fb-pixel Skip to main content

Stem cells offer evidence on cancer’s growth

For a decade, it has been one of the most hotly debated questions in cancer: Do ­tumors return despite powerful treatments because they harbor a small number of stem cells that evade drugs?

If these rare, resilient cells are seeding new tumors, it would suggest a radical change in strategy for fighting cancer: drugs should primarily target these rare stem cells, rather than aiming to shrink a tumor. But research has produced conflicting results about the existence and importance of such cancer stem cells.

Now, two of the world’s top science journals are publishing evidence that should help settle the question. Three independent teams of researchers, working in mice with different types of cancers or precancerous tumors, have used new genetic techniques to find populations of stem cells that cause those tumors to grow.


In one case, researchers were able to show that treating glioblastoma, a deadly brain tumor, with a commonly given chemotherapy leaves behind those rare cells, which spark regrowth of the tumor.

“What these three papers have done, through elegant strategies, is demonstrate, indeed there are cancer stem cells,” said Robert Weinberg, a member of the Whitehead Institute for Biomedical Research in Cambridge and a biology professor at MIT who was not involved in the new studies. “It makes it more and more difficult for people to doubt the existence of cancer stem cells.”

Stem cells are better known for their ability to become any of the normal types of tissue and cells in the body, and they hold promise in a wide array of treatments, from regenerating spinal cord to curing diabetes. Cancer stem cells, on the other hand, highlight the malevolent side of such regenerative abilities — instead of growing healthy tissue, they cause tumors to emerge.

The new work does not answer all questions about stem cells or suggest that they are at the root of every cancer, researchers said. Rather, it solves many of the problems that led scientists to question earlier work.


Sean Morrison, director of the Children’s Medical Center Research Institute at the University of Texas Southwestern Medical Center, who has cast doubt on some claims about cancer stem cells in the past, called the new work an important step forward but said it remains unclear what portion of cancers are seeded by stem cells.

“We need to do a lot more experiments,” Morrison said.

He and others said the studies address a major flaw in past research, which typically used the same strategy to find evidence for cancer stem cells: grind up a human tumor, separate out cells, and put those into mice without an immune system to see which ones grow new tumors.

Although the technique identified stem cells in a wide array of cancers, critics pointed out that the system was artificial. Cells were likely to behave differently when they were stressed, separated from their normal surroundings, and implanted at different spots in a whole different animal. And separate laboratories did not always get the same results.

“I would say the cancer stem cell field has been controversial, because for every claim there is a disclaim,” said Luis Parada, chair of developmental biology at UT Southwestern Medical Center, senior author of one of the new papers.

The new research, published Wednesday in the journals Science and Nature, avoids such problems by having tumors develop in their natural environments, in mice. One team, led by Parada, used mice that are engineered to spontaneously develop glioblastoma. A team of Belgian and British researchers smeared a carcinogenic chemical on mouse skin to cause tumors to develop. A third team of Dutch researchers triggered a cancer-causing gene that caused a tumor to grow in the gut of mice.


Each team used a different set of experiments, but depended crucially on genetic techniques that allow scientists to label a specific subset of cells, marking them and the cells they give rise to. That allowed them to trace the family tree in the tumor, and successfully find a subset of stem cells in each case that drove the tumor’s growth.

Parada took an additional step, treating the mice with the drug commonly given to human patients with glioblastoma, called TMZ. The drug wiped out most of the cancer cells, but not the relatively rare stem cells, which regrew the tumor — a result that suggests how much cancer therapy might need to shift.

“If I were giving a therapy to my mice and saw the tumors shrank by 30, 40, 50, 60 percent, I’d be indifferent to that — the only thing that would matter to me is what is happening to the cancer stem cells in the tumor,” Parada said. “If your job is to prune a tree so it retains a certain shape, and you only prune dead branches, you might as well not be pruning at all.”


Locally, Cambridge-based Verastem, a 21-person company that went public earlier this year, is working on developing drugs that target this subset of cells. The company hopes to have three experimental drugs being tested in patients within a year. One of the early cancers its scientists hope to target is mesothelioma, which typically starts in the lungs.

Supporters of the idea of cancer stem cells say many unresolved issues remain to be solved. Weinberg and collaborators have discovered that non-stem cells appear to be able to morph into stem cells, suggesting that both populations of cells will need to be targeted by drugs.

Carolyn Y. Johnson can be reached at cjohnson@globe.com. Follow her on Twitter @carolynyjohnson.