Using biochips to analyse the genetic makeup of these tumours – which comprise around 22,000 genes – implanted in laboratory mice, they managed to differentiate malignant cells from cancerous ones using mathematical and statistical tools.

By observing the mice, they were able to see which cancerous cells would not spread to other parts of the body, even though they would still need to be removed.

An analysis of 32 genes in breast cancer tumours and 31 in lung cancer tumours out of the 22,000 or so present in each allowed the scientists to work out how the cancer cells would develop over time.

Project leader Ramón García Escudero, of the Energy, Environment and Technology Investigation Centre (CIEMAT) said that a simple diagnostic method for use in hospitals would need to be developed before the results could benefit patients, a process which would take at least two years.

“There are many patients who would not need chemotherapy because they have cancerous cells which are not the spreading type,” explains García Escudero.

“At present, many patients who undergo chemotherapy merely do so because there have not been any clear methods to allow doctors to predict how different types of cancer cells would evolve.”

In addition to the Molecular Oncology Unit at the CIEMAT, experts from the Catalán Oncology Institute and the Madrid-based hospitals Gregorio Marañón and 12 de Octubre took part in the research.

The results have been published in the latest edition of the USA-based science magazine, PloS One.