Cancer cells reconnect with their inner child to evade immunotherapy

Turning on an embryonic gene may help some tumors resist immune checkpoint inhibitors
Brightly colored blobs of blue, green and pink with a black background
The green and red glow in an osteosarcoma cell shows that DUX4 is present. Osteosarcoma is the most common type of bone cancer. Image courtesy of the Tapscott and Bradley labs

Reconnecting with a gene from a long-forgotten youth may help adult tumor cells fly under the immune system’s radar, according to work from Fred Hutchinson Cancer Research Center published Thursday in the journal Developmental Cell.

A team of scientists found that a gene called DUX4 — usually turned on just after an egg is fertilized — allows tumor cells to become invisible to attacking immune cells. DUX4 may help to explain why checkpoint inhibitors, a type of immunotherapy that unleashes a patient’s own immune system against their cancer, do not work in all patients. The Hutch scientists saw that higher DUX4 levels were associated with significantly poorer patient survival after checkpoint-inhibitor treatment.

“We identified a new factor that was previously unknown to be present in cancer cells,” said Dr. Robert Bradley, a computational biologist. This factor influences responses to immunotherapy. Bradley co-led the project with Hutch colleague Dr. Stephen Tapscott, an expert in DUX4 who studies the connections between developmental biology and cancer.

The Food and Drug Administration has approved seven checkpoint inhibitors as cancer therapies. The list of cancers for which a checkpoint inhibitor has been approved continues to grow, and it now includes melanoma, bladder cancer and small cell lung cancer, among others. The percent of patients who benefit from checkpoint-inhibitor therapy can vary widely, even among patients with the same tumor type. One study of checkpoint inhibitors in breast cancer estimated that between 5% and 30% of patients benefited.

“It’s very important to understand why some patients respond [to immunotherapy] and some don’t, both in order to be able to give patients advice on what medication would be most effective, and in order to try to design additional therapies that will let more patients respond to immunotherapy,” Bradley said.

How a muscular dystrophy gene got linked to cancer

DUX4 is a transcription factor, a protein that helps turn on genes. Normally, it helps regulate early embryonic development. DUX4 is supposed to shut off permanently in almost all adult tissues. If it doesn’t, it can wreak havoc. Tapscott has spent years studying why DUX4 causes an inherited muscular disorder known as facioscapulohumeral muscular dystrophy, or FSHD, when it’s turned on in adult muscle cells.

He also suspected a connection between DUX4 and cancer. Many DUX4 target genes fall into a category of genes that are only turned on in three contexts: early during embryonic development, sperm development or cancer. These types of genes can be potent stimulators of a cancer-specific immune attack. When Bradley’s team unexpectedly linked DUX4 to cancer, he immediately reached out to Tapscott, and a new collaboration was born.

Bradley and his team were searching for tumor-associated genes that could influence the immune response to cancer.

“We used computational analyses of data from about 10,000 patients to look for genes in cancers that shouldn’t be there,” Bradley said.

Postdoctoral research fellow Dr. Guo-Liang Chew and staff scientist Dr. Amy Campbell drew on molecular analyses of tumor samples stored in The Cancer Genome Atlas database, a joint effort between the National Cancer Institute and the National Human Genome Research Institute. These samples ranged across 33 types of solid tumors, including cancers of the breast, prostate, lung and pancreas. In cancers from 26 different types of tissue, DUX4 was one of the three most highly cancer-associated genes.

But when muscle cells turn on DUX4, they die. So why was it turned on in so many tumor types? Cancer cells often gain growth and survival advantages by reawakening long-dormant genes. Some help them act more like stem cells, which can replicate themselves indefinitely without gaining specialized functions. Muscle cells, in contrast, are highly specialized cells that will never divide again. Clearly, context matters.

To discover exactly what advantages DUX4 confers to tumors, the team looked closer. First, they confirmed that in most cancer cells, DUX4 was active and turned on the same genes in cancer as it does in early development. Then, they began exploring potential connections between DUX4 and the immune response to cancer.

DUX4 turns tumors invisible

Given what Tapscott knew about DUX4 target genes and their potential to draw an immune attack, Chew and Campbell decided to look at immune activation in tumors, expecting to see that DUX4 was associated with higher immune activity. Instead they saw just the opposite: signs that fewer immune cells had mounted attacks on tumors with DUX4.

