Skip to main content

A subset of cancer-associated fibroblasts deposits perlecan in pancreatic tumors driving metastasis and chemoresistance

Pancreatic cancer is one of the most aggressive forms of solid cancer, with a 5-year survival rate of less than 10%. By the time patients are diagnosed, the tumor is often inoperable and has spread to distant sites. A hallmark of a pancreatic tumor is the highly remodeled stroma, a tissue that surrounds pancreatic cancer cells. This tumor stroma comprises in part a dense matrix, which is thought to protect cancer cells against chemotherapy, and provides a highway for dissemination to distant sites.

In a recent study published in Nature Communications, HFSP fellow Dr. Claire Vennin and co-workers interrogated how the genetic features of pancreatic cancer cells shape their surrounding stroma. In particular, the researchers focused on how alterations of the gene p53, which is commonly mutated in pancreatic cancer, influence the properties of Cancer-Associated Fibroblasts, or CAFs. CAFs are an important population of the tumor stroma which produce most of the matrix. 

Figure: Intravital (in vivo) imaging of pancreatic cancer cells (green) surrounded by a collagen-rich matrix (magenta) and blood vessels (red). Credit: Dr. Max Nobis. 

Using mouse models of pancreatic cancer, the authors isolated cancer cells and CAFs from both metastatic (spreading) and non-metastatic (non-spreading) pancreatic cancers, which are characterized by distinct mutations in p53. By mixing these different CAFs with cancer cells, the researchers discovered that cancer cells from a non-metastatic tumor start to spread when mixed with CAFs from a metastatic tumor. These results suggest that some aggressive pancreatic cancer cells, with high metastatic ability, can ‘educate’ the fibroblasts around the tumor. This in turn induces the fibroblasts to remodel the matrix and to interact with other, less aggressive cancer cells to support their ability to spread. This also means that in a growing tumor, where cancer cells have heterogenous abilities to spread, even a small number of aggressive metastatic cells can help increase the spread of other, less aggressive cancer cells.

The researchers then asked how the CAFs can induce the spread of cancer cells. Using mass spectrometry, the authors discovered several molecules that CAFs from metastatic tumors deposit in the stroma at higher levels than CAFs from non-metastatic tumors. The researchers next employed gene-editing techniques to interrogate how reducing the level of one of these molecules, called perlecan, would influence the fate of pancreatic cancer cells in mouse models of the disease. Remarkably, reducing perlecan in the tumor stroma significantly reduced the spread of pancreatic cancer cells. In addition, using advanced live imaging through optical windows, the authors tracked individual cancer cells and revealed that lowering the levels of perlecan not only reduced the spread of cancer cells, but that tumors also responded better to chemotherapy. This was confirmed by long-term experiments, where reducing perlecan in combination with chemotherapy significantly increased mouse survival rates in comparison with chemotherapy alone.

The study suggests that there are important benefits in targeting the aggressive CAFs of a tumor in combination with targeting the cancer cells themselves with chemotherapy. If targeting the aggressive fibroblasts in patients harboring precise genetic changes could be achieved, this would make them more susceptible to currently approved treatments, which would significantly change how we treat this aggressive cancer. In addition, targeting perlecan, or other matrix molecules that help remodel the tissue of metastatic tumours, may be effective for not just pancreatic cancer, but also prostate and breast cancers.


CAF hierarchy driven by pancreatic cancer cell p53-status creates a pro-metastatic and chemoresistant environment via perlecan. Claire Vennin, Pauline Mélénec, Romain Rouet, Max Nobis, Aurélie S. Cazet, Kendelle J. Murphy, David Herrmann, Daniel A. Reed, Morghan C. Lucas, Sean C. Warren, Zehra Elgundi, Mark Pinese, Gabriella Kalna, Daniel Roden, Monisha Samuel, Anaiis Zaratzian, Shane T. Grey, Andrew Da Silva, Wilfred Leung, Australian Pancreatic Genome Initiative (APGI), Suresh Mathivanan, Yingxiao Wang, Anthony W. Braithwaite, Daniel Christ, Ales Benda, Ashleigh Parkin, Phoebe A. Phillips, John M. Whitelock, Anthony J. Gill, Owen J. Sansom, David R. Croucher, Benjamin L. Parker, Marina Pajic, Jennifer P. Morton, Thomas R. Cox & Paul Timpson. Nature Communications, volume 10, Article number: 3637 (2019).

Media contacts

Guntram Bauer
Director Scientific Affairs and Communications

Rosalyn Huie