By Rakesh Gollen and Varsha Bhatt-Mehta
In our previous blog post, we highlighted the challenges related to drug disposition in the pediatric population. The pharma industry uses three different approaches to characterize PK/PD during the drug development process: namely, the pharmacometric or “top down” approach (PK/PD), the physiologically-based pharmacokinetic (PBPK) characterization of drug disposition or “bottom up” approach, and allometric scaling. Each has its own advantages and limitations.
All of these approaches use a combination of data available from both the preclinical and clinical settings (i.e., existing or newly generated data). The “top down” (PK/PD) approach evolves from the initial data-driven learning phase to the prospective hypothesis-confirming phase by using one or more pharmaco-statistical models. Using this approach, various changes in parameters (such as age, body weight, organ function, race, ethnicity, etc.) are included as covariates in the models to assess the influence of these non-heritable, as well as heritable, covariates on drug disposition. Other covariates, such as changes in intragastric pH, bile salt formation, gastric emptying time, intestinal motility, bowel length, and effective absorptive surface (which can affect the ionization, stability, and absorption and thus bioavailability) of many drugs can also lead to altered PK/PD and are mostly ignored using this approach. The other limitation is the need for large clinical data to have sufficient predictive power, which is not possible in the case of specific population such as pediatrics.
The “bottom-up” approach is based on mechanistic, physiological-based pharmacokinetic (PBPK) modeling, which relies on the previously established parameters including systems-based (anatomical and physiological parameters such as organ/tissue size, blood flow connecting different organs/tissues, protein binding concentrations, glomerular filtration rate, organ-specific drug metabolizing enzymes, and drug transporters), drug-based (physiochemical properties determining absorption through membranes, partition to various tissues, binding to blood/plasma protein, and the affinity towards various enzymes and transporters), and study-specific parameters (formulation, dose, route, and frequency of administration, along with food intake while incorporating various other changes in the gut physiology as a function of age). The development of models that include various enzyme and transporter data as a function of age and maturation are not easily available in various experimental settings, including children, resulting in severe limitation of pediatric PBPK models. Ironically, this is the population where such models could be most helpful.
Lastly, the allometric scaling method, which is more of a traditional approach to scale pediatric data, uses various allometric relationships to quantitatively describe the relationship between various physiological process (GFR, enzyme expression) as a function of body weight or body surface area (which is still a common practice in clinical drug development especially in oncology). This approach is complicated by the numerous assumptions in development of the allometric model that overpower its advantage of simplicity.
As pharmaceutical scientists, it has become evident that we need to evaluate the available data using a variety of approaches to find the right (safe and efficacious) dose for this specific “orphan” population, giving full consideration to the research question, properties of the drug molecule, targeted population age group, and available information. If you are using any of these approaches, please feel free to share your experiences or suggestions in the comment box below.