In silico human-based methodologies for evaluation of drug cardiac safety and efficacy

Lead Research Organisation: University of Oxford
Department Name: Computer Science

Abstract

Cardiotoxicity is one of the leading causes of failure during drug development and also, more worryingly, after marketing approval. Withdrawal due to cardiotoxicity has increased from 5.1 to 33%, including compounds to treat cardiovascular problems as well as drugs not intended to affect the heart. Cardiotoxicity manifest itself as adverse effects on cardiac electrophysiology such as arrhythmias and sudden death, and alterations on contractility and structure including, for example, increased incidence of heart failure and myocardial infarction. In spite of the variety of animal methods for preclinical screening for drug safety, 20-50% of all advanced candidates have to be abandoned due to adverse outcomes, even late in the drug development process. These figures could be drastically reduced, and at a smaller cost of animal experimentation, by the adoption of in silico technologies in the earlier phases of the drug development process.

The main aim is to accelerate the uptake of human-based in silico methodologies for evaluation of cardiac drug safety and efficacy in industry, regulatory and clinical settings.
The specific objectives include:

1) Review, collation and implementation of a comprehensive database of human electrophysiology and contractility in silico models for specific disease conditions such as heart failure, myocardial ischemia, genetic disorders and cardiomyopathies.
2) Development and qualification of in silico human models for the prediction of adverse outcomes in human cardiac electrophysiology and contractility for specific disease conditions, based on existing models, and calibration with in vivo and ex situ recordings.
3) Evaluation studies to compare in silico human-based predictions to clinical outcomes, current animal methods, and in vitro methods including stem cell derived cardiomyocytes.
4) Planning and development of workshops and dissemination activities to identify and overcome barriers for the uptake of in silico methods in industrial, clinical and regulatory settings.

The proposal will consider as an important focus the fact that cardiotoxic adverse events are more likely to occur in patients with compromised cardiac electromechanical function due to coronary artery disease, myocardial infarction and/or heart failure. These are rarely reproduced in animals, which are however systematically used for drug evaluation. We will create the infrastructure required to facilitate access to expertise and in silico multiscale human cardiac models from ion channels to the whole organ, for a range of conditions including heart failure, myocardial ischaemia, and genetic mutations. This amazing technology has the obvious advantage of a focus in human, but also the ability for personalisation to specific patients' conditions, with a potential 3Rs impact in drug safety and efficacy evaluation in industry, and also in clinical applications that also rely on animal testing. In addition, to delivering in silico modelling technology, the project will facilitate inter-sectorial working groups including academics, clinicians and industrial partners for the definition of evaluation criteria to increase the credibility and refine the in silico human models for cardiotoxicity, and to design strategies to overcome barriers for their uptake to refine, replace and reduce animal experimentation for drug safety and efficacy evaluation. The project is strongly supported by wide membership across regulatory, industrial, clinical and academic sectors located in at least ten countries, who will contribute to raising the profile of the in silico models for the 3Rs in animal experimentation, internationally.

Technical Summary

Cardiotoxicity is one of the leading causes of failure during drug development and also, more worrying, after marketing approval. Withdrawal due to cardiotoxicity has increased from 5.1 to 33%, including compounds to treat cardiovascular problems as well as drugs not intended to affect the heart such as antihistamines. Current used strategy to screen for adverse contractility effects involves a combination of preclinical in-vitro pharmacological profiling, cardiomyocyte assays and in-vivo cardiovascular (CVS) studies, and uses a variety of animal species (rats, mice, rabbits, guinea-pigs, dogs, pigs and non-human primates).
In spite of the variety of animal methods for preclinical screening for drug safety, 20-50% of all advanced candidates have to be abandoned due to adverse outcomes, even late in the drug development process.
The main aim is to accelerate the uptake of human-based in silico methodologies for evaluation of cardiac drug safety and efficacy in industry, regulatory and clinical settings. The specific objectives include:

1) Review, collation and implementation of a comprehensive database of human electrophysiology and contractility in silico multiscale mechanistic models for specific cardiac disease conditions.
2) Development and qualification of in silico human models for the prediction of adverse outcomes in human cardiac electrophysiology and contractility for specific disease conditions, based on existing models, and calibration with in vivo and ex situ recordings.
3) Evaluation studies to compare in silico human-based predictions to clinical outcomes, current animal methods, and in vitro methods including stem cell derived cardiomyocytes.
4) Workshops and dissemination activities to identify and overcome barriers for the uptake of in silico methods in industrial, clinical and regulatory settings.
Project membership involves key partners across 11 countries who will raise the profile of in silico human models for the 3Rs.

Publications

10 25 50