Pushing toward personalized pancreatic cancer treatments

// Biomedical Engineering

Tags: Faculty, research

Photo of pancreatic cancer tissue

A fluorescence lifetime image shows pancreatic cancer, with collagen fibers in red and yellow and cancer cells in green and blue. Photo courtesy of Melissa Skala/Optical Microscopy in Medicine Lab.

Share this story:

The numbers don’t lie: Pancreatic cancer is the deadliest form of the pernicious disease. Its five-year survival rate of 8.5 percent, per the most recent data from the National Cancer Institute, is the worst among all cancer sites.

“It is the most lethal cancer because we have no really effective way to treat it,” says Melissa Skala, an associate professor of biomedical engineering at the University of Wisconsin-Madison and a principal investigator at the Morgridge Institute for Research.

Skala and Paul Campagnola, a professor of biomedical engineering at UW-Madison, hope to make inroads toward improved drug therapies through a two-year National Institutes of Health Exploratory/Developmental Research Grant. They plan to use Campagnola’s novel imaging and micropatterning techniques to create a 3D model of pancreatic ductal adenocarcinoma, which will improve their ability to identify effective drugs.

Current testing relies either on mouse models, which don’t match the human disease in aggressiveness or the start of tumor development, or on two-dimensional cultures that generally fail to account for interactions between cancer cells and those in the stroma, the tumor’s hefty supportive tissue. That’s a key missing ingredient, because the stroma protects the cancer cells from drugs while also encouraging tumor growth.

To build their model, the UW-Madison engineers will take high-resolution images of cancerous tissue using second-harmonic generation microscopy, then print small-scale 3D replicas of the tissue’s extracellular matrix, part of the stroma that creates structure. By bringing together cancer cells and support cells in that realistic environment, they’ll see how the cancer cells move, divide and produce energy—and which drugs stop them.

In time, that could allow researchers to model a particular patient’s cancer and test different treatment strategies, from chemotherapy to targeted therapies.

“This is really going after personalized medicine,” says Campagnola. “We’ve spent years and years building up the technology to do all this stuff and now we’re really ready for primetime.”

Campagnola and Skala are both members of the UW Carbone Cancer Center.

NSF grant for ovarian cancer imaging and analysis

Campagnola and collaborator Vikas Singh, an associate professor in the Department of Biostatistics and Medical Informatics and the Department of Computer Sciences, also recently earned a National Science Foundation Early Concept Grant for Exploratory Research.

For that project, Campagnola will further develop a method of 3D imaging tissues in ovarian cancer. Singh will then analyze the imaging data and develop algorithms aimed at sharpening diagnoses and prognoses.

“Deep learning is normally done with large datasets, like millions of people. We don’t have that,” says Campagnola. “It’s a whole different way of doing deep learning and we are excited to see how it’s going to work.”

Author: Tom Ziemer