About 5 years ago, researchers discovered that they could create what they called cancer stem cells by forcing mammary epithelial cells through a developmental program typically associated with embryogenesis. Robert Weinberg, PhD, director of the Ludwig Center for Molecular Oncology at the Massachusetts Institute of Technology in Cambridge, Massachusetts, describes the research as the most important result to come out of his lab in the past 20 years.

Dr. Weinberg, also a member of MIT's Whitehead Institute for Biomedical Research, says the interesting and unexpected finding faced such resistance from a top journal that the editors refused to even send the manuscript out for review. Eventually, the study appeared in Cell in 2008 and bolstered earlier research raising the specter of cancer stem cells.1

Since then, the pace has quickened noticeably in research to identify subsets of cancer cells with the properties of quiescent, nondividing stem cells that can evade conventional chemotherapies. Last year, major articles suggested these stem cells may drive the growth and persistence of skin, brain, and intestinal tumors. The accumulating data are raising an important, if controversial, question: is chemotherapy targeting the right cell population? And, if not, how can the true source of cancer renewal be exposed and wiped out?.

Not everyone considers the existence of cancer stem cells to be a settled matter, and the research is still controversial enough that one well-respected scientist declined to speak on the record with “CytoSource” to avoid risking a public spat with critics.

Dr. Weinberg, however, says the preponderance of evidence is making such skepticism harder to justify without countervailing observations. “I used to be apologetic or tentative about using the term “cancer stem cell.” I no longer feel that way,” he says. Even so, he cautions that researchers must be rigorous in providing evidence of the cells' existence. The most critical experiment, he says, is to implant putative cancer stem cells in a mouse model and observe whether a cancer outgrowth ensues.

In one such experiment in mice, researchers identified a subpopulation of cells that fuels the growth of established intestinal adenomas, a result they claim provides “direct, functional evidence for the presence of stem cell activity” within the precursor to intestinal cancer.2 A second study similarly found a persistent population with stem cell-like characteristics in squamous skin cell tumors.3 In a third article published concurrently, researchers suggested that recurrence of the malignant brain tumor glioblastoma multiforme in mice likewise was driven by a reservoir of quiescent cells. The putative stem cells, the authors maintain, appeared to reseed the tumor after treatment with the chemotherapy drug temozolomide.4 In combination with the drug ganciclovir, an antiviral agent more commonly associated with treating patients who have cytomegalovirus, tumor development was significantly impeded.

Ganciclovir, the scientists concluded, likely goaded the nondividing stem cells into becoming active again, and thus more susceptible to chemotherapy. “I think that cancer stem cells have a heightened resistance to traditional therapies,” Dr. Weinberg says. “How one gets a clinical response is not totally clear, but I think it will be so in the future.”

A Question of Malignant Behavior

  1. Top of page
  2. A Question of Malignant Behavior

Connie Eaves, PhD, a distinguished scientist at the Terry Fox Laboratory at the British Columbia Cancer Agency in Vancouver, Canada, says the current controversy over cancer stem cells has arisen from a polarized and mostly crude understanding of how malignant cell populations behave. “It's sort of like how physics was about 100 years ago, when particle theory and wave theory butted heads,” Dr. Eaves says. Unlike cancer research, she says, the field of physics has since been able to accommodate the chameleon-like nature of complex scientific phenomena that can diverge wildly depending on their context.

As cancer researchers have learned, what is considered perfectly normal in one stage of development or in one cell type, such as rapid cell division or a blastocyst burrowing into the uterus, can be considered a malignant feature in a different place or time. What scientists are still learning is that a genetic mutation is not necessarily required for that shift in context; altered regulation of a gene's expression due to epigenetic changes often does the trick. As such, the mechanisms dictating a cell's fate are not set in stone. “They're reversible, and actually they don't require enormous manipulations of the cell,” Dr. Eaves says. “It's actually pretty trivial if you get the right things in the cell to cause these changes.”

