulations of patients, "right dose" highlights the need to maintain an optimal drug exposure (local or systemic) in line with a subject’s phenotype. This phenotype can be a collection of the "-omics" data, notably on metabolomics, lipidomics, proteomics, transcriptomics, and genomics. Although individualization down to the "-omics" level may prove challenging and resource intensive, clusters or groups of patients may be formed by common elements characterized through similar pharmacokinetic exposure and pharmacodynamics response. In doing so, physiologically based pharmacokinetic (PBPK) modeling and simulation can serve as the only meaningful in silico tool that can capture the biological variability and translate it into exposure and efficacious response.

PBPK models are staples for predicting human exposures in advance of clinical trials, formulation development for better drug absorption, food effect, and an a prior knowledge towards de-risking strategy for drug-drug interactions. However, the tool can be extended to provide much deeper insight into the understanding of drug disposition in and efficacious behavior of a given patient. In the context of personalized medicine, the aim of PBPK would be to develop the relationship between dose, physiology, and local or systemic exposure for a given subject or subpopulation. This session will focus on these caveats of incorporating big data on patient variability into a mathematical model and a PBPK simulation tool to characterize the exposure across an entire group or sub-population as opposed to a mean individual who does not necessarily represent the collective. Overall such virtual trials or population simulations hold the key to understanding the range of applicability of the safety and efficacy profile for a given drug product.

The session will also focus on the collaborative and integrative nature of the exercise. Using PBPK tools for personalized medicine involves the integration of several disciplines, including biology, pharmacokinetics, biomarkers, clinical pharmacology, and biostatistics, to name a few. This integration of different functional areas and disciplines is central to achieving the goal of delivering a robust drug product that is both safe and efficacious for everyone.