Posted on: 03/11/2004

Stem Cells in Tumors May Help Explain Some Cancer Mysteries

If anyone still has doubts that the 1950s are a bygone era, consider some experiments scientists conducted then without raising eyebrows (let alone lawsuits).

To probe the basics of cancer, a team of scientists injected cells from patients' colon, lung, ovarian and other tumors into the patients' own thighs. After trying this with varying numbers of cells, the scientists found that the transplants regenerated a tumor in the thigh (or in a mouse) only if at least one million cells were injected.

If the bioethics of the 1950s fell short of today's, the science was a harbinger of an upheaval now shaking cancer biology. After going nowhere for decades, those old results have finally led to a revolutionary insight.

"Within a tumor mass, there is only a small population of cells that can spawn more tumor; other cells in the tumor can't," says cancer biologist Robert Weinberg of the Whitehead Institute, Cambridge, Mass., who discovered the first human oncogene.

The existence of these "tumor stem cells" promises to explain one of the more perplexing, and tragic, mysteries of cancer treatment. Biologists have made great strides in identifying molecular pathways by which cancer cells grow and spread. But although Iressa shrinks non-small-cell lung cancers, Erbitux shrinks advanced colorectal cancers and the just-approved Avastin shrinks metastatic colorectal cancer -- in each case, in only some patients -- the drugs prolong survival by mere months or not at all.

The reason, says Dr. Weinberg, may be that "killing off the majority of cells in a tumor will still leave it with the ability to regenerate another tumor, from these stem cells."

Eradicating the non-stem cells in a tumor "may result in a remission" and even the appearance of being cancer-free, explains molecular biologist John Dick of the University of Toronto, but "the disease will relapse if the tumor-initiating cells are not eliminated."

Discovering that only some cancer cells are able to generate more tumor has been a struggle, as the lag between those 1950s experiments and today shows.

"The idea of cancer stem cells has been around for a long time," says biologist Irving Weissman of Stanford University. But no stem cell (normal or cancerous) had been isolated until he and his colleagues identified blood-forming stem cells in mice in 1987, and in humans in 1992. Seizing on this method for differentiating stem cells from others, Dr. Dick and his team discovered in 1994 that, in leukemia, only some cells have the ability to spawn more tumor. The Toronto team dubbed them "leukemic stem cells."

It was an uphill struggle to figure out whether blood cancers are unique, or whether solid tumors have cancer stem cells, too. But after begging the university for a crucial $500,000 machine, biologists led by Michael Clarke and Muhammad Al-Hajj of the University of Michigan, Ann Arbor, found last year that breast cancer consists of a few cancer-initiating cells that can make more breast-cancer cells, seemingly forever, surrounded by an ocean of noncancer-initiating cells.

Although the two kinds of cells looked identical, on closer inspection there were several "markers" -- think of them as microscopic flags poking up from the cell's surface -- that distinguish a breast-cancer stem cell from the cells in a breast tumor that can't generate more tumor. When the Michigan team injected as few as 100 cells of the former into mice, breast tumors grew every time. Not even tens of thousands of noncancer stem cells produced a tumor. Since the Michigan discovery, researchers in Japan and Canada have found brain-cancer stem cells; stem cells likely exist in other solid tumors, too.

Dr. Clarke estimates that breast cancer stem cells make up as little as 3% to 5% of some tumors. That's the good news. Those are the only cells you have to kill to cure cancer, because non-stem cells eventually die off on their own. Even when non-stem cells spread, they don't pose much danger because they die after dividing a few times. The goal of current chemotherapy, to kill as many cells as possible, can probably be dialed back.

The bad news is that standard chemo hits tumor cells at a vulnerable point in their life cycle, but cancer stem cells don't seem to cycle this way. "I think the reason cancer therapy does not cure all cancers has to do with the unique properties" of the cancer stem cell, says Dr. Dick.

Ever more sobering, the molecular pathways that scientists have recently identified, with much fanfare, as allowing cancer cells to grow may not be critical after all. If such a pathway promotes growth in the zillions of cells that are not cancer stem cells, it may be inconsequential. The only pathways that matter are those that keep the miscreants alive and thriving. Or as Dr. Weinberg told me, "tumor stem cells may explain why you can have tumor shrinkage but not life extension. If current chemotherapies don't target tumor stem cells, the cells keep making more tumor."

On the bright side, knowing which cells they have to kill gives scientists a better shot at finding therapies that do that. "I'm optimistic we can figure this out," says Dr. Clarke.


Click HERE to return to brain tumor news headlines.