With cancer stem cells, Dr. Eaves says, the idea is that a cancer can arise from a small subset of long-lived cells within a particular tissue that eventually gain dominance by acquiring the ability to initiate a runaway assembly line. Mutant daughter cells can quickly roll off that assembly line and some may develop into cancer cells. Because many of the genes involved in determining a cell's fate are the same genes involved in malignant transformation, Dr. Eaves says, it does not take much of a stretch to imagine that the process by which cells become cancerous through epigenetic changes might be reversible as well. The big question, of course, is how to engineer that reversal.

It does not help that most methods used to evaluate a cancer intervention often include groups of patients in very diverse, latestage malignancies. Also, scientists have only partially characterized the individual cells and populations that may give rise to cancer. “In the meantime, you can't just tell all of the patients in the world to go into a cupboard while we figure all of this out,” Dr. Eaves says.

Despite the remaining gaps in knowledge, researchers are pressing ahead with 2 main strategies to attack putative cancer stem cells. Some groups are seeking agents that are directly cytotoxic, a scheme that Dr. Weinberg says has been bolstered by work in his own laboratory suggesting that some compounds will preferentially target cancer stem cells. The finding has given him some initial encouragement that the tactic is not merely a “pie in the sky” idea. Similar to the recent glioblastoma study in mice, other efforts are aimed at forcing the stem cells to differentiate, thereby making them more vulnerable to chemotherapy. Dr. Weinberg cautions that it is still too early to say which cancers may respond best to the combination therapy strategy. A retrospective analysis of more than 150 cancer patients treated at the University of Texas MD Anderson Cancer Center in Houston and the University of Washington in Seattle, however, has offered another glimpse into how such a tactic might work.

When treated with the chemotherapy drug capecitabine and the anti-inflammatory arthritis medication celecoxib, researchers found that patients with stage IV colorectal cancer survived a median of nearly 94 months, more than twice as long as similar patients on more conventional treatments. The tentative conclusion, yet to be confirmed in animal studies, is that celecoxib can target the relatively dormant, nondividing, stem cell-like cells and somehow reactivate them, making them newly susceptible to capecitabine.

“We noticed this phenomenon in the clinic and we've been instituting the protocol for patients for some time,” says Edward Lin, MD, a principal investigator in the study and a medical oncologist at the University of Washington and Fred Hutchinson Cancer Research Center, also in Seattle. He adds that the therapy seems to be the only effective strategy for increasing the proportion of very long-term survivors.

Dr. Lin refers to the normally inactive cells as “army reserve cells” that can be called up and pressed into service by a tumor when needed. So far, he says, preliminary research is consistent with the idea that the reservists are actually cancer stem cells that resist the chemotherapy, and that celecoxib may selectively target the stem cell pathway within the tumors.

He and his collaborators are now trying to continue the story in mice by reverse-engineering the observation into a research strategy aimed at activating and then draining the pool of putative cancer stem cells. They expect to present their experimental results in mice at the Digestive Disease Week conference in May, and Dr. Lin says he's optimistic that follow-up studies will confirm his group's initial observations. “We're peering into the window of what's happening in tumors,” he says.

Cytosource Reader Poll #10:

Cancer Stem Cells

Q: Which therapeutic strategy seems the most promising?

A. Directly targeting cancer stem cells with cytotoxic agents.

B. Forcing cancer stem cells to differentiate, thereby making them more vulnerable to chemotherapy.

C. Another strategy targeting the stem cell pathway that has yet to be discovered.

D. None of the above; cancer stem cells don't exist.

Take the poll online at The results will be published in the June 2013 issue.


Q: What is the biggest obstacle facing whole genome sequencing's bid to become a viable clinical service?

71% Upfront costs and insurance reimbursement.

14% Bioinformatics and interpretation logistics.

0% Demonstration of clinical usefulness.

14% Ethical and legal challenges.


  1. Top of page
  2. A Question of Malignant Behavior