Advances made in using computers to predict the side effects of developmental drugs
Posted: 11th February 2011
Over 40% of proposed drug compounds fail to make it to market due to side effects on the heart. Currently, drug cardiac safety is assessed according to set of laboratory tests, with a tightly limited range of acceptable answers. There is broad agreement within the drug development community that these tests are not as sensitive or accurate as they should be, but credible alternatives have not been found. Computer Scientists from Oxford University are helping tackle this problem.
Now the preDiCT project has developed the computer models and tools to simulate drug effects on the heart, which offers the possibility of a better way to predict drug safety. It is now possible to simulate drug action in the human heart from sub-cellular to whole-organ scale.
On 7-8 February, drug regulators and pharmaceutical industry researchers met researchers from Oxford and Valencia Polytechnic Universities, to discuss the next steps to integrate computerised modelling and simulation in drug safety assessment.
Members of the preDiCT team and Scientific Advisory Board, the European Medicines Agency, Health Canada and the US Food and Drug Administration, along with representatives from AstraZeneca, GSK, Notocord, Fujitsu Laboratories of Europe, and clinicians participated in the discussions. It was agreed that modelling and simulation technology could help in:
- Resurrecting “good” drugs which had previously failed
- Investigating drugs’ likely safety profiles during early-stage development
- Simulation results adding weight/explanatory power to new drug applications
- Drug safety assessment by regulatory bodies
- Identifying drugs which are “risky” or “safe” for specific groups of patients
As a result of this meeting, the project will now energetically pursue further discussions with regulators and pharma representatives, to agree criteria for the acceptance of models in the drug development process.
The preDiCT project links researchers in Oxford University’s Computing Laboratory and the Department of Physiology, Anatomy and Genetics, as part of a consortium of eight European businesses and universities. The project is funded by DG-INFSO, as part of the European Commission’s FP7 “Virtual Physiological Human” initiative. www.vph-predict.eu / www.vph-noe.eu