IBM’s Watson Tackles Genetics-Driven Brain Cancer Treatments

Watson Goes Live at Cleveland Clinic

IBM’s Watson supercomputer is again expanding its reach into medicine, this time through a partnership between IBM and the New York Genome Center. The new project will analyze genetic mutations present in tumor genomes, with hopes of then discovering treatments that could target those specific mutations.

This new genome-focused approach to cancer treatment is not entirely science fiction. Researchers have been studying tumor genomes and the mutation within them for years. Throughout that research, the focus has been first on identifying mutations, and then on discovering whether those mutations are harmful. Pharmaceutical companies are piggy backing on that early work to begin creating new treatment options that can target and neutralize the harmful mutations.

At this point, there is a significant amount of data available on harmful genome mutations. The problem with applying these findings in the clinical setting is that there is no central database of genome information. As each mutation was studied, a research paper was published with the details of what was discovered about it. The data has not been consolidated into a searchable database to date, leaving oncologists with no way of sequencing a patient’s tumor, and then reviewing the individual mutations, their positive or negative affect, and any potential therapies that have been identified thus far for combating those negative mutations.

Enter Watson. With 23 million medical research studies within its database, researchers are hoping that it can power a new application that will analyze a patient’s genetic mutations and then present oncologists with recommended treatment plans that will target the most damaging of those mutations with medications that have been shown to work in other trials. Each patient’s overall treatment plan would likely be different, but would be optimized for their tutor’s genetic makeup.

To start the project will focus on a brain cancer known as glioblastoma. Robert B. Darnell, an oncologist and president of the New York Genome Center explains, "It’s as close to a death sentence as you can get." Patients typically survive a year after diagnosis, and surgery, radiation and chemotherapy have only been able to extend like by an average of two months.

For the project, 20 patients will be selected, their tumors will be genetically sequenced, and then Watson will have the opportunity to search its database of medical literature to create a treatment plan that will address the tumor’s harmful mutations. Watson will be free to recommend any treatments, including drugs that may have been studied in conjunction with other cancers, but never glioblastoma.  Oncologists and pharmacists will review the recommendations and move forward with the treatment plan based on the their clinical judgment.

A major drawback with Watson is that the supercomputer has only been loaded with the abstracts from the 23 million studies it has access too, rather than the full text papers. Some oncologists question whether Watson will be able to generate clinically viable treatment plans given the limited amount of detail within its data set. Heidi L. Rehm, a molecular geneticist at Brigham and Women’s Hospital, commented, “It’s only as good as the data going in.”

Because this particular cancer is so devastating, researchers are hopeful that they will know whether Watson’s recommendations are helping in a relatively shot period of time.


Enjoy HIStalk Connect? Sign up for update alerts, or follow us at @HIStalkConnect.

↑ Back to top

Founding Sponsors

Platinum Sponsors