The study will look at developmental cancer drugs
A project to improve the effectiveness of tests into cancer treatments could limit the need for animal testing, researchers have claimed.
Scientists at the University of Dundee, who will lead the £10m Europe-wide project, said it may also boost safety for those involved in clinical trials.
They want to isolate at an earlier stage those drugs most likely to produce cancerous effects themselves.
Doing so could cut the amount of time and money spent on ineffective drugs.
Researchers said that one of the key drivers behind the plan was the "3 R" issue - the reduction, refinement and replacement of experimental animal use.
The study, known as the MARCAR project and involving 12 organisations from business and academia, will explore the use of non-invasive imaging techniques such as MRI scanning to study the effects of developmental treatments.
As such techniques are non-invasive, long-term studies can be carried out on the same animal over weeks or months.
The method also has the potential to detect pre-cancerous lesions and tumours sooner without sacrificing the animals, meaning smaller numbers of subjects could be used.
Prof Roland Wolf, from the university's biomedical research institute, said: "This would potentially markedly reduce the numbers of animals needed for this kind of research and provide a much more reliable prediction of the rates of toxicity of drugs in development in man."
The researchers will focus on a group of drugs known as non-genotoxic carcinogens (NGCs).
These are drugs which when tested are shown to promote biochemical processes which lead to cancer.
At present such potential cancer-causing compounds only tend to be identified following prolonged biological trials.
Prof Wolf said the research would also improve safety for those taking part in drug trials.
He added: "The development of new drugs is a very costly process, partly because of the large number of drugs which never make it to market due to the discovery of cancerous effects during drug development.
"Predictions regarding safety of drug compounds can be imprecise and sometimes incorrect.
"If we could make better predictions at an early stage of drug development it would save a lot of time and money and make the whole process more efficient.
"To achieve that we need to identify early biological indicators, known as 'biomarkers', that can be used to predict the effects of drugs and reliably and robustly predict later cancer developments."