$3.1 million National Cancer Institute grant will support research on colorectal cancer at IU

The award to Paul Macklin of IU and colleagues at the University of Southern California will advance search for new cancer treatments using computer simulations

A $3.1 million grant from the National Cancer Institute will advance an Indiana University researcher's work in the fight against colorectal cancer.

Paul Macklin, an associate professor at the IU School of Informatics, Computing and Engineering and a member of the IU Simon Cancer Center, is one of three recipients on the award. The funds will support the use of computer simulations to explore the complex cellular interactions that fuel the growth and spread of tumors in the body.

Paul MacklinView print quality image
Paul Macklin. Photo by James Brosher, Indiana University

Macklin created PhysiCell -- software that simulates cellular activity -- which will be used in the study. The grant's collaborators are the University of Southern California's Stacey Finley, an expert on studying cellular metabolism in a single cell, and Shannon Mumenthaler, who specializes in creating cell cultures that mimic bodily organs. About one-third of the award will support activities in Macklin's lab at IU.

Over 1.3 million people in the United States are living with colorectal cancer, according to the National Cancer Institute. The disease is the third leading cause of cancer death in men and women in the country, with about 10 percent of those diagnosed dying within five years.

"We chose to focus on colorectal cancer in part because it's one of the leading causes of cancer death," Macklin said. "We also chose it because there are so many unknowns about how the cancer develops, especially at the earliest stages of metastasis."

The first phase of the research will focus on adapting a mathematical model of the metabolic pathways inside cells from Finley's research to Macklin's PhysiCell system. The team will then run simulations that mimic the ways in which various substances resulting from these metabolic processes are created and drive growth inside cancer cells. The system will also examine how tumor cells exchange these substances with other tumor cells and surrounding support cells called fibroblasts.

The interactions between these cells are expected to determine how tumors form and spread. The team will then focus on verifying the software's predictions against bioengineered tissues in the lab that mimic a colorectal tumor.

This phase will take place inside "mini-organs" -- called organoids -- created in Mumenthaler's lab. Organoids are created by seeding 3D scaffolds with cells that form into structures resembling rudimentary organs.

Organoids are an increasingly popular biomedical research tool that allow scientists to safely experiment on highly accurate tissue types without risking patients' health. Mumenthaler's lab uses automated microscopes to take many highly detailed organoid images over time, allowing Finley and Macklin to calibrate and test their software.

"This period is where we can really start to collect data that puts our simulations to the test," Macklin said. "It's here where we can learn whether our models are accurate predictors of tumor growth -- or whether we need to go back and tweak the software."

If the simulations suggest that a specific compound may reduce tumor growth, the organoids will also allow the researchers to try it on real cells -- but only in a few select cases.

Colorectal cancer organoidView print quality image
An image of an organoid representing human colon cancer cultured from a colon tumor. The colored stains reveal the structure's cellular architecture, including nuclei, in green, and cytoskeletal structures, in red. Photo courtesy Shannon Mumenthaler, University of Southern California

"The big advantage of conducting research in a computerized environment is you can afford to fail," Macklin said. "By comparison, lab research is extremely expensive in terms of time and money."

He estimates that PhysiCell can simulate 500 experiments in just a couple of days. In a lab, even one experiment could take weeks.

"If you've got the freedom to experiment, you can really narrow your ideas down until you're spending your experimental budget on the most promising ideas," he said. "You're not trying to go all the way to a clinical trial phase and then failing after spending millions or billions of dollars."

PhysiCell is especially suited to this "fail fast and try again" approach since it's one of the quickest systems of its type.

"There's a lot of ways to simulate a tumor; we think we've found the sweet spot between speed and accuracy," Macklin said.

He said most software either simulates a few cells in high detail or simulates a large number of cells with little detail or mechanical realism. Adding detail increases the simulation's realism, but it also significantly increases the total run time and doesn't always make a major difference from experiment to experiment. Macklin said PhysiCell focuses its energy on simulating the key details that make a difference.

PhysiCell also is the first software to efficiently simulate tissue environments with many chemical factors like oxygen, glucose and metabolites -- a critical need when studying cancer metabolism.

"This means we can simulate a bunch of factors in a fraction of the time required by traditional systems," Macklin said. "Rather than trying to predict everything, we're focused on solving a really narrow class of problems really, really fast."