Lynda Chin, MD, used the Lila and Murray Gruber Memorial Cancer Research Award and Lectureship to explain how advanced computer technology can help physicians better understand complex cancer treatments.
Genomic advances have greatly improved the options for treating melanoma and other cancers, but the science is complex, making it difficult for clinicians to keep up. With that in mind, a leading cancer treatment center is teaming up with IBM to use its Watson computer technology to make those advances more accessible through the “Oncology Expert Advisor.”
“Melanoma has become, in the matter of a decade, the most exciting field in cancer. It is the poster child for genomic-based therapy and immune-based therapy,” said Lynda Chin, MD, who delivered the Lila and Murray Gruber Memorial Cancer Research Award and Lectureship Sunday. “Understanding the molecular mechanisms can completely change the way we think about how to target a disease and achieve a significant response rate and even a durable cure.
“But that rapid change all comes down to understanding the biology and having the therapy that goes with it. How do we make sure the practicing oncologist learns this new knowledge coming out in such an explosive way?”
Appropriately, Dr. Chin frames the challenge in the form of a question because one proposed answer has links to a television game show that wants its answers presented in the form of a question — Jeopardy. It was on the iconic game show that the IBM Watson computer competed against humans, and won.
“How do we take that capability and develop it to be a cognitive clinical decision support system?” asked Dr. Chin, professor and chair of genomic medicine and scientific director of the Institute for Applied Cancer Science at the University of Texas MD Anderson Cancer Center, Houston. Her lecture, “Genomic Medicine: Transforming Research and Patient Care,” addressed both the advances in genomic-based treatments for melanoma and the proposed system to share that knowledge with clinicians to improve outcomes.
Dr. Chin presented an update on the most recent melanoma data from The Cancer Genome Atlas (TCGA). Melanoma is now molecularly classified into four genotypic subtypes — the BRAF mutation, the NRAS mutations, NS1 loss-of-function mutations, and a group that is wild type for those three signature mutations.
“We named the fourth category a wild type because it didn’t have any of the other three mutations,” she said. “This is the first time we have a pretty clear molecular subclassification of melanoma based on the TCGA data, which is a large enough cohort for us to do that.
“Along with this genomic revolution is the ability of advanced analytics to extract insights from these complex data sets. We recently demonstrated the power of such a system biology approach in sophisticated model systems with omic technology to define a non-obvious combination treatment strategy for NRAS mutant melanoma. Our preclinical data has led to human clinical trials showing efficacy.”
Translating those insights into clinical trials and eventually clinical standards requires collaboration between clinicians and researchers, improved communication, and education of the practicing physicians. That is where the Watson-powered Oncology Expert AdvisorTM comes in.
The Oncology Expert Advisor offers up-to-date, evidence-based knowledge personalized to a patient on demand. It is not a static system that gets periodic updates. Instead, it is powered by Watson, a third-generation cognitive computing technology that has artificial intelligence-like capability that learns as it runs, so updated clinical and research data can be incorporated. It also is scalable so its capacity can grow. The latest version of Watson is 240 percent faster and significantly smaller than the version that won on Jeopardy.
“We want to make the knowledge and expertise of an expert for a specific type of cancer available and accessible by any physician who can tap into that knowledge at any time — on demand — and then tailor it to the patient,” Dr. Chin said. “If we can succeed in doing that, it will allow us to share the rapidly progressing and exploding knowledge broadly to all practicing physicians. I call it ‘democratization’ — making sure the physician community can get access to level the playing field when it comes to the knowledge base so they can make better decisions for their patients.”
The prototype of MD Anderson’s Oncology Expert Advisor powered by IBM Watson is built for leukemia.
“The goal is to expand the Oncology Expert Advisor to other major tumor types, and to augment and enhance functionalities of the system as Watson core technology improves. In parallel, we will pilot democratization using this prototype in MD Anderson’s network. The hope is that in 18 to 24 months we will have a cognitive clinical decision support system that can help a lot of oncologists in community settings make evidence-based decisions tailored to their patients. Of course, that is making the assumption we get the support to do it and everything goes right,” Dr. Chin said.