Leukemia researchers strive for a cure

Why is cancer so hard to treat and why does it return even after all visible signs are gone? Some researchers trying to answer these questions believe that cancer stem cells may be “the roots” that feed at least some cancers.

Cancer researchers have several competing visions of tumors. In one vision, all tumor cells are pretty much the same, or closely related, and have an equal capacity to divide and form new tumors. In a second vision, only a few cells have the capacity to initiate new, full-fledged tumors. These “bad seeds” are “cancer stem cells.” Some researchers would probably add a third vision, a “clonal vision,” where groups of cells descend from a common clone, and the cells that make up a clonal group all behave similarly, but different clonal groups may behave differently.

What are stem cells, and why are they so important? There are several different types of stem cells in adults, including hematopoietic stem cells, mesenchymal stem cells, neural stem cells, epithelial stem cells and skin stem cells. All stem cells — regardless of their source — have three general properties: they are capable of dividing and renewing themselves for long periods; they are unspecialized; and they can give rise to specialized cell types — for instance, hematopoietic stem cells are capable of producing all of the different types of blood cells. When needed in adults, stem cells are able to re-supply the body with many different types of tissues.

Finally, stem cells often exist in a quiescent state, meaning that the cells are not dividing. Quiescent, nonproliferating cells are insensitive to traditional chemotherapy that kills fast-growing cells. Some believe that these quiescent cells may be resistant to Gleevec as well.

Some leukemias, including chronic myelogenous leukemia (CML), acute myelogenous leukemia (AML) and some solid tumors, including brain cancers and breast cancers, have been shown to originate from cancer stem cells. The March 2004 issue of Science News magazine reported on the work of John E. Dick from the University of Toronto. A decade ago, Dick led a research team that showed that only some cancer cells from leukemia patients could reproduce leukemia in rodents. Science News also cited the work of Michael F. Clarke of the University of Michigan Medical School, who reported similar results for breast cancer cells. Clarke found that perhaps one in 100 breast cancer cells forms tumors when implanted into mice. In 2003, two research teams presented evidence that cancer stem cells underline brain tumors as well. “I think the cancer stem-cell hypothesis will apply to every kind of cancer,” Dick told Science News.

Researchers studying CML and AML are finding ways to target their respective stem cells. Craig T. Jordan, Ph.D., of the University of Rochester, and Monica L. Guzman, Ph.D., have suggested that in AML conventional chemotherapy kills the leukemic blast cells (the progeny of the leukemic stem cells), but does not kill the leukemic stem cells (the parent cells). While this provides initial control of the disease, patients relapse as the leukemic blast cells.

Since these leukemia blast cells greatly outnumber the stem cells, response to treatment is almost always measured by a drug’s effect on the blast cells. Thus, while treatments often reduce the bulk of disease, they often fail to target the rare stem cells that many researchers feel must be killed to prevent recurrence of the disease. It is very difficult to assess the effect of treatment on these stem cells because they are so rare.

Targeting leukemia stem cells

Monica L. Guzman, Ph.D., and Craig T. Jordan, Ph.D. Reproduced with permission. The leukemic stem cell (LSC) model at left proposes that leukemic blasts originate from a common primitive progenitor that has the capacity to self-renew. Conventional therapy regimens for leukemia, right, have been designed to eliminate leukemic blasts. These regimens may not effectively ablate the LSC population, which eventually regenerates the disease. The ability to design therapies that can target LSCs should yield more effective eradication of the disease.

Guzman, Jordan and others have suggested that in addition to targeting the bulk of the disease, stem cells must be targeted to prevent relapse. They have found that leukemic stem cells in AML are dependent on a survival protein, NFkappaB, while normal hematopoietic stem cells are not. They devised a strategy where they stress these cells with idarubicin, a traditional chemotherapy, while at the same time inhibiting the survival protein, NFkappaB, by adding a “proteasome” inhibitor, MG-132. This strategy of stressing the cell and at the same time inhibiting an important survival protein worked well in the lab.

CML and GIST have so far proven remarkably similar in their biology, treatment with drugs, and resistance mechanisms. If you want to know what’s going to happen in the GIST world, sometimes you can look at what’s happening in the CML world. The existence of a primitive quiescent stem cell population that is resistant to Gleevec has been detected in CML. It is speculated that failure to eliminate this stem cell population may be why Gleevec eventually fails some patients.

A recent paper, “Punish the Parent, Not the Progeny” by Lucy J. Elrick, Heather G. Jorgensen, Joanne C. Mountford, and Tessa L. Holyoake, from the University of Glasgow, extends the cancer stem cell theory to CML. These researchers noted that a small population of quiescent leukemic cells exists in CML patients and this population cannot be eliminated with Gleevec. These cells remain after Gleevec therapy, even when apparently complete responses are achieved, and probably explain molecular disease persistence.

