What is PBPK Modeling & Simulation?
Physiological based pharmacokinetic modeling and simulation (PBPK) is a computer modeling approach that incorporates blood flow and tissue composition of organs to define the pharmacokinetics (PK) of drugs. The concept of PBPK was first described by Teorell in 1937. Simply put, PBPK is a tool to assess factors responsible for patient variability that impacts the PK of drugs.
Alterations in PK properties, such as, absorption, distribution, metabolism, and excretion (ADME), can have a substantial impact on achieving the desired therapeutic concentration of a drug. Too low of a concentration results in ineffective therapy, and too high may result in side effects or even toxicity. PBPK is a powerful tool that requires modern computational power to handle the intense and complicated mathematical equations necessary to make quantitative PK estimations and predictions.
PBPK provides a mechanistic approach to study and predict the PK of drugs based on physiologic and anatomic characteristics, as well as the physical and chemical properties of a given drug. The goal of any drug or therapeutic intervention is to be effective without causing harm. Depending on the drug and the individual taking it, this can be a difficult task; nonetheless, this is an essential and actionable area for translational science. The utilization of intelligent PBPK models and simulations presents countless opportunities for improvements in drug development.
Methodology & Components of a PBPK Model
Non-compartmental analysis (NCA) is a standard PK modeling tool. NCA is an empirical model that lacks physiological meaning. NCA assumes that the dose administered is distributed uniformly throughout the body and that the elimination of the drug is defined through a rate constant that does not account for physiology. This modeling approach allows for rapid generation of PK parameter estimates, but does not account for any physiological mechanisms or biological processes that drive or alter PK.
Population pharmacokinetic analysis (popPK) is another standard tool to describe the PK of a drug. Typically population PK models are empiric as well, but can also be semi-mechanistic.
Alternatively, a PBPK model is based on physiology (Figure 1), biological processes, organ function, enzyme/transporter abundance and function, and blood flow, etc. PBPK modeling and simulation incorporates physiologically driven parameter sets that are responsible for PK variability among patients.
The PBPK model allows us to include:
- Physio-chemical properties of the drug
- Specific physiological differences
- from patient groups and demographics
- Trial design information
- dosing route and frequency
A population can be defined as nearly any group of people or clinical scenario such as:
- Healthy patients
- Patients with a tumor or disease that affects or alters organ function
- Life event/stage (childhood, pregnancy, or post-surgery patients)
The patient’s characteristics (age, sex, weight, body composition, organ function, genetics, etc.), can also be utilized and integrated in a PBPK model and simulation. The drug properties used in PBPK models include: molecular weight, logP, pKa, protein binding data, blood/plasma ratio, metabolism, permeability/solubility, transport mechanism, lipophilicity, etc. The model then can predict the PK of a drug before a study is conducted and the study can then be used to verify the prediction.
Each cycle of prediction and verification – from animal → healthy subjects → to patient → to special population – is accomplished by changing the physiological parameters within the model. Regulatory agencies have begun to accept PBPK modeling in place of many drug-drug interaction (DDI), pediatric, special population studies and more. PBPK modeling can also be referred to as bottom-up or mechanistic modeling and simulation.
Benefits and Applications of Using PBPK Models
The benefit of using PBPK modeling is that it is a cost-effective and robust predictive tool that is devoid of the ethical challenges associated with clinical trials in sensitive populations (e.g. cancer patients, pediatrics, pregnant women, etc.). The physiological effects that alter PK are vast and, in some cases, they are compounding. There is a critical need for PBPK investigation, especially for drugs with a narrow therapeutic window and sensitive patient populations. For example, applications can include pregnancy populations, organ transplant populations, and bariatric surgery.
Many PBPK models have been utilized to study pregnancy. There are numerous physiological changes that occur throughout gestation. Changes in fat and water volumes can directly impact hydrophilic or lipophilic compounds and there are changes in the expression of cytochrome P450s, specifically 1A2/2D6 and 3A4 during pregnancy.
