Most researchers focus on the ways cancers — and the people who have them — are similar. For Thomas Yankeelov, there’s hope in the distinctions.
Thomas Yankeelov, PhD, was already an aspiring cancer killer when, on a Florida beach five years ago, he read “The Origins of Computer Weather Prediction and Climate Modeling.”The paper, by Peter Lynch, PhD, a professor of meteorology at University College Dublin, recounts the human failure during the last hundred or so years to predict the weather. Yankeelov, then a cancer researcher at Vanderbilt University, noticed that people’s historically weak understanding of the weather bore surprising similarity to their struggle to grasp the so-called Emperor of All Maladies.
“What people call ‘cancer’ is really a group of more than 100 different diseases, and obviously every person is unique,” says Yankeelov, who received a $6 million recruitment grant from the Cancer Prevention and Research Institute of Texas and holds a joint faculty appointment at the University of Texas’ Dell Medical School and the Cockrell School of Engineering’s Department of Biomedical Engineering. “It becomes much easier to fight when we understand the mechanisms driving it — how a particular type of cancer is likely to grow in a particular person. We can only do that with patient-specific information.”
An Intuitive Approach — and a Massive Supercomputer
Yankeelov brings a low-key kinetic energy to the cancer fight. In a uniform of rumpled shirts, faded jeans and canvas shoes, he is at turns enthusiastic about the possibilities of his work and grave about its stakes.
His personal style belies the profoundly technical nature of his research, which revolves around complex mathematical models, sophisticated imaging equipment and, in the Texas Advanced Computing Center, one of the most powerful supercomputers in the world.
Beyond that, it’s pretty intuitive. He uses the best available imaging and science to learn everything possible about a patient’s cancer. He then runs thousands of computer simulations using complex statistical forecasting equations and a methodology similar to the one that meteorologists use to predict the weather. Each simulation draws data from the patient’s unique biology and the cancer’s specific traits, considering both how similar tumors have behaved in others and how this particular tumor is likely to respond in this particular patient.
The result is a prediction — based on data, biology and mathematics — of how one tumor or set of tumors is likely to respond to a phalanx of available treatments. The forecast will ultimately help guide physicians’ decisions about how to best fight the disease, especially as the cancer responds to treatments and provides yet more data to the people fighting it.
Every physician brings experience, education, learning and intuition to a patient’s cancer fight. Yankeelov — working alongside Stephanie Eldridge, PhD, David Hormuth, PhD, and Ernesto Lima, PhD — wants to add math and data to their arsenal.
“Once a science is mathematized, we become much better at making testable predictions,” Yankeelov says. “That’s when we make much more rapid progress. It is not unreasonable to believe that oncology will resemble other sciences in this way.”
The disease, a scourge everywhere, has a special resonance in Austin and Central Texas. In most places, heart disease is the leading cause of death; in Travis County, it’s cancer. And Austin is home to the LIVESTRONG Foundation, which has pushed to reorient cancer care around patients, their lives and their wishes. The foundation donated $50 million to the Dell Medical School in order to create the LIVESTRONG Cancer Institutes, where Yankeelov leads cancer imaging research. He also serves as director of the Center for Computational Oncology at Texas’ Institute for Computational Engineering and Sciences, directed by world-renowned computational scientist J. Tinsley Oden, PhD.
Yankeelov’s work “leverages the immense computational and biological expertise at UT Austin to develop novel ways to select the best treatment for cancer patients, assess resistance mechanisms and ultimately translate this information into effective therapies,” says S. Gail Eckhardt, MD, inaugural director of the LIVESTRONG Cancer Institutes. “A big and complicated problem like cancer requires a big and complicated approach to improve therapy. Investigators like Dr. Yankeelov are well-positioned to take on this challenge and work with oncologists to develop more effective treatments for cancer patients.”
Others agree with Eckhardt. “Dr. Yankeelov’s work is exciting. Using advanced computational methods to better understand imaging information — and to integrate this data into other medical realms — will help advance the fight against cancer,” says John D. Hazle, PhD, who holds the Bernard W. Biedenharn Chair in Cancer Research and chairs the Department of Imaging Physics at MD Anderson Cancer Center in Houston. “We are already generating an immense amount of information regarding cancer characteristics and response to therapy. We need to integrate imaging data into these platforms as well to help us understand how cancer evolves and responds to therapy.”
Cancer as Math
President John F. Kennedy’s challenge to put humans on the moon by the end of the 1960s launched perhaps the most publicly successful effort to solve a science problem. It certainly inspired President Barack Obama and Vice President Joe Biden, who invoked it in naming their Cancer Moonshot initiative to accelerate cancer research and treatment.
But to Yankeelov, the differences between a cancer cure and a moonwalk illuminate why the former is so elusive.
By 1961, when President Kennedy set out his challenge, laws of motion and gravity had been clear since Isaac Newton wrote them down in 1687. Spaceflight and moon landings required an enormous amount of engineering know-how, but the rigorous mathematical theory guiding the effort was well-established.
Cancer is a bigger mystery. Tumors evolve and mutate from a huge variety of triggers inside and outside the body, and they can react very differently to identical treatments.
“We don’t have a mathematical description of cancer the way we did of spaceflight,” Yankeelov says. Finding that mathematical description is the focus of Yankeelov’s work. But he also wants to flip a system in which healers and institutions tend to be seen as unique and special — products of a physician’s learning and experience — while patients tend to be treated as standardized vessels for a generic disease.
Instead, he wants to create a new model that:
Researchers have spent generations sifting for cancer-fighting clues among the commonalities of survivors and victims. For Yankeelov, there’s also hope in the distinctions.