One of the most important parts of science is asking the right questions. As an engineer trying to break into medical research, Joseph Butner, Ph.D., found that some of the questions he was trying to answer weren’t actually all that important to oncologists in the clinic and they weren’t translating into new or better treatments for patients.

But Butner was certain that if the right mathematical methods that have traditionally been used to solve engineering and physics problems were applied to medicine, then he could tap into opportunities and advancements in therapeutic approaches that could change patients’ lives.

He felt compelled to keep chasing the biggest problems in cancer treatment. He knew if he was going to get anywhere, he needed to team up with clinicians who had a better understanding of diseases, treatments and the unique challenges they face in clinical practice. If they could combine their vast knowledge of cancer with his deeper understanding of mathematical models, Butner was certain they could make advancements.

Partnering with oncologists

In October 2023, Butner came to MD Anderson, where he is now an assistant professor in Radiation Oncology. He said clinicians haven’t always been open to the idea of working with mathematicians. But that’s starting to change, and MD Anderson is leading the way.

“MD Anderson is full of innovative thinkers who were not only willing but excited to think outside the box to give mathematical modeling a chance, and this forward thinking has now led to the exciting new Institute for Data Science in Oncology,” Butner says.

Butner has used mathematical equations to predict how effective checkpoint inhibitor therapy, a type of immunotherapy, will be in solid tumors, and how to more accurately determine surgical margins from calcification seen in mammograms in patients with ductal carcinoma in situ, a type of breast cancer.

Looking to the future

Now, he’s looking to use equations to better understand the way combination therapies interact to revolutionize how individual treatment plans are designed.

“It’s common for mathematical models to describe only one type of treatment, but in the clinic, patients often receive multiple therapies to better address their individual needs,” Butner says. “These therapies may interact in complex ways, and if we can quantify these interactions and how they affect the patient and the disease, then we can use known optimization methods to engineer personalized treatment plans to better achieve long-term disease control.”

He and his team are now building combination models of immunotherapy used together with radiation, and they are already planning models of other combination therapies.

Butner hopes that the right equations will eventually change the way treatment plans are currently designed by providing oncologists with one or more optimized personalized treatment plans that can be recalculated and adapted to the unique needs and goals of each patient as their disease changes over time.

Butner often hears others ask, “Math is so complicated. Can’t we just use artificial intelligence (AI) instead?”

Butner believes that artificial intelligence, like mathematical equations, is another tool we can use to make progress on cancer treatment.

“In the future, we’ll use both AI and computational models,” he says. “AI excels when large amounts of data are available, but this is not always the case for new technologies or approaches, like those used in the groundbreaking clinical trials happening every day at MD Anderson. Equations offer a powerful way to fill this gap; often, they can be used reliably with data from only a single patient. It’s up to us as cancer researchers to determine how we think innovatively to really go beyond the limits of each method. I think that the future involves combining AI and equations in clever ways to leverage the unique strengths of each approach to exceed what we can currently do.”

Learn about research careers at MD Anderson.