TORONTO - Canadian researchers have developed a technology that analyzes breast cancer tumours in a new way, allowing them to predict with more than 80 per cent accuracy a patient's chance of recovering.
The goal of the computerized tool is to eventually help doctors better target treatment to an individual patient, based on their tumour's profile.
Called DyNeMo -- for Dynamic Network Modularity -- the system analyzes how proteins and other components within cancer cells interact with each other to form networks.
The way the networks in cells occur can indicate how the tumour is likely to behave.
"What you find is that those proteins actually form a network, so they're not just individual hubs doing their own thing," said co-inventor Marc Wrana, a senior investigator at the Samuel Lunenfeld Research Institute at Mount Sinai Hospital in Toronto.
"They're all interlinked, in the same way humans are interlinked in a social network."
In a study of 350 women with breast cancer, published online Sunday in the journal Nature Biotechnology, the researchers found that patients who survive breast cancer have a different organization of their protein network within tumour cells than those who succumb to the disease.
"So what we were able to do was by looking at how different proteins are expressed ... we're able to identify changes in the global structure of that network that could predict outcomes from disease," Wrana said.
"Our hope with this technology is to eventually provide individualized analysis to breast cancer patients and their oncologists so that they are better informed and empowered to select a treatment best suited to them."
That could mean using certain drugs instead of others, he explained.
Co-inventor Ian Taylor, a PhD candidate in molecular genetics in Wrana's lab, said DyNeMo is intended to add another piece of information to other prognostic technologies used by doctors to determine the size, stage, grade and other traits of a woman's breast tumour.
"This technology doesn't replace those, but it does complement them ... so we have more accurate predictions," he said, noting that as worldwide studies on protein networks accrue, "we expect our accuracy of prediction will increase."
The researchers are looking for partners in the biotech or pharmaceutical industries to commercialize the product, said Wrana, who hopes it will be in widespread use for breast cancer patients within five years.
They also plan to apply the technology to other types of cancer to see if it could predict a patient's response to particular drugs before treatment is chosen. Cellular networks may play a critical role in other diseases, as well.
"Our idea is that these global network changes might not be just important in cancer, but other kinds of complex diseases, such as inflammatory bowel disease," Wrana said.
Marc Vidal, an associate professor of genetics at Harvard Medical School, called the Toronto research an important step in the evolution of personalized medicine.
"They made a really important discovery in cancer research," he said from Boston.
Vidal said the rapidly evolving field of network biology suggests that it is not a person's individual DNA that matters so much, but how that DNA makes products like proteins and how those products interact with each other "in amazingly complex ways."
"That's what creates a cancer cell, that's what creates a cancer cell that will respond -- or not -- to a particular compound (drug)."
Vidal said the exciting aspect of the research is the use of network biology to analyze and better understand cancer cells.
While that won't lead to fully personalized treatment right away, "it's definitely a great foot in the door in that direction."
