A team of researchers has stalled tumors in mice by killing a small portion of cancer cells that seem to fuel malignant human melanoma. The report appears to be the first published data in support of the most alluring aspect of the increasingly popular cancer stem cell hypothesis: the prospect of identifying and eradicating the primary cancer cells that stoke and spread tumors, but often elude treatment.
“We showed in experimental models that specific targeting of a subset of cells is sufficient to inhibit tumor growth,” said senior author Markus Frank, HMS assistant professor of pediatrics at Children’s Hospital Boston. “We provided proof of principle that targeted ablation has potential therapeutic utility.” The study is published in the Jan. 17 Nature.
Stem Cell Defense
By the strictest definition, cancer stem cells are rare mutated cells that can renew themselves and also give rise to the heterogeneous mixture of cells that make up the bulk of tumors. Broad spectrum chemotherapy may eliminate the tumor mass, but it may spare the hardier stem cells, which survive to seed new tumors, according to the hypothesis. Unfortunately, cancer stem cells seem more resistant than the rest of the tumor to cancer drugs and radiation.
The cancer stem cell hypothesis contrasts with the prevailing paradigm of the last half century. The long-standing clonal evolution model says all cancer cells have equal potential to morph into more chemoresistant and invasive cells.
In the last decade, other scientists have identified cancer stem cells in blood cancers and several solid tumors. The new study from researchers based mostly at Children’s and Brigham and Women’s Hospital takes the evidence one step further in melanoma by targeting the cancer stem cell with an antibody that halts its growth.
“Even if this antibody doesn’t make it all the way into the clinic, what they’ve done is valuable,” said Michael Clarke, a researcher at Stanford University who has identified cancer stem cells in breast and colon cancer. “Identifying the cancer stem cells will enable people to dissect what’s going on at the molecular level and to identify other effective therapeutic targets.”
Scientists do not know the relationship, if any, between normal stem cells and cancer stem cells, but the intellectual path to the new paper began with studies of normal stem cells. Frank and his wife, Natasha, an HMS instructor in medicine at BWH, first discovered and characterized the latest known cell surface transporter, ABCB5, five years ago in experiments on the biology of the pigmented skin cells known as human epidermal melanocytes and found high co-expression with stem cell markers.
That led them to investigate adult stem cells for their potential to regenerate tissue or to downregulate immune responses to transplanted tissue. They discovered that the glycoprotein ABCB5 peppered the surface of progenitor melanocytes that had fused with differentiated cells, perhaps revealing a mechanism to regenerate tissue (see Focus Nov. 21, 2003).
Like other members of the large ABC family, the glycoprotein transports unwanted molecules out of cells. These cell surface transporters concern cancer biologists because their ability to pump toxins out of a cell can confer multidrug resistance to tumors. Two years later, the Franks and their colleagues reported that ABCB5 explains the resistance of human malignant melanoma to the cancer chemotherapy doxorubicin, which is known to be ineffective in treating melanoma. Blocking ABCB5 significantly enhanced the cytotoxic efficacy of doxorubicin in malignant melanoma cultures, they showed.
Testing Seeds of Disease
The most recent study, led by pharmacologist Tobias Schatton, a postdoctoral fellow in Frank’s lab, explored the cancer stem cell potential in cells marked by ABCB5. “Three main characteristics needed to be proven,” Schatton explained. “They are capable of self-renewal, or giving rise to more copies of themselves. They can differentiate into other cell types, including cancer cells that don’t form tumors in mice. And they must be tumorigenic, responsible for tumor growth and initiation.”
The key findings came from serially transplanted human melanoma cells in mice, conducted in collaboration with researchers in the BWH lab of Thomas Kupper, who have expertise in melanoma xenotransplantation models. Starting with metastatic tissue samples from four patients, the researchers sorted the tumor cells into two groups, ABCB5-positive and ABCB5-negative. The researchers transplanted each group of cells plus an unsegregated mixture into three groups of 23 mice.
ABCB5-positive cells generated tumors in 14 mice, including all mice injected with the highest dose. The mixed melanoma cells grew into cancers in seven mice. Only one mouse injected with ABCB5-negative tumor cells developed cancer. The researchers provide evidence that positive cells among the highest negative dose given might explain the tumorigenicity in this mouse.
Taking the tumors generated by positive cells in mice, the researchers again separated the cells and transplanted them into 18 more mice each. Ten of the positive cells generated secondary tumors, compared to none of the ABCB5-negative.
The researchers tracked the lineage of dividing cancer cells in mice using genetic tags that glowed red or green under the microscope. These colors showed that ABCB5-positive cells gave rise to vigorous growth of both positive and negative cells and that ABCB5-negative cells gave rise only to negative cells that did not grow well.
“The next steps are to look into pathways that are operative in ABCB5-positive cells to find out how they mediate enhanced tumorigenicity,” Schatton said. “We have an efflux pump on these cancer stem cells, but that alone doesn’t explain why they drive tumor growth and why there is an enhanced abundance of melanoma stem cells in more advanced disease.”
Years earlier, the Franks had devised an antibody to block ABCB5. Now they used it to kill the cancer stem cell it decorated. “Under the microscope, we could see the results,” said George Murphy, BWH dermatopathology chief and HMS professor of pathology. “The antibody localized to the stem cell component of the tumors and elicited an immune response by the mouse directed against the antibody coating.” Murphy used a series of molecular probes to locate the stem cells, track the antibody, identify the immune cells that swarmed in to kill the antibody-tagged cell, and correlate the sequence of events with tumor growth.
“This answers the question people have been asking for a number of years: if we target cancer stem cells, do we stop or reverse the cancer’s growth?” said Kupper, the Thomas B. Fitzpatrick professor of dermatology at BWH and HMS. “I don’t think anyone expects an antibody to cancer stem cells will work as a single therapy to kill all melanomas, but as part of a rational combination approach, it may be uniquely effective. Chemotherapy or immunotherapy plus an antibody directed at cancer stem cells may give a one–two punch that cancer can’t recover from.” Kupper heads the National Institutes of Health grant to BWH and Harvard that funded the study.
It is too soon to know if this approach will be effective in people, but researchers are encouraged by the results in melanoma and the implications for other cancers. Frank’s earlier paper also found ABCB5 to a lesser degree in malignant breast cancer, lung, colon, stomach, and renal tissues.
“I would not try to induce false hopes in patients currently suffering from melanoma,” he said. “I wouldn’t say we have the actual drug in hand. This demonstrates proof of principle in vivo, in experimental model systems using human xenografted tumors, that we can halt tumor growth by targeting a prospective marker of cancer stem cells.” The findings provide a rationale for clinical drug development and stir up many additional questions in the basic biology, including ones with further potential relevance for cancer therapy.