The University of Washington Medical Center is teaming up with IBM’s Watson super computer to provide faster treatments based on cancer patients’ genetic profile.

Share story

Cancer patients at the University of Washington Medical Center could soon see faster DNA-based analysis of personalized, precision treatments thanks to a new collaboration with IBM’s Watson super computer.

UWMC is among more than a dozen cancer centers nationwide to join a new effort that could speed up genetic profile results from weeks to minutes, organizers announced Tuesday.

“What we’re hoping with Watson is that it might provide additional insight, but more important will be a tool in the future to make this simpler, faster and more widely available,” said Dr. Jonathan Tait, a professor of laboratory medicine and director of the genetics and solid tumor diagnostic testing lab at the UWMC.

The UW and a dozen other participating institutions will use Watson to assist doctors in identifying cancer-causing mutations and potential treatments. It typically takes weeks to gather and test tumor specimens and then sequence the genes, Tait said.

“Then there’s a gigantic mountain of raw data,” he added. “But once Watson gets it, it can be fast.”

After Watson comes up with genetic profiles and potential treatments, the results must be reviewed by a doctor for accuracy and feasibility, Tait said.

About 1.6 million Americans are diagnosed with cancer each year. Most receive surgery, chemotherapy or radiation to treat the disease. Increasingly, however, genetic sequencing is becoming an option for patients whose cancers don’t respond to standard treatment.

The UW currently uses its own Onco-Plex system that detects mutations in 234 cancer-related genes. The test, which costs $3,000 to $4,000, is typically used for patients with advanced cancer or those with unusual and hard-to-treat types of disease. About 40 patients a month are treated with Onco-Plex now, Tait said.

The new Watson collaboration will enroll an additional 20 patients per month to start, Tait said. That will include patients with brain, breast, prostate and lung cancer. In the future, patients with melanoma skin cancer and liver cancers will be added.

About 10 to 20 percent of patients who receive genetic analysis of their tumors benefit immediately, with different drugs or therapies to target their disease. For the rest of the patients, the analysis offers information that’s helpful for future decision-making, Tait said.

“When you look at individual cases, we’ve had some truly remarkable successes,” he said. “We’ve had people who thought they were at the end of their options and we tested them and found something that wasn’t expected that allowed them to be eligible for a drug. When the drug was used, they responded.”

IBM reached out to leading cancer centers as part of its broader Watson Health initiative focusing on precision or personalized medicine. Watson is an artificially intelligent computer system that uses a human-style cognitive framework to inform analysis and decisions.

“This collaboration is about giving clinicians the ability to do for a broader population what is currently only available to a small number — deliver personalized, precision cancer treatments,” Steve Harvey, vice president of IBM Watson Health, said in a statement.