AUSTIN (KXAN) — Dr. Thomas Yankeelov of the University of Texas at Austin might specialize in cancer research, but it was a deep dive into a meteorology book a decade ago that helped inspire new research into detecting the effectiveness of breast cancer treatments in patients.
Before math and physics factored into meteorology research, the field was reliant on studying the environment and subsequent changes. Yankeelov said he was drawn to the prospect of incorporating mathematical research into oncology to help better serve cancer patients.
“The idea was to try to build mathematical forecasting models that you can then apply to the individual patient,” he said. “We try to build these mathematical models that describe the underlying physics and biology of cancers and how they grow and how they invade — how they respond to treatment.”
That was the impetus behind new research that can determine the efficacy of various cancer treatments — like chemotherapy, hormone therapy and immunotherapy — in patients after one therapy treatment cycle.
The research was developed through a collaboration between UT Austin’s Oden Institute for Computational Engineering, the Livestrong Cancer Institutes at Dell Med, Texas Oncology, Dell Seton Medical Center and the Cockrell School’s Department of Biomedical Engineering.
When patients are initially diagnosed, researchers run an assessment on the current status of cancer in their body before treatment is introduced. Those measurements serve as a baseline before therapy starts.
After one round of therapy treatments — which varies from seven to 21 days, depending on the therapy — researchers take a second round of measurements to see the success rate and build a prediction on whether the patient should continue this specific therapy type.
So far, the accuracy rate is significantly high.
“Using those two time points, we’re getting an accuracy of about 88% of the time we can we can predict who’s going to respond and who’s not going to respond,” he said.
From there, Yankeelov’s team is looking to expand that research into predicting the efficacy of a patient’s treatment response, before treatments even begin. This approach would combine this oncology forecasting model with a patient’s individual biological makeup and the current cancer growth to predict the efficacy of various treatment options before spending time undergoing said treatments.
“We’ve done the retrospective stuff that shows that we can use this technique to make a pretty accurate prediction of who’s going to respond and who’s not. Now the next step is to say, ‘Okay, do we really trust this and will it actually change the patient’s therapy?'” Yankeelov said. “Because, if you can predict who’s going to not respond correctly about 88% of the time, then you have good motivation for saying, ‘Okay, we need to get off of this therapeutic regimen.'”
The Oden Institute’s research team has taken their analyses and partnered with Texas Oncology and Austin Radiological Association to serve oncology patients at community cancer treatment centers.
The vast majority of patients don’t receive their treatments at a research facility. Yankeelov said his hope is that more awareness on the research, paired with this community collaboration, will help expand the Oden Institute’s research subject pipeline and further the study.
When Yankeelov began his research a decade ago, he said mathematics wasn’t as prevalent in cancer research. Now, he said he’s inspired by the growing interest in computational oncology and the advancements being made for cancer patients.
“It is definitely changing the way that people think about cancer,” he said, adding: “When you bring in people from different backgrounds, they have different experiences, different expertise. You have them start talking together, you can go in new directions.”