To discover why, the team focused on how T cells, a specialized type of immune cell, kill cancer cells. T cells can only attack what they can see. Cells make themselves visible to T cells by using molecular “handles” to stick bits of their proteins to their surface. When deciding whether to attack, T cells examine the protein pieces, called peptides, in the molecular handle to see if they are from diseased cells. 

The handle is called molecular histocompatibility complex, or MHC, class I. No MHC, no immune attack. Sure enough, when the team looked at MHC-I levels in tumor cell lines, they found that high levels of DUX4 correlated with low levels of MHC I.

“DUX4 is preventing any peptide presentation, so the [cancer] cell is invisible to the immune system,” Tapscott said.

DUX4 hobbles checkpoint inhibitors

Several types of immunotherapy work through the action of T cells. This includes checkpoint inhibitors such as Keytruda and Yervoy (pembrolizumab and ipilimumab, respectively). Checkpoint inhibitors disrupt the molecular brakes that cancer cells use to check T-cell attack. Once the brakes are removed, anti-tumor T cells are free to kill off cancer cells — in theory. In reality, checkpoint inhibitors are hit and miss: Some patients’ tumors respond beautifully, while others never shrink.

The team wondered if the invisibility caused by DUX4 could be behind some patients’ lackluster responses to the drugs.

The researchers looked at tissue samples donated for research by cancer patients who had taken ipilimumab, also sold under the brand name Yervoy. They saw that patients for whom the drug didn’t seem to work had much higher levels of DUX4 than did patients whose tumors shrank. The researchers also linked DUX4 levels to patient survival after ipilimumab treatment; patients with little to no DUX4 in their tumors survived longer than patients with DUX4-high tumors.

The findings suggest that the presence of DUX4 could help identify patients for whom checkpoint inhibitors may not work well, though that would need further clinical testing to confirm, Bradley said.

Four people standing in a row
Drs. Stephen Tapscott, Amy Campbell, Guo-Liang Chew and Robert Bradley showed that DUX4 helps cancer cells evade immune attack. Photo courtesy of the Tapscott and Bradley labs

Basic biology and better immunotherapy

The early embryo must surmount many challenges to survive and develop into a human being, not the least of which is the mother’s immune system. The findings are “a very good example of how cancer cells are recapturing an early embryonic program to do what early embryos do really well: fly under the immune system’s radar,” Tapscott said.

This benefit may be just the beginning of the benefits tumor cells gain from DUX4, the researchers hypothesize. They’re currently studying whether DUX4 also helps tumors select the fittest cells, a process that also occurs during early embryonic development when the embryo ruthlessly prunes defective cells. The work could shed more light on how a cancer cell’s ability to connect to a distant past sets them up for future survival.

The most obvious clinical application of the work is to test whether targeting DUX4 improves responses to checkpoint inhibitors, Bradley said. Drugs targeting DUX4 in FSHD, including one developed by Tapscott, are already in the pipeline.

“The question is, could the same approach or the same drug work when applied to cancer?” Tapscott said. The team is starting to test the effectiveness of such drugs against DUX4 in cancer cells in the lab.

Sabrina Richards, a staff writer at Fred Hutchinson Cancer Center, has written about scientific research and the environment for The Scientist and OnEarth Magazine. She has a PhD in immunology from the University of Washington, an MA in journalism and an advanced certificate from the Science, Health and Environmental Reporting Program at New York University. Reach her at

Are you interested in reprinting or republishing this story? Be our guest! We want to help connect people with the information they need. We just ask that you link back to the original article, preserve the author’s byline and refrain from making edits that alter the original context. Questions? Email us at

Related News

All news
Study uncovers new proteins involved in regulating muscular dystrophy-linked gene CRISPR-based proteomics technique reveals potential therapeutic targets for FSHD, a rare but devastating disease March 13, 2018
Machine learning provides new insights into tumor suppression Algorithm illuminated surprising metabolic effects of important cancer gene May 7, 2018
Revealing a new way that cancer can evade immunotherapy — and, maybe, how to stop it How a new technology solved a mystery and honors one man’s legacy September 24, 2018

Help Us Eliminate Cancer

Every dollar counts. Please support lifesaving research today.