Population PK/PD modeling is an extremely valuable tool and has been extensively used in the pharmaceutical industry to facilitate drug development. I gained extensive hands-on experience in building various types of population PK/PD models when I worked as a Senior Clinical Pharmacokineticist in the Department of Clinical Pharmacology and Pharmacometrics at Abbott (currently known as AbbVie). Since I became an independent investigator in academia, developing and utilizing population PK/PD models for dose regimen optimization for new drug candidates, also known as model-informed drug development (MIDD), continues to be one of my major research interests. 

Besides new drug development in the pharmaceutical industry, pharmacometrics (i.e., PK/PD modeling) also plays an important role in clinical practice in hospital setting. This is particularly the case for special/vulnerable populations (e.g., infants, pregnant women, critically ill patients), where traditional PK study with frequent serial blood sampling is no longer feasible. As only a few limited blood samplings per subject may be collected in clinical studies, the sparse individual data makes the PK/PD analysis considerably more challenging and the ONLY way to handle this type of data is to perform population PK modeling. This strategy is known as opportunistic clinical study design combined with population-based pharmacometrics models, which has proven to be powerful when used in tandem as they can provide comprehensive information about drug kinetics and guide dose regimen optimization in vulnerable populations. My group is fortunate to work on many interesting projects in this direction, and our modeling & simulation results have played important roles in identifying rational dose regimens of various drugs in these special/vulnerable populations.  

Some drugs have complicated PK, and simple compartmental models won't be able to characterize their disposition; this brings the need of building mechanism-based PK models.

 Sometimes we get luxury tissue data on hand; this brings the opportunity of building physiologically-based PK (PBPK) model.  My group has extensive hands-on experience in building various mechanism-based models and PBPK models. 

QSP:  represents the youngest branch of the PK/PD discipline, and it has immense potential in drug discovery and development.  For QSP models, some of them are essentially an “enhanced” version of mechanism-based PK/PD modeling, which may contain multiscale structures to characterize mRNA/protein data, biomarker data, drug concentration-time profiles, clinical responses simultaneously; QSP models could also include complicated network model (commonly seen in systems biology) combined with drug kinetics component.   This is a promising and meanwhile challenging research direction. My group has just started research in this direction.