Could a computer program treat triple negative breast cancer?

Monash University

Wednesday, 20 June, 2018


Could a computer program treat triple negative breast cancer?

A computer program developed by Monash University researchers has revealed a previously unknown combination of drugs that may be the answer to treating triple negative breast cancer, using genetic and treatment data from cancer cells grown in labs and from hundreds of patients worldwide. The work has been published in the journal PLOS Computational Biology.

Triple negative breast cancer is aggressive and deadly, with no targeted drug therapy. Patients are currently treated by chemotherapy but there is no guarantee of success — and in cases where chemotherapy does not work, the survival rate remains only 12 months.

Doctors are currently turning to combination therapies — cocktails of drugs — in an effort to kill the cancer. However, there is no reliable way to predict which combinations, among hundreds, will work (and work quickly) for an individual patient with triple negative breast cancer.

Monash researchers have used genetic and treatment data from triple negative cancer cells grown in the lab and from hundreds of patients worldwide to develop a computer program, which has revealed a previously unknown combination of drugs that may be the answer to the disease. Dr Lan Nguyen and his team at the Monash Biomedicine Discovery Institute believe the computer model will eventually become an app that clinicians can use to match the best combinations of drugs for individual patients who present with the disease.

Triple negative breast cancer cells can develop resistance to a single targeted drug within days, sometimes hours — largely by rerouting the signalling pathways within the cells. “It’s similar to when there’s a car accident, and the traffic manages to reroute itself around it without causing gridlock,” Dr Nguyen said. “But how exactly these cancer cells find new routes to avoid the drug effect remains largely unknown.”

The Monash team, in collaboration with Israel’s Weizmann Institute of Science, characterised a key signalling network that drives the growth of triple negative breast cancers and developed a computer model that can predict how the network reroutes in response to a particular drug agent. This new model and its predictions allowed them to rank various combinations of drugs as to which are the most likely to defeat the cancer, by blocking the new route undertaken by the cancer cells.

Using data from The Cancer Genome Atlas, a database of cancer genes and patient histories run by the US National Institutes of Health, the researchers tested their league table of drug combinations to determine their success in people who had survived triple negative breast cancer. By inputting a patient’s genomic and proteomic information into the computer model, they can tell who may benefit from this drug combination and who may not — so that precious time is not lost in treating a patient with the wrong drugs.

The researchers found a previously unknown combination of two drugs that the model predicts could be successful in treating this previously untreatable disease, with Dr Nguyen saying they hope to have this new combination in clinical trials in two to five years. He added that the computer model can also be adapted and used to determine effective drug combinations for other serious cancers, such as lung and melanoma, where network rerouting in order to evade the drug effect has been observed.

Image credit: ©stock.adobe.com/au/caleb

Related News

Deep knowledge AI system could speed up drug research

Known as Pattern to Knowledge (P2K), the technology uses artificial intelligence to leverage deep...

New algorithm could improve rare disease diagnosis

Stanford scientists have developed an algorithm, called Phrank, that automates the most...

Deep-learning algorithm controls paralysed arms

By analysing the brain activity of tetraplegia patients, a new deep-learning algorithm can help...


  • All content Copyright © 2018 Westwick-Farrow Pty Ltd