New ultrasound technology research by a Carle Illinois College of Medicine professor could change how fast breast cancer patients find out if their initial chemotherapy is working. Some breast cancer patients wait months to learn if their tumors are responding to initial chemotherapy, but Michael Oelze’s new research could shorten the wait to a week.
Professor Oelze and his Canadian research partners were awarded a $2 million grant from the National Cancer Institute to study the use of quantitative ultrasound imaging to identify early response to initial chemotherapy administered to reduce tumor size prior to the patient’s primary treatment.
“The current project isn’t looking at trying to diagnose breast cancer. It’s looking at how early and how well we can detect early response of locally advanced breast cancer patients to neoadjuvant chemotherapy,” said Oelze.
Oelze, who is also a professor in the Department of Electrical and Computer Engineering at the University of Illinois at Urbana-Champaign, is working with Dr. Gregory Czarnota from Sunnybrook Health Sciences Centre in Toronto on the project that will employ quantitative ultrasound imaging technology.
Ultrasound imaging can provide evidence of whether cancer cells in a tumor are undergoing cell death by revealing changes in shape, size, and structure of cells within the tumor. “However, when we scan a tumor at different time points, differences in the way the scans were taken at each time point can reduce the ability to observe changes due only to cell death and the subsequent structural changes in the cells,” said Oelze. “To mitigate these differences, we exploit the use of radiological clips already embedded in the tumors. Radiological clips help radiologists locate, image, and evaluate the significance of changes in the tumor.”
Professor Oelze and his team are using the clips to differentiate imaging changes that result from tumor response from those caused by inconsistencies in how the images were taken. The result is increased accuracy of quantitative ultrasound imaging to detect tissue signal changes in tumor tissue.
Machine learning has played a role in this research. In the past, image processing models have been used to analyze the image signals to assess masses and identify cell death. These models alone are helpful for interpreting the image, but the addition of machine learning allows for a more accurate assessment of the patient’s response by picking up on features that earlier models missed.
The study also includes ultrasound analysis of tumor changes early-on. “We are planning on a study of around 150 patients. Each will have multiple three-dimensional scans of the tumors and multiple image frames we can look at.We will scan patients multiple times at baseline (just before the onset of chemotherapy), one-week post-therapy, two weeks, four weeks, and so forth. The improved analysis and imaging technology will provide more accuracy in identifying early responders and nonresponders at one-week post initiation of neoadjuvant chemotherapy, which is a much earlier time point than can be provided by conventional imaging procedures,” said Oelze.
Oelze credits the Cancer Center at Illinois (CCIL) and Carle Illinois with helping him establish connections at a local hospital to test imaging techniques. “The College of Medicine, the CCIL, and all these things combined will spur this sort of clinical research and collaboration that we really need here at Illinois,” said Oelze.
This new grant and Oelze’s research will also open new doors for enhanced learning experiences for medical students. “Carle Illinois students involved in this work will see the result of the fundamental research going from the lab to the clinic to solve a medically significant problem in the early identification of the response of locally advanced breast cancer patients to neoadjuvant chemotherapy,” Oelze explained.
Oelze is currently working on other collaborative projects for treating tumors with ultrasound-based therapy techniques. His lab members are also working on detecting micro-calcifications in lesions and tumors to improve detection rates in cancer patients. He is affiliated with the Beckman Institute, CSL, and the HMNTL.
The original article from Carle Cancer Center can be found here.