To trace the cause of metastasis in pancreatic cancer, the CSHL team used cells sampled from six mice to grow paired sets of pancreatic organoids—balls of cells that develop in 3D culture. The mice had gene mutations that induced pancreatic ductal cancer. In each of the mice, normal pancreatic ductal cells were compared with cells from the same animals in three distinct stages: premalignant, malignant, and metastatic. Experiments revealed that cancer cells became capable of metastasis via an epigenetic reprogramming of gene enhancers, reverting them to an earlier developmental state characterized by proliferation and increased mobility. [Vakoc Lab/CSHL]


Pancreatic cancer is a killer; 85% of patients die within nine months of diagnosis. A new study sheds light on how the cancer spreads throughout the body.

The study (“Enhancer Reprogramming Promotes Pancreatic Cancer Metastasis”), published in the journal Cell by researchers at Cold Spring Harbor Laboratory, reports that the cancer’s spread is controlled by epigenetics—changes that aren’t hardwired into DNA, but affect how genes are expressed. To make this discovery, scientists grew and tested balls of cells that mimic the shape and behavior of the pancreas, known as pancreatic organoids. These organoids may one day lead to personalized cancer treatments.

In a few years, pancreatic cancer will become the second-leading cause of cancer death in the United States—eclipsing colon and breast cancer—according to Howard Crawford, Ph.D., director of the pancreas research program at University of Michigan. But pancreatic cancer garners far less public attention than other malignancies.

“That’s because we don’t have any survivors,” Crawford tells GEN. “We don’t have people that can…bring a lot of press. We all have to rely on the patients’ families and loved ones to raise awareness. And that’s a challenge.”

The cancer is so deadly because pancreatic tumors regularly break off and spread to far-flung regions of the body—a process known as metastasis. Scientists have tried to identify genes that control the cancer, but genetics don’t tell the whole story.

“We have a pretty good understanding of how pancreatic cells become pancreatic tumor cells,” said Chang-Il Hwang, Ph.D., postdocoral fellow and co-first author of the study. “We don’t know how they metastasize to distant organs.”

To understand the cancer’s spread, Hwang and colleagues collected pancreatic tumors and their metastases from mice and grew the cells in a dish. The cells formed tiny 3D structures known as organoids, which looked and acted like pancreatic cells.

When the researchers compared organoids from the initial tumor to organoids from the metastases, they didn’t find major genetic differences. But they did see that metastatic organoids had more active enhancers—short regions of DNA that boost gene expression by binding to proteins.

The roughly 800 enhancers active in metastatic organoids were linked to embryonic pancreas formation. In effect, metastatic cells were turning back the clock and reverting to an earlier state in order to leave the pancreas.

The researchers analyzed the DNA sequences of the enhancers to find the protein that binds to them, and found FOXA1. When they expressed high levels of FOXA1 in organoids and injected them into the tails of mice, the organoids spread to the lung—a sign of metastasis. But when the researchers injected mice with organoids lacking FOXA1, they didn’t metastasize.

The scientists also checked human pancreatic tissue samples and found that FOXA1 increased with disease severity—consistent with its role in metastasis. Hwang is now working to better understand how FOXA1 works in order to develop future therapies.

“The future goal will be to try to utilize this information to benefit metastatic pancreatic cancer patients,” said Hwang.

Because organoids are grown from a patient’s cells, Hwang and others may be able to use them to personalize cancer treatments. A researcher could grow organoids from a tumor, treat those organoids with a variety of drugs, and see which drugs work best before administering the drug to a patient. But this takes time—something that pancreatic cancer patients have in short supply.

“It takes almost a month or more to establish a good organoid culture from a pancreatic patient,” said Crawford. “If a patient has six to nine months to live, that’s not a lot of time.”

Crawford believes the key is earlier diagnosis. Ten percent of patients have a family history of the disease and genetic markers that put them at risk. He thinks these people should be screened early and often. But screening the rest of the population will be a challenge.

“We have to have a fairly perfect way to screen [the] population,” he said. “Even with a 98% or 99% success rate…there’s a large number of people there that would falsely be diagnosed and a few that would be missed.”

Previous articleCell-Based Therapies Soften Up Cancer by Targeting Tumor Stiffness
Next articleA Gene Fix for a Mouse Model of Duchenne?