IBM’s Watson takes on cancer
NEW YORK—Watson is the cognitive computing system best known for its success vs. human contestants on the quiz show “Jeopardy!,” but now it’s taking on an even more formidable challenge: Identifying effective treatments for cancer.
To that end, IBM has announced a new collaboration between its Watson Health initiative and several leading cancer institutes. The partnerships are intended to accelerate the ability of clinicians to identify and personalize treatment options for their patients. The institutes will gain access to advanced cognitive capabilities that quicken the process of personalizing treatment options by translating DNA insights, analyzing a person’s genetic profile and gathering relevant information from medical literature.
“We believe that Watson could be cancer’s biggest enemy,” Lauren O’Donnell, vice president and general manager with IBM Life Sciences, tells DDNews.
The project is part of a broader strategy by IBM to extend its reach into the healthcare industry and specifically expand its Watson Health initiative, a program with the goal of advancing patient-centered care and improving health. Earlier this year, the company announced that it was establishing a cloud-based system called Watson Health Cloud that will provide a secure and open platform for physicians, researchers, insurers and companies focused on health and wellness. The program, which is aligned with the Health Insurance Portability and Accountability Act, will enable secure access to personal health data so that clinicians can develop more individualized treatment options.
“IBM is really getting serious about getting into the business of health, and we believe that we are going to transform the industry,” says O’Donnell. “We’re seeing big disruptions in the marketplace right now—including the increasing importance of the cloud and secure access to data—that create exciting opportunities for us.”
The newly announced collaborations with cancer institutes will enable clinicians to use Watson with a much broader set of patients by the end of 2015, making personalized medicine more accessible to cancer patients. Experts see cancer treatment as an area of medicine that is particularly well positioned to benefit from more individualized therapies. Most of the 1.6 million Americans who are diagnosed with cancer each year receive surgery, chemotherapy or radiation treatment, but when these standard treatments fail, some patients are beginning to benefit from treatments that target their specific cancer-causing genetic mutations. These kinds of treatments are expected to become possible for an even greater number of patients as genetic sequencing becomes increasingly accessible and affordable.
However, the process is time-consuming and requires clinicians to sift through and reconcile a deluge of genetic information. A single patient’s genome, for example, represents more than 100 gigabytes of data. Moreover, clinicians must also process large amounts of other health information, such as electronic medical records, journal studies and clinical trial information.
Steve Harvey, vice president of IBM Watson Health, tells DDNews that one of the most significant challenges preventing broader adoption of personalized medicine solutions is the volume of the data that must be processed and analyzed. Watson, he says, can help clinicians quickly sift through this data and provide valuable insights on cancer-causing mutations and medical literature. “It’s very challenging to scale these kinds of treatments,” he says. “The cognitive computing and natural language processing abilities of Watson allows us to offer a unique value proposition that addresses this problem of scale. Watson can aggregate both standard databases that already exist and also read through the literature to find specific items that we can point clinicians to very quickly.”
It often takes weeks for clinicians to analyze the specific mutations found in a patient, review the available medical literature and then identify a tailored treatment option. Watson has the ability to complete the genetic material and medical literature review process in only a few minutes. The result of this process is a report and data visualization of the patient’s case that includes evidence-based insights on potential drugs that may be relevant to an individual patient’s unique DNA profile identified in the medical literature. The clinician can then evaluate the evidence to determine whether a targeted therapy may be more effective than standard care for the patient.
IBM views the cloud-based nature of the program as a major strength. Watson’s rationale and insights are expected to continually improve as participating institutions use Watson to assist clinicians in identifying cancer-causing mutations. This will make the latest combined wisdom of the world’s leading cancer institutes available to oncologists. Moreover, the insights produced by the system will be available to all participating institutions, increasing access to up-to-date genetic information.
“You used to have to go to Mayo Clinic or another top cancer institute if you wanted the best data or treatment options,” O’Donnell says. “This program has the potential to allow patients to have access to good medical data anywhere in the world.”
The institutions that are among the first to subscribe to the program include Ann & Robert H Lurie Children’s Hospital of Chicago, BC Cancer Agency, City of Hope, Cleveland Clinic, Duke Cancer Institute, Fred & Pamela Buffett Cancer Center, McDonnell Genome Institute at Washington University in St. Louis, New York Genome Center, Sanford Health, University of Kansas Cancer Center, University of North Carolina Lineberger Cancer Center, University of Southern California Center for Applied Molecular Medicine, University of Washington Medical Center and Yale Cancer Center. IBM is currently communicating with other institutions that may join the program.
“Determining the right drug combination for an advanced cancer patient is alarmingly difficult, requiring a complex analysis of different sources of Big Data that integrates rapidly emerging clinical trial information with personalized gene sequencing,” said Dr. Norman Sharpless, director of the University of North Carolina Lineberger Comprehensive Cancer Center. “We are partnering with IBM in an effort to solve this decision problem with the help of cognitive technology and to improve the decisions we make with our patients to maximize their chance for cure.”