“The emergence of drug resistance with imatinib (Gleevec) monotherapy also argues in favor of complete disease eradication that we believe should remain the ultimate therapeutic goal in CML,” noted the authors. New approaches to the elimination of these primitive CML cells may thus be crucial to the development of curative strategies.” Several theories have been proposed to explain why Gleevec doesn’t kill these cells: — A greater role for multi-drug resistance proteins. — The quiescent state of the cells — Pre-existing kinase mutations. — Unknown mechanisms?

The Glasgow researchers, led by Dr. Tessa Holyoake, seem to get a little more speculative on their Web site (as opposed to their paper) about why the quiescent CML stem cell population might be insensitive to Gleevec. “Indeed, we have demonstrated that, in vitro, quiescent CML stem cells are completely insensitive to imatinib at concentrations up to 10-fold higher (10mM) than those achievable in vivo, whilst proliferating cells are exquisitely sensitive to less than 1mM. One possible explanation for these findings is the conformation of the Bcr-Abl kinase in the quiescent versus proliferating stem cells. Recent studies suggest that Bcr-Abl conformation is absolutely critical for imatinib binding and function. Active Bcr-Abl is in an open (non-accessible) conformation, thus sensitivity to imatinib in CML is presumed to result from a dynamic switch between open and closed conformations possibly linked to cell cycle progression. This switch may not be triggered in quiescent cells; hence, imatinib may not be the optimal choice of agent to eradicate this population. A new generation of combined Src/Bcr- Abl kinase inhibitors that do not appear to be conformation sensitive and are 10-20-fold more potent than IM is now available and should therefore be more effective than imatinib.”

This new generation of Src/Bcr-Abl inhibitors mentioned on the Glasgow University Web site includes the new Bristol-Myers Squibb drug, BMS- 354825. If the Glasgow research group’s theory is correct, then BMS- 354825, or a similar drug, might be able to kill the resistant CML stem cells. It is interesting to note that in mouse models of CML, BMS-354825 is curative over a 40-fold dose range, while Gleevec is not curative, even at the maximum tolerated dose. If correct, this theory has implications for GIST — if both the active and inactive forms of KIT or PDGFRA are inhibited by BMS-354825, and residual quiescent tumor tissues that are still viable after treatment with Gleevec have activated KIT due to an active kinase formation. (this would be true whether or not the residual tissue had a stem cell origin or a clonal origin).

If the theory that quiescent cells have an active kinase conformation and are therefore resistant to being killed by Gleevec is correct, it would present a number of interesting questions:
— Could a drug that inhibits both the active and inactive kinase conformation have better efficacy than Gleevec? Could it be curative in some cases if used as front line treatment?
 — In some patients with stable disease, are there tumors that are in a quiescent state, and therefore resistant to  apoptosis?
— What is the right drug or drug combination for adjuvant therapy?
— Are researchers getting not only Gleevec-resistant tissue, but also residual viable tissues from patients responding to Gleevec?

One of the biggest challenges facing cancer stem cell research is the ability to separate the suspect cancer stem cells from the overwhelming majority of non-stem cell cancer cells. As hard as this is in leukemia, it is even more difficult in solid tumors like GIST. A Jan. 20 article on the NewScientist. com Web site describes new tests to identify cancer “ringleaders.” New techniques to do this have been developed at the University of Cambridge, U.K., and Kumamoto University, Japan, and have been licensed for commercialization to Stemline, a biotechnology company in New York.

“Once we have eradicated the cancer stem cells, in essence we have destroyed the engine responsible for treatment failure and disease recurrence, the major problems for fighting cancer,” says Ivan Bergstein, chief executive of Stemline.

It seems evident that there are at least two target populations in GIST and most cancers: those tumors/cells that respond to treatment (which often form the bulk of the tumors), and those that don’t respond to treatment. Proponents of a clonal vision of cancer might argue that there are many different target populations, each representing a different clone, and therefore each might require a separate drug or drug combination. The heterogeneity noted in GIST tumors to date might argue for the clonal vision. Whether these non-responding tumors/cells have a stem cell origin or a clonal vision, they still form a separate, often much smaller, population, and the effect of a drug is typically measured by its effects on the larger population. A drug or drug combination that might work perfectly on a second or third, smaller, population could appear to have no effect because response would be measured on the larger population.
Norman Scherzer, Life Raft Group executive director, contributed to this report