Data analysis, Modelling and accelerating development of Long-acting therapeutics

Posted on: 13 January 2023 by Andy Sloan in Developing the Long-acting Pipeline

Data analysis and modelling in pharmacology is an essential technique for understanding and predicting the effects of different drugs on the human body. It allows researchers to interpret large amounts of data from clinical trials, lab experiments, and other sources, in order to understand how drugs interact with different biological systems and how they might be used to treat various diseases or conditions. It is an important part of the drug development process, and helps researchers optimise drug therapies and improve patient outcomes.

Pharmacological data and Modelling approaches

Pharmacological data can generally be classified into two types:

Pharmacokinetic (PK) data is concerned with the timecourse of the amount (or concentration) of drug in the body following a dose, and pharmacokinetic models applied to PK data aim to describe and predict how a drug is absorbed, distributed, metabolized, and eliminated in and from the body. Pharmacokinetic models are often used to optimize dosing regimens, predict drug-drug interactions, and understand the PK profile of a drug in different populations.

Pharmacodynamic (PD) data is concerned with the timecourse of the biological response of the body to a drug, usually regarding its intended efficacy, and is intimately connected to the pharmacokinetics of a drug as the amount of drug in the body will drive the size of its effects. Pharmacodynamic models applied to PD data involve a much wider variety of data than pharmacokinetic analysis, describing and include predicting how a drug can affect various receptors, enzymes, or other biomolecules, as well as how it influences physiological processes such as blood pressure, heart rate, and blood glucose levels or even final clinical outcomes.

PK and PD data are often analysed jointly in “PKPD” analysis to allow the data on drug quantity to inform the understanding of observed drug effect.

There is a range of mathematical models that can be applied usefully to both PK and PD data, ranging from simple statistical and empirical models (e.g. compartmental PK models) to more complex models aiming for greater mechanistic detail in the description of the biological processes involved (e.g. physiologically-based pharmacokinetic modelling (PBPK)). Choice of model will depend on the requirements of the analysis and the degree of detail and information content in the data being analysed, with more complex models not necessarily always being preferable or required.

Challenges for long-acting drug development

One of the key challenges of data modelling in pharmacology is the need to account for variability and uncertainty in the data. This can be due to factors such as differences in patient characteristics, differences in drug formulation, delivery mode or differences in measurement techniques. Researchers must therefore use statistical techniques to account for these sources of variability and ensure that their models are robust and reliable.

Dr Henry Pertinez is a mathematical modeller at CELT who conducts quantitative data analysis using various techniques such as compartmental and non-compartmental pharmacokinetic analysis, physiologically-based pharmacokinetic modelling (PBPK), deconvolution and population mixed effects analysis to provide insight into the meaning and interpretation of data, and for simulation and extrapolation of drug use scenarios.

Henry said “Focussing specifically on long-acting therapeutics, one of the key aspects of these drugs’ (or specific formulations’) PK behaviour is the nature of their slow release from the administration site following dosing - in many cases this in fact becomes the dominant factor in affecting the PK profile of the drug or formulation in question. PK modelling methods are an important tool in seeking to describe, quantify, understand and predict the nature of this long-term release.”

Watch Henry describe Data Modelling in this short video

Our Impact

The various modelling techniques utilised by data modellers in the CELT Pharmacology Group have been applied to long-acting drug formulations intended to treat various infectious diseases such as HIV, Malaria, Tuberculosis, Hepatitis C and SARS-CoV-2 and have helped accelerate the development of these drugs for real world application in healthcare programmes.

A PBPK modelling study was published in 2021 about dose prediction of nitazoxanide repurposing for treatment of SARS-CoV-2 ¹. Using the published data, an open label phase 1 trial was later conducted in healthy volunteers and found that the pharmacokinetic data from the earlier published predictions using the PBPK model were comparable to the observed data ². This demonstrates the advantages and usefulness of the modelling techniques to inform dose prediction.

Teoreler

Researchers at CELT have developed Teoreler – a web-based physiologically based pharmacokinetic (PBPK) model platform for the simulation and extrapolation of drug use scenarios. This application uses complex mathematical equations to simulate and predict the efficacy of drugs in various subpopulations of humans and animals.

Dr Rajith Rajoli, who is the lead developer for Teoreler, said “Teoreler is an online PBPK platform for researchers to simulate and predict pharmacokinetics of their drug of interest without distressing about the mathematical complexity of the models. Teoreler is completely web-based, so the user just needs to have an internet connection to execute simulations and eliminates the need to install any software, nor does it require a high spec computer which is typically needed with desktop based applications. Simulations can be performed on mobile devices such as smartphones and tablets as the computational load is completely taken care by the server. The current objective of Teoreler is to provide users with a hands-on experience with PBPK modelling and also assist them in understanding the concepts of pharmacokinetics and its role in drug development.”

Teoreler was launched in the third quarter of 2022, in BETA (test mode) until the end of the year during which a feedback campaign was held to gather data from the mathematical modelling community for quality assurance and to ensure that the system is beneficial for a wide range of researchers. Currently, Teoreler features a human adult model and two animal models – rat and mouse with a few more in pipeline for release in the coming months.

Every effort has been made to simplify the complex concepts of PBPK modelling and allow a wide audience to understand and use this advanced subject in their day-to-day research.

Find out about the CELT Pharmacology Group

Dr Henry Pertinez and Dr Rajith Rajoli are members of the CELT Pharmacology Group. Led by Professor Andrew Owen, CELT Pharmacology bring to bear cutting-edge preclinical expertise to a diverse portfolio of projects.

The team work closely with local, national, and international partners to better understand mechanisms that underpin clinical phenotypes and to translate new interventions through the preclinical and clinical challenges. The teams’ expertise spans in vitro, in vivo, bioanalytical and pharmacokinetic modelling capability that is harmonised to achieve maximum value for their projects.

1. Rajoli, R. K. R., et al. (2021). "Dose prediction for repurposing nitazoxanide in SARS-CoV-2 treatment or chemoprophylaxis." British Journal of Clinical Pharmacology 87(4): 2078-2088.
2. Walker, L. E., et al. (2022). "An Open Label, Adaptive, Phase 1 Trial of High-Dose Oral Nitazoxanide in Healthy Volunteers: An Antiviral Candidate for SARS-CoV-2." Clinical Pharmacology & Therapeutics 111(3): 585-594.