Understanding the cancer stem cell hypothesis

Dr. John Dick (CPimages/S. Lake)

The idea that cancer is caused by certain cells with properties similar to stem cells is being debated with renewed vigour, following the publication of several studies suggesting that early predictions about progenitor cells in some cancers need to be re-assessed in light of new evidence.

Research has shown that progenitor cells commonly referred to as “cancer stem cells” are a subset of cancer cells that are responsible for initiating and sustaining the growth of cancer. Current treatment approaches often successfully destroy the bulk of the cells in a tumour, causing the cancer to enter remission. However, cancer stem cells may be resistant to most current forms of cancer therapies, providing a possible explanation for the high rate of relapse for many types of cancer.

When it was first put forth in 1994, this hypothesis was a departure from what scientists previously thought about cancer. For many decades, the scientific community accepted the “stochastic” model of cancer development, which views cancer as a disease brought on by normal cells that mutate and then divide. In this model, tumours are homogeneous, with no hierarchy among cancer cells.

Some scientists suspected that some or all cancers may actually be heterogenous, consisting of cancer stem cells that are specialized to create normal cancer cells. These normal cancer cells are what make up the bulk of tumours and cancers of the blood. However, this idea was not generally accepted until Dr. John Dick, a Senior Scientist at the Ontario Cancer Institute, the research arm of the University Health Network’s Princess Margaret Hospital in Toronto, showed in the mid-1990s that certain cells are indeed cancer stem cells.

In the cancers Dick studied, these cells appeared to be very rare. When human cancer cells were transplanted into immuno-compromised mice, only 0.1 to 0.0001 gave rise to new tumours. In other words, cancer stem cells only appeared to account for a very tiny fraction of the cells in these types of cancer. If cancer stem cells are very rare in all cancers, it could open the door for new treatments that selectively target the relatively small population of cancer stem cells. If these cells are destroyed, then it might be possible to stop cancer at its root.

However, recent studies suggest that in some types of cancer, cancer stem cells are not as rare as originally predicted. Research led by Sean Morrison, a scientist at the University of Michigan, showed that in certain types of melanoma, one-quarter of cells could give rise to new tumours. Morrison used a different technique than Dick and his colleagues to transplant the cells into mice. In essence, he suggests that in earlier studies, some progenitor cells may have been prevented from giving rise to tumours because of the techniques that researchers use to take the cells from human tumours and transplant them into mice.

As in any debate, there are different ways to interpret the hard data.

Some commentators have argued that the high frequency of cancer stem cells in the tumours studied by Morrison’s group suggests that the cancer stem cell hypothesis is invalid. Their argument is that the apparent rarity of cancer stem cells could actually be an artefact on researchers’ inability of researchers to transplant human cells into mice. Carried far enough, this argument could even suggest cancer stem cells are actually just normal cancer cells.

However, this is not the view that most scientists have taken – including Morrison. He points out that research by Dick and others has definitively proven that cancer cells are heterogeneous, and that in some cancers such as leukemia, the evidence that certain cells have a progenitor role is very convincing.

"There are some people out there who look at the paper and say that this completely invalidates the cancer stem-cell field," he told Nature in December 2008. "That's an over-interpretation." His message is that researchers who report cancer stem cells in solid tumours should use techniques that could help human cells survive in mice when validating their experiments to avoid under-counting.

Meanwhile, Dick cautions that cancer stem cells are still a new field in need of further investigation. His original studies in the 1990s, which proved the existence of cancer stem cells in acute myeloid leukemia, did not use the methods that have been recently criticized for under-counting cancer stem cells – suggesting that in some cancers, cancer stem cells are indeed rare.

In a response to research by Morrison and others, Dick, along with colleagues Jean Wang and Catherine O’Brien, also notes that the cancer stem cell model never predicted that cancer stem cells would be rare – it only predicts that there is heterogeneity within tumours.

“The critical test is whether cancer-initiating cells can be identified and prospectively separated from non-tumorigenic cells — an affirmative answer negates the stochastic model,” they write in a commentary published on the Nature website. Although some early assumptions about cancer stem cells may prove to be incorrect, the evidence still supports the thrust of the cancer stem cell hypothesis: that certain cells, with stem cell-like properties, are responsible for the growth of other cancer cells.

Where to go from here?

The debate over the cancer stem cell hypothesis has led to another debate in the scientific community.

Some scientists argue that evidence supporting the entire cancer stem cell hypothesis is not strong enough to justify devoting extensive resources to answering questions about cancer stem cells. Others argue that findings in support of the hypothesis and the tremendous benefits that research in this area could bring to patients justifies further inquiry.

Many scientists in Ontario take the latter view, and are pursuing research that will help refine the cancer stem cell hypothesis and develop therapies that selectively target progenitor cell populations.

“As Dr. Dick has pointed out, there is still very solid evidence that cells in at least some cancers are heterogeneous – in other words, that all cancer cells are not created equal,” says Dr. Bob Phillips, Deputy Director of the Ontario Institute for Cancer Research. “Furthermore, there are many reasons to believe that for some cancers, therapies that target cancer stem cells could be far more effective than current therapies. OICR’s investment in cancer stem cells research will accelerate our understanding of cancer stem cells and help translate our knowledge into new treatments.”