A new approach to mathematical modeling of prostate cancer has made it possible to spot potentially life-threatening tumors among the ones that are less threatening, allowing physicians to selectively offer aggressive treatment to those in need. The findings, published in the journal European Urology Focus, may help reduce the severe side effects of cancer treatment among men whose cancers do not yet need to be treated. "Curative treatment of early prostate cancer by surgery or radiotherapy needs to ideally be targeted to the minority of men with significant cancers, so that the remainder are spared the side effects of treatment, which frequently includes impotence,” Colin Cooper, a professor of cancer genetics at UEA's Norwich Medical School, said in a press release. Discriminating between aggressive and more benign tumors in prostate cancer has been challenging because of the range of variability seen when examining thes tumors under a microscope. Earlier attempts to use mathematical modeling to classify prostate cancer have failed. This time researchers at the University of East Anglia used a method called Latent Process Decomposition (LPD). This computational technique assesses the structure of a data set, without making use of clinical outcomes information. Using this approach, researchers studied gene activity in prostate cancer samples This allowed them to identify a pattern found only in tumors likely to have severe health consequences. The research team named these cancers DESNT.