In the decades since President Nixon's 1971 declaration of war on cancer, scientists have made great progress in their battle against many cancers, such as childhood leukemia, testicular cancer and Hodgkin's disease.
But progress against most other types of cancer has been less dramatic, according to National Cancer Institute statistics. The percentage of women with late-stage breast cancer who die within five years of diagnosis is still about the same -- nearly 70%. And the story is similar with other deadly cancers, such as those of the lung, prostate and colon. Better screening and prevention have reduced the risk of dying from them, but once they take firm hold, the most aggressive therapy often fails.
Researchers believe they now understand why this happens. In focusing their efforts on trying to shrink tumors, cancer treatments may be missing a vital target.
Mounting evidence points to a handful of special cells within a tumor as the true culprits that trigger the disease and cause it to recur and spread. These so-called cancer stem cells seem to act like the more-familiar adult or embryonic stem cells in their ability to renew themselves while churning out cells of other types. What's worse, they may be more resistant than the bulk of the tumor cells to traditional cancer therapy.
With that growing knowledge comes a new approach to fighting cancer. Scientists hope that if they redesign drugs to specifically kill or disable the cancer stem cells, they could stop the cancer for good. Shrinking the tumor could come later.
This insight has triggered a race to develop a new generation of cancer drugs, some of which are already in human trials. "We are on the verge of a new era in cancer medicine where we target not the bulk of the tumor but its seeds," says Dr. Owen Witte, head of the Eli and Edythe Broad Center of Regenerative Medicine and Stem Cell Research at UCLA. "We can then definitively treat the tumor instead of continuously suppressing its growth."
The idea that cancer originates from malignant stem cells has been around since the 19th century, when scientists observed that tumor tissue resembled embryo tissue under a microscope. Concrete evidence had to wait for recent advances in biotechnology such as ways to rapidly scan the activity of thousands of genes and techniques for breeding rodents with specific gene mutations.
Armed with these modern tools, in 1994 a research team led by John Dick, a senior scientist at the Toronto General Research Institute in Ontario, studied the effects of injecting human leukemia cells into mice lacking an immune system. "We found that only one cell in a million had the ability to initiate leukemia," Dick says.
Cancer stem cells rare
These rare, cancer-initiating cells displayed special molecules, or markers, on their surface that distinguished them from other leukemia cells. Furthermore, they were similar in certain ways to normal stem cells found in the body. "These leukemic stem cells are a caricature of normal development," Dick says. They resemble stem cells but give rise to tumor cells instead of healthy tissue.
Researchers have now implicated stem-like cells in many other types of cancer. In 2003, a team of researchers at the University of Michigan at Ann Arbor found that only about 1% to 10% of breast cancer cells had the ability to form new tumors. When injected into a mouse, as few as 200 of these distinctive cells were enough to form a tumor. Yet even 20,000 of the other tumor cells failed to form one when injected.
"Only a small fraction of the cells in a tumor drive the cancer," says Dr. Max Wicha, professor of internal medicine and a member of the research team. "The rest are dead-end cells."
A similar pattern has been found in brain, prostate and colon cancers, as well as other solid tumors. Wicha suspects that cancer stem cells in different tumor types may turn out to have common features. "A treatment that works against one type of cancer stem cell may work against other types as well," he says.
If a single, universal malignant stem cell type exists, it would be an attractive target for cancer therapy. But some researchers say that care is needed in interpreting the findings, which are mostly based on injecting human tumor cells into mice and other animals. Dr. Richard Hill, professor of medical biophysics at the University of Toronto, says this method might not be identifying cancer stem cells, but be identifying human cells that are able, for some reason, to survive in the alien mouse environment.
Hill also thinks it may be hard to pinpoint and eradicate the subpopulation of cells responsible for perpetuating cancer in humans. "There may be many such cell types," he says. "Targeting them is going to be more problematic than we think."