Woman's cancer genome decoded

— -- Talk about personalized medicine: For the first time, scientists have mapped all the genes in a single person's cancer, allowing them to uncover eight new genes that could lead to better ways to treat the disease.

Researchers used malignant blood cells from a 50-something woman who died of acute myeloid leukemia, a cancer of blood-forming cells in the bone marrow, according to a paper in today's Nature. Doctors mapped all the genes in her tumor cells, the compared them — side by side — with the genes in a normal cell from her skin.

That allowed them to see exactly how the DNA of cancer differs from healthy DNA, says author Timothy Ley, a professor of medicine and genetics at the Washington University School of Medicine in St. Louis.

Ley and his colleagues found just 10 key genetic changes in the woman's leukemia. Eight were new genes never before linked to this kind of cancer.

Ley says his findings are a reminder of cancer's daunting complexity.

He examined tumor samples from 187 other patients with the same type of leukemia, hoping to find many genes in common — something that could make it easier to design one drug that would work for everyone. But none of the other patients had the same eight new mutations found in the woman's tumor.

That suggests that even cancers that look alike may actually be caused by completely different genetic changes, Ley says.

"There's a lot to learn," Ley says. "We're just getting started."

It's possible, Ley says, that doctors may find that there are dozens of genes that contribute to this type of leukemia. He compares the pathways to cancer development like the roads that lead to a city. Although these roads may follow different paths, they all converge at once place. In the future, doctors will try to design drugs that block the parts of these pathways that cells have in common, much as police set up roadblocks on commonly used highways.

Ley and colleagues are already mapping another patient's "cancer genome," as the collection of genes is called, and hopes to quickly map 10 more.

"We need to do hundreds, if not thousands, of each of the major cancer types," he says.

But researchers won't need to map every patient's genome, says Ley, who adds that sequencing this patient's genes cost $1.6 million.

After mapping enough individual genomes, Ley says, scientists will get a good idea of all the mutations that can make a cell malignant. That might allow them to design drugs that block the bad genes.

Doctors now use a designer drug to treat another type of leukemia, called chronic myeloid leukemia, as well as rare gastrointestinal tumor. That drug, Gleevec, