Researchers at the MD Anderson Cancer Center hope their latest research platforms will enable in-depth understanding of tumor behaviors
By Ivy Ashe
Population Health Scholar
University of Texas System
PhD Student in Journalism
UT Austin Moody College of Communication
Sometimes in research, it’s hard to see the forest for the trees.
That’s often the case in the study of cancer, where investigators focus their studies on specific groups of patients or hone in on the distinct natures of individual tumors. Until recently, there were few frameworks in place for looking at aggregate data from these studies.
At The University of Texas MD Anderson Cancer Center in Houston, Dr. Andrew Futreal and the Adaptive Patient-Oriented Longitudinal Learning and Optimization (APOLLO) project are building a new system for analysis that brings research data and clinical data together in one platform. APOLLO is aimed at longitudinal study of tumor behaviors.
“Tumors, when you treat them, change over time,” Futreal said. “We’ve not done a great job of assessing that change. We know that molecularly it’s very important to understand what goes on in the tumor, but we’ve never really exploited that.”
APOLLO is a multidisciplinary team effort that organizes pre- and post-treatment biopsies, tissue processing and quality assessment, expert pathology qualification, and sample processing for deep, uniform molecular profiling. To date, more than 1500 patients have been part of the effort, comprised of more than a dozen cancer types. The longitudinal plan was funded as part of MD Anderson’s Moon Shots Program, a collaborative effort to accelerate the development of scientific discoveries into clinical advances that save patients’ lives.
The Moon Shots Program consists of 13 disease-focused initiatives and utilizes 10 different research platforms to provide infrastructure and technical support that enable the Program to move more quickly toward clinical advances.
Futreal also works with the Translational Research Accelerator, a purpose-built data platform whose creation was the logical next step from APOLLO. Once you have a large collection of molecular data, what do you do with it? What’s the best way to access it? Most importantly, how do you combine clinical data with molecular data?
“Clinical data comes in a variety of shapes and sizes and stripes,” Futreal said. “Some of the richest is in the physician notes, in the text. Immediately you can see that this presents a problem if you want to extract that information and process it.”
The MD Anderson team went back to 2012, a date that coincided with the start of the Moon Shots Program (2012), and began to clean and organize the hundreds of thousands of records produced since then: diagnoses, lab reports, testing data, and research trial data.
“We want to learn as much as possible from every patient we see, not just clinical trials,” Futreal said.
With the data getting more organized, the team is now looking for partners who can help accelerate development of a platform powerful enough to handle all of the data points and flexible enough to adapt to change if a new algorithm or new style of machine learning is developed. The final TRA platform should also be able to facilitate extensive collaboration, Futreal said.
“There’s a growing and I think admirable push in the cancer community to think about how we can interact more efficiently between institutions and countries,” he said.
That would be a particular boon to researchers working on rare tumor types that may only be seen occasionally at a given institution. With the right platform, the researcher would have access to additional data from cases at other locations.
“We have to provide something that helps researchers do their jobs faster,” Futreal said. “If we can provide a platform that provides an answer to their question in minutes rather than diving into medical records for 2 months, that’s a win.”
Futreal spent more than 12 years working in the United Kingdom prior to joining MD Anderson.
“I think everyone in the oncology space is thinking about these things,” he said. “We have to keep an eye out about what other folks are doing. Are we really going to be collaborative in a way that the sum is greater than the individual parts? We’re going to be able to ask questions across cancer types that have never been able to be asked before.”