Physiological changes in pregnancy include altered expression of Cytochromes (CYPs). For example, the expression of CYP3A4 increases in the first trimester and then decreases during the second and third trimesters but remains higher than the normal non-pregnant enzyme expression. Additionally, CYP2D6 enzyme expression increases during pregnancy. These changes in the CYP enzyme expression can alter the metabolism of several commonly prescribed drugs throughout gestation, thus illustrating the importance of accounting for these differences when prescribing medications during pregnancy.
The effect of these changes and other parameters on the PK of drugs in pregnancy can be evaluated utilizing PBPK studies. Some additional physiologic changes altering PK include:
- Variations in cardiovascular physiology
- Increased blood volume/fat/plasma/amniotic fluid and placental volume
- Decreased plasma protein/serum creatinine, etc.
Organ Transplant Populations
Another powerful application of PBPK has been in organ transplant populations. Organ transplants are extremely complicated procedures with an intense recovery period. Of particular interest for PBPK modelers is liver transplant patients and heart transplant patients. The altered architecture of the liver can impact the PK of drugs, such as:
- Decreased organ volume
- Transporter expression
- Drug metabolizing enzyme expression
- Albumin production
The changes in cardiac output can also have profound alterations on the blood flow throughout the body and in turn effect all organs and the PK properties of drugs. Both populations (liver transplant and heart transplant) can be studied, modeled and simulated to evaluate necessary alterations in drug choice, as well as dosing level and frequency prior to dosing the patient.
Bariatric surgery is another very compelling application. Obese patients are known to have higher fat mass, altered cardiac output, increased liver and plasma volume and increased activity of CYP2E1 and α-1 acid glycoprotein. The increased incidence of obesity worldwide has led to an increase in bariatric surgery procedures. There are a variety of procedures, from jejunoileal bypass, biliopancreatic diversion with duodenal switch and gastric bypass, all varying in invasiveness. The consequences to any of these procedures will be:
- Reduced gastric capacity and emptying time
- Altered GI pH
- Reduced absorption area
- Altered bile flow and small intestinal transit
- Altered generation of and exposure to metabolizing enzymes and efflux transporters
Patients undergoing bariatric surgery are on a variety of post-surgical drugs and there are most certainly physiological effects that impact the PK, which can be modeled and simulated for safer and more optimal drug usage.
PBPK models utilize assumptions about the rates of each individual process and sometimes these rates may be unknown. In these cases, sensitivity analyses can be undertaken to understand the consequences of uncertainty. Another limitation is that PBPK models tend to describe the average person with a disease of interest, but does not describe inter-individual variability and unexplained variability (in contrast to population PK modeling). This limitation can be overcome by sensitivity analyses with high and low values for important characteristics.
The general idea behind PBPK is that fluctuation or alterations in one’s physiology that can be described mathematically, can alter the PK properties (ADME) of a compound and in turn have an impact on the desired therapeutic outcome. PBPK modeling and simulation allows us to investigate the effect of these changes in silico and prior to dosing patient populations. PBPK is a powerful tool and can be iteratively modified and updated as new knowledge of the compound and physiology are discovered. Through “predict-learn-confirm” cycles, one can not only use these models to describe the clinical observations, but more importantly to predict, via simulations, the untested clinical outcomes.
PBPK is a strategic tool being used by drug developers to predict and evaluate the effect of previously unrecognized physiological parameters that directly impact the PK of drugs. These parameters include various intrinsic factors (pregnancy, genetics, disease, organ dysfunction, race) and extrinsic factors (diet, smoking, alcohol use, other medications). The power of PBPK allows us to examine the impact of these parameters that are critical in drug development.
Ultimately, PBPK can provide researchers and physicians with a clearer understanding of how dose selection should be altered throughout diseases, life stages and other physiological altering events that impact the PK of compounds. Contact us to learn more about Nuventra’s PBPK and modeling and simulation services.