Q&A with Will Ganley, Senior Specialist, Nanopharm.
Can you Explain ‘in Silico Modelling’ and What a PBPK Model is?
In silico modelling involves using computer simulations to predict how drugs behave in the human body. An example is physiologically based pharmacokinetic (PBPK) modelling, which uses drug property, physiology, and biochemical data to predict absorption, distribution and excretion (ADME) processes using pharmacokinetic models.
The outcome is a system of equations employing a ‘bottom-up’ strategy that begins with modeling each specific process that moves the drug throughout the body. Starting from these foundational elements and progressing to a comprehensive model that integrates data from various sources, such as in vitro tests and clinical trials, we can replicate observed clinical outcomes. By developing a model that assigns significance to input data based on physiological relevance, we create solutions that accurately predict clinical outcomes, resulting in a trustworthy PBPK model.
How is the PBPK Model Applied Specifically to Inhaled Drug Products?
A typical whole-body PBPK model can be regarded as a large flow diagram, which maps mass transport processes across different types of tissue and blood vessels. If all the drug mass starts in the lung after a subject has inhaled, the model shows that the drug mass moves through the rest of the body and is exposed to the different tissues and blood compartments before it is ultimately metabolised and eliminated. For orally inhaled products, we are particularly interested instead in the details of what happens in the lungs.
The lung can be divided into two broad regions: the central region, which encompasses the upper airways where the blood flow goes from the arterial blood through to the venous blood, and the lower peripheral airways where the blood flow is reversed. We can further segment both regions into more granular compartments.
Within each compartment, the mucus layer is where the drug will typically start its journey as either a completely dissolved system, for example from a nebuliser, or as undissolved particles. For the latter, we directly simulate the particles’ dissolution in the mucus layer before they permeate down into the lung tissue.
In the mucous layer, dissolved drug will passively diffuse into the epithelium, then subepithelium, before entering the blood vessels and being carried away to the rest of the body. Knowing this, it is important that we consider the inputs to the model when developing and designing studies that allow us to generate the model’s input parameters.
There are two key types of input parameters for orally inhaled PBPK models. Deposition data that is obtained through deposition models or imaging data. Commonly used deposition models include computational fluid dynamics (CFD) or semi-empirical methods (such as the National Council on Radiation Protection and Measurements model). The advantage of CFD models is that computed tomography (CT) scans of healthy volunteers and patient airways can be used in the simulations to generate realistic and subject specific deposition predictions in the lungs.
