Each project in the 2022-2023 program in Oncological Data and Computational Science will be awarded $50,000 and allocations of 12,500 core hours of supercomputer time. The new research projects are as follows:
Forecasting the lung’s functional response to cancer radiotherapy
Edward Castillo, PhD, associate professor of Biomedical Engineering at UT Austin
Julianne M. Pollard-Larkin, PhD, associate professor of Radiation Physics at MD Anderson
Standard image-guided radiotherapy (RT) planning for lung cancer delivers a prescribed dose to the tumor while protecting critical organs, but current planning approaches assume lung function to be equal throughout. These researchers showed that RT plans designed to avoid functional regions of the lung can reduce both the incidence of lung toxicity and the magnitude of post-RT pulmonary function loss.
They propose developing a post-RT pulmonary function forecasting model that could ultimately be used to determine RT plans that optimally preserve lung function.
Developing patient-specific models to predict triple-negative breast cancer response to neoadjuvant chemotherapy
Gaiane Rauch, MD, PhD, professor of Abdominal Imaging, MD Anderson
Ernesto Lima, ScD, research associate at the Oden Institute’s Center for Computational Oncology and TACC
Patients with localized triple-negative breast cancer (TNBC) typically are treated with neoadjuvant chemotherapy. However, approximately 50% of patients do not respond well to standard-of-care regimens. Thus, new methods are needed to improve targeting and evaluation of therapy response in these patients. The project’s goal is to develop a personalized clinical-computational model as a ‘digital twin’ to predict response to neoadjuvant chemotherapy in patients with TNBC. The proposed model could allow personalized planning of neoadjuvant chemotherapy before treatment begins and identify patients unlikely to benefit, possibly saving them from unnecessary therapies and side effects.
Single-cell network-based transfer learning model for designing precision medicine in colorectal cancer
Stephen Yi, PhD, assistant professor of Oncology and director of Bioinformatics at UT Austin
Scott Kopetz, MD, PhD, professor of Gastrointestinal Medical Oncology at MD Anderson
A mechanistic understanding of cancer requires finding and understanding the molecular profiles and cellular behaviors underlying tumor heterogeneity. In single cells, distinct signaling circuits control biological signaling. Laboratory studies to dissect these pathways can be time-consuming and expensive, but a deeper understanding can reveal opportunities for more effective combination therapies.
The researchers aim to accelerate the discovery of innovative, actionable combination therapies that can be prioritized for clinical evaluation through integrative network-based computational identification of functional alterations in cancer patients to facilitate therapeutics.
To tackle this challenge, they propose to use single cell network-based transfer learning to identify functional alterations in tumor cells from patients with colorectal cancer in order to best match cell lines with known susceptibility to drug combinations.
Achievements So Far
There already has been significant research output from the collaboration with several papers published in the last 12 months.
In a recent edition of Cancer Research the research team led by Chengyue Wu of the Oden Institute and Gaiane Rauch of MD Anderson, funded in round one (2020-21), reported the development of a patient-specific mathematical model that integrates magnetic resonance imaging (MRI) data with biological data to forecast treatment response and design optimal therapy strategies for triple-negative breast (TNBC) cancer patients. The researchers will build upon this early success to improve and validate the model in the newly funded research project.
Further, the research team led by David Hormuth of the Oden Institute and Caroline Chung of MD Anderson, published new results in Scientific Reports that describe an imaging-based approach to predict responses to high-grade gliomas following combination chemotherapy and radiation treatment. Also funded in the first round of projects, the team developed mathematical models capable of incorporating detailed MRI data to follow tumor progress and forecast individual responses. Based on this initial work, the research team recently was awarded a research grant from the Cancer Prevention and Research Institute of Texas (CPRIT) to continue developing this approach to personalize therapy.
Tom Yankeelov is Scientific Lead of the collaboration from the Oden Institute. He is also Director of the Oden Institute’s Center for Computational Oncology.
“This third round of pilot projects demonstrates that our tri-institutional collaboration is truly thriving as we move into new areas of research. It is so exciting to see not just how well the different groups are working together, given their different areas of expertise, but also how they are producing significant discoveries that are leading to major breakthroughs in cancer.”