All too often, cancer that seems to have been wiped out by treatment comes back. Some scientists have blamed so-called cancer stem cells, a subset of cancer cells that might be able to remain dormant, evading chemotherapy or radiation treatments, only to form new tumors months or years later. The idea has been controversial, but three papers published today report evidence that in certain brain, skin, and intestinal tumors, cancer stem cells are the source of tumor growth.
The cancer stem cell model differs from the traditional idea that tumor growth is equal opportunity—that is, any and all cancerous cells can divide and cause the tumor to grow and spread. The stem cell model says that tumor growth is more hierarchical, mainly driven by a subset of cells that can make new copies of themselves and give rise to the other cell types the tumor contains. Some of the first evidence for cancer stem cells came from studies of leukemia in the 1990s, which showed that only a small subset of the cancerous blood cells could propagate the disease in mice. But it has been harder to test whether cancer stem cells fuel the growth of tumors in other tissues.
In the new studies, three independent groups used genetic cell-marking techniques to trace the proliferation of certain cells within growing tumors. The method gives researchers a glimpse of "what happens in the real life of a tumor," says Cédric Blanpain, a stem cell researcher at the Université Libre de Bruxelles in Belgium. He and his colleagues report online in Nature that in mouse papilloma tumors, a precursor to skin cancer, most of the tumor growth came from a few cells , which in some ways resembled the stem cells that maintain healthy skin.
In a second paper, also published online today in Nature, developmental biologist Luis Parada and his colleagues at the University of Texas Southwestern Medical Center (UTSMC) in Dallas show that in mice that develop glioma, a form of brain cancer, tumor growth seems to come from a small subset of cells in the tumor . They find that the cells can remain dormant during chemotherapy that kills off most of the cancer and can give rise to new tumors once the drug treatment stops.
And in the third paper, published online today in Science, developmental biologists and stem cell researchers Hugo Snippert, Arnout Schepers, Hans Clevers, and their colleagues at the Hubrecht Institute in Utrecht, the Netherlands, used mice with multicolored intestines to look at the kinds of cells that form intestinal adenomas, a precursor to intestinal cancer. The rodents, which the scientists have nicknamed confetti mice, carry genetic markers that can label intestinal cells blue, green, red, or yellow depending on which cell they originate from. The team reports that the adenomas grow from cells that express a gene called Lgr5+, which is also active in normal intestinal stem cells . "The tumor is really like a caricature of normal tissue," Snippert says.
Such cell-tracing techniques are the right approach to test the cancer stem cell model, says Sean Morrison, who studies stem cells and cancer at UTSMC and who was not involved in any of the studies. There is now enough evidence to be fairly sure that the model explains at least some types of cancer, he says. Morrison cautions, however, that the studies on papilloma and adenoma looked at precancerous tumors. Indeed, when Blanpain and his colleagues examined mice with squamous cell carcinoma, a malignant outgrowth of the papilloma, they found that most of the cells were actively dividing, not just a small subset of stem-cell-like cells.
Understanding which cancers might grow from—or simply harbor—cancer stem cells is key to more effective treatments, the researchers say. That is no easy task, however. Morrison notes that tumor growth differs even among patients with the same type of cancer. Still, says Parada, having three examples in which tumors seem to harbor cancer stem cells suggests there will be more. "I hope it will bolster and stimulate the community" to figure out how to better study the cancer stem cell model, he says. "Let's bring this level of scrutiny to all solid tumors."