The PDAC Key

At the same time, hundreds of miles away in Houston, a group of researchers from the MD Anderson Cancer Center, Baylor College of Medicine and Rice University were collaborating on a closely related piece of work. Using specialized mouse models of pancreatic cancer with impaired EMT, Raghu Kalluri and his colleagues were investigating the transition’s role in mediating metastasis and chemoresistance. The unexpected conclusion they reached mirrored the one from Dingcheng Gao’s group – namely, that pancreatic cancer, like breast tumors, can metastasize without undergoing EMT (1).

The group began by creating transgenic mouse models of pancreatic ductal adenocarcinoma (PDAC) in which either Snail or Twist, two of the transcription factors responsible for inducing EMT, were knocked out. Deleting the Snai1 and Twist1 genes had no effect on the development or appearance of pancreatic tumors, but the researchers noted a significant decrease in cells undergoing EMT. Immunolabeling of the primary tumor showed far fewer epithelial cells that expressed either αSMA (a mesenchymal marker indicating EMT-positive status) or Zeb1 (another EMT-inducing transcription factor similar to Snail), and global gene expression profiling revealed a decrease in the expression of EMT-associated genes. What was increased, on the other hand, was the degree to which cancer cells proliferated when the transition was suppressed. With no change in the timing of tumorigenesis and local invasion, it’s clear from these experiments that PDAC doesn’t rely on EMT to initiate and progress.

But metastasis is at the heart of the question. Do these cancers rely on EMT in order to spread to distant areas of a mouse’s – or a patient’s – body? The researchers compared circulating tumor cells in control and EMT-suppressed mice and found that the numbers were unchanged. Histopathology and immunostaining in livers, lungs and spleens (the major target organs of metastasis) revealed approximately the same frequency of cancer spreading in both groups – and, when examined more closely, the metastases all proliferated at about the same rate and were largely negative for EMT-inducing factors Twist, Snail, Zeb1 and αSMA. The take-home message? Removing EMT from the equation doesn’t affect the cells’ ability or inclination to metastasize.

So it appears that the Texas group’s pancreatic tumors behave much like the Cornell group’s breast cancers. Is the same true of the cells’ ability to survive chemotherapy? Previous studies have established a link between EMT and gemcitabine resistance in PDAC (2)(3)(4). “Gemcitabine works primarily on cancer cells that are dividing or proliferating. When cancer cells suspend their proliferation – such as when they launch an EMT program – then anti-proliferation drugs like gemcitabine do not target them well,” says Kalluri (5). The next step, then, was to test sensitivity to the drug in cells with suppressed EMT. The researchers administered gemcitabine to control, Snail- and Twist-knockout mice and discovered that, with the EMT-inducing factors removed, the chemotherapy-treated animals showed improved histopathology and survival. This held true across different mouse models of pancreatic cancer, all of which showed better responses to gemcitabine after EMT suppression – decreased tumor burden and proliferation, increased cancer cell death, and extended survival times.

“We found that EMT program suppressed drug transporter and concentrative proteins, which inadvertently protected these cancwer cells from anti-proliferative drugs such as gemcitabine,” says Kalluri. “The correlation of decreased survival of pancreatic cancer patients with an increased EMT program is likely due to their impaired capacity to respond to chemotherapy, leading to overall poor prognosis and higher incidence of metastasis.” (5)

Are there other possible explanations? The research still has gaps; it’s possible that other EMT-inducing transcription factors are replacing Snail and Twist in knockout mice, or that EMT suppression from birth (as in the mouse models) has a different effect to EMT suppression only at or after the onset of disease. It doesn’t look like the transition plays a significant role in PDAC metastasis – but in order to make that statement conclusively, more research, and probably more fierce debate amongst researchers, is needed.

But at the moment, the findings are fairly clear with respect to chemoresistance, and it seems clear that – by reducing proliferation and decreasing the expression of genes involved in transporting and concentrating drugs – the transition confers resistance to treatment and thus compromises patient survival. What does that mean for the clinic? Ultimately, that establishing a patient’s EMT status may provide insight into the potential for treatment – and that although treatments targeting the transition may not prevent metastasis, they offer a potential way of enhancing the effectiveness of existing therapies.

PROCEED WITH CAUTION

TRACKING THE TRANSITION

THE PDAC KEY

RESEARCH TIMELINE