By comparing cells from a set of twins, one of whom has a common form of childhood leukemia, a team of scientists from Britain, Japan and Italy have been able to isolate not just the cancer stem cells, but also precancerous cells that give rise to the disorder but have not yet become cancer cells.

That work is published in the Jan. 18, 2008, issue of Science.

"It's the earliest yet that anyone has been able to see," senior author Tariq Enver, professor at Oxford University's Weatherall Institute of Molecular Medicine, told BioWorld Today, calling it "first base."

Acute lymphoblastic leukemia, or ALL, is one of the most common pediatric cancers, and needless to say, the cellular transformations that ultimately lead to ALL are of keen interest to researchers and drug developers alike.

The problem, as with many cancers, is that by the time cancerous cells call any attention to themselves clinically, they have accumulated a series of mutations. And at the stage where the first mutations occur, the child does not yet show any clinical symptoms.

"You could line up a school of children in front of me, and I wouldn't be able to tell who has preleukemia," Enver said.

But the set of twins whose cells are described in the Science paper, because they shared a placenta, passed cells back and forth in the womb - among them, a preleukemic cell that had the fusion of two transcription factors, TEL and AML. That fusion is the first step on the road to developing ALL.

Both girls have a population of cells with fused TEL and AML. In one twin, the preleukemic cell has gone on to leukemia, but in the other, the leukemic cells are missing a second copy of TEL in addition to the TEL-AML fusion. That unusual instance gives the researchers a unique, if grim, opportunity to identify and study blood samples from a child with preleukemia.

The researchers first studied blood samples from the twins and identified a B precursor cell that is present in the bone marrow of leukemia patients, but not found in normal children. When they transplanted those cells - known as CD34-positiveCD38-negative/lowCD19-positive, for the surface markers they express - into mice, the cells were able to engraft in the bone marrow and proliferate, confirming them as cancer stem cells.

The researchers then isolated the same cell type from both the leukemic and the preleukemic twin, and transplanted it into mice. Like the leukemic stem cells, the cells from the preleukemic twin were able to engraft, showing that, though they are not cancer cells, they do have stem cell potential. The researchers were unable to detect cells with the same combination of surface markers in control blood samples from healthy children.

On the medical side, the identification of a cancer stem cell "has a bunch of therapeutic implications for the future," Enver said, including the possibility of specifically targeting cancer stem cells, or monitoring them to determine whether a therapy is successful.

"Current therapies work pretty well, but they are harsh," Enver said. "It's like using a sledgehammer to crack a nut." The leukemic twin is the best example of that. Though in remission, she is blind in one eye and shorter than her sister as a result of the chemotherapy.

New therapies might be able to target the cancer stem cells directly - and if they do, they might be specific enough to justify their use in cases of preleukemia, where the risk of the treatment currently is not justified given that the preleukemic cells have a low chance of progressing to outright leukemia.

In addition, "we can now generate the preleukemic cell and test . . . ideas" about how cancer develops from the original molecular changes, he said. "Perhaps, this will give us insight into what triggers the transformation."

The Science paper was published a day after researchers identified a melanoma stem cell in a separate paper published in the Jan 17, 2008, issue of Nature. In it, researchers from Harvard Medical School showed that a subpopulation of melanoma cells that express the surface marker ABCB5 were more tumorigenic in serial transplantation experiments than other cells, and were able to make different tumor cell types.

If mice were transplanted with the melanoma stem cells and then treated with an antibody to ABCB5, tumor growth was inhibited.

The authors noted in the paper that "identification of tumor-initiating cells with enhanced abundance in more advanced disease but susceptibility to specific targeting through a defining chemoresistance determinant has important implications for cancer therapy."