Researchers identify human pancreatic cancer stem cells

Feb 1 2007

Researchers at the University of Michigan Medical Center have, for the first time, identified human pancreatic cancer stem cells. Their work indicates that these cells are likely responsible for the aggressive tumor growth, progression, and metastasis that define this deadly cancer.

In the February 1 issue of Cancer Research, the researchers demonstrate that only 100 of these stem cells are needed to produce human pancreatic cancer in half of mice tested. They also found these cells are at least 100 times more tumorigenic than cancer cells that did not have one of three protein markers they believe to be associated with pancreatic cancer stem cells.

The findings could help advance development of new therapies for this cancer, which has a five-year survival rate of only three percent - the worst prognosis of any major cancer, said the study's lead author, Diane M. Simeone, M.D., an associate professor of surgery and molecular and integrative physiology.

"The cells we isolated are quite different from 99 percent of the millions of other cells in a human pancreatic tumor, and we think that, based on some preliminary research, standard treatments like chemotherapy and radiation may not be touching these cells," said Simeone. "If that is why pancreatic cancer is so hard to treat, a new approach might be to design a drug that specifically targets pancreatic cancer stem cells without interfering with normal stem cell function."

While such a drug has not been developed, ongoing research suggests it is possible to do so, she added.

The study also advances the emerging notion that stem cells may lie at the heart of some, if not all, cancers, Simeone said. That theory suggests that only cells that have the properties of "stemness"− that is, cells that can self-renew and differentiate into other types of cells − are the only ones capable of producing tumors. These "cancer stem cells," could derive from normal adult stem cells in organs that have mutated, or from a differentiated cell that has devolved to take on the qualities of stem cells. They are resistant to traditional therapy designed for cells that rapidly turn over because stem cells don't, according to some researchers. Thus, they remain after tumors shrink and may be responsible for cancer recurrence and metastasis.

This study confirms at least one of those concepts, the researchers said. "Our study demonstrates that the very small subset of cells in a human pancreatic tumor that cause the cancer to grow and propagate have stem cell-like features," Simeone said.

To look for cancer stem cells in pancreatic cancer, the research team implanted cancerous tissue from human pancreatic specimens removed during patient surgery in "xenograft" mice with compromised immune systems. Researchers removed tumors after they grew, and then sorted millions of cancer cells to isolate those that had one or more of three surface protein markers − CD44, CD24, and ESA. They chose these markers, called cell adhesion molecules, because they'd recently been found on isolated breast cancer cells by study co-authors Max Wicha, M.D., from Michigan and Michael Clarke, Ph.D., from Stanford University School of Medicine.

The researchers then implanted about 100 of each type of cell in mice, and found that tumors would grow in a subset of the animals, but cells that expressed all three markers were the most potent, producing tumors in six of 12 mice tested. If more cells are used, "we can get tumors to grow 100 percent of the time," Simeone said. "These cells are highly tumorigenic, which reflects the biology of this cancer."

Additionally, the tumors derived from the highly tumorigenic pancreatic cancer cells "appeared remarkably similar to the appearance of tumors taken directly from patients," Simeone said. The purported cancer stem cells produced a diverse mixture of cells, some of which are not cancerous, that reflected an actual human pancreatic tumor, she said.

The markers that define the highly aggressive pancreatic cancer subtype are not identically matched to those found in aggressive breast cancer, the researchers also discovered. The difference is in one marker, CD24, which is negative in breast cancer and positive in pancreatic cancer, according to Simeone.

"These studies suggest that several stem cell markers may be shared by cancer stem cells in different tumor types, such as CD44 and CD133, but it is possible that each tumor cancer stem cell has its own set of unique markers," Simeone said.

The study was funded by the Lustgarten Foundation, the Els Pardee Foundation, the Michgan Life Sciences Corridor, and the American Diabetes Association.