Casey Stowers Wins 2026 Oden Institute Outstanding Dissertation Award

Triple-negative breast cancer (TNBC), an uncommon subtype of cancer, gets its name from what it lacks: overexpression of three receptors that could have facilitated targeted treatment. Recent Ph.D graduate Casey Stowers’s dissertation is also structured as a trio: three aims that address what can be done in the wake of these vacancies.

Stowers’s dissertation, “Integrating mechanism-based and data-driven modeling to predict the response of triple-negative breast cancer to therapy,” was the culmination of her Ph.D. research at the Oden Institute for Computational Engineering and Sciences and earned her the institute’s 2026 Outstanding Dissertation Award. The award recognizes exceptional research that is impactful, original, and cutting-edge. Her advisor, Tom Yankeelov, who directs the Center for Computational Oncology, said she deserves this award for “the thoroughness and completeness with which she attacked and solved problems.” He added,  “When Casey goes after a problem, if humans are capable of solving it, then Casey will solve it.”

While pursuing her Computational Science, Engineering, and Mathematics (CSEM) Ph.D., her goal was to build a more accurate and efficient model to improve the predictions of TNBC response to therapy. In TNBC, the patient lacks overexpression of those key receptors, which means that they cannot benefit from targeted drugs. Instead, these patients generally need systemic therapies like chemotherapy that attack all of their cells, resulting in dramatic side effects. Stowers wanted her model to support therapy optimization: killing the most cancer while minimizing the side effects. 

Professor Yankeelov said, “As there are currently no accepted methods of predicting the response an individual patient will achieve with a particular therapeutic regimen, it is difficult to overstate the importance of predicting and optimizing the response of cancer to therapy.” He also added that “the mathematical and computational formalism she developed applies to any cancer for which the requisite data is available. Thus, her work has an impact beyond “just” breast cancer.”

To accomplish this, she structured her research around three distinct aims, and through the process, she became a more independent researcher. “For the first project, Professor Yankeelov led the process of coming up with the idea. For the second project, we were more collaborative, and for the third project, I tried to be as independent as possible,” said Stowers, sharing the collaborative relationship she had with her advisor.

For the first aim, she implemented a convolutional neural network (a type of deep learning model that is well-suited for imaging data) to predict treatment response from a single pre-treatment image, rather than requiring imaging data across the patient’s whole cancer journey. While it successfully predicted responses, it was less accurate than using the full time course of data. 

As she began the second aim, Stowers poached an idea from ecology: habitats. Just as different habitats shape how organisms behave, different regions within a tumor might grow or respond to therapy differently. By grouping parts of the tumor into habitats, Stowers discovered she needed fewer parameters to describe the system, making her model more efficient. Her analysis showed that without specifying where these regions were in a tumor, the model produced habitats that were spatially connected, an exciting result that confirmed it was capturing something biologically real. 

Stowers’s third aim investigated the mechanics of how a tumor shrinks and grows under stress. The existing model focuses on the impact of tissue mechanics on tumor growth, but Stowers wanted to explore its effects in cases where the tumor shrinks. Her final results discuss trade-offs between the accuracy and efficiency of various model options. 

Wrapping up her thesis and defending her research, Stowers said she “felt like sprinting to the end of a really long race and then coming to a super sudden stop when I passed. It was rather jarring to me, and the dissertation award reminded me to celebrate the end a bit more. Since I am not planning to pursue an academic career, it feels like my time in research ended on a high note.” Having completed this phase of her career, Stowers has pivoted to working as a Data Analyst for a company that develops math educational software for children, a different kind of problem, but one that continues utilizing her computational skills to help others. 

Information regarding the criteria for the Outstanding Dissertation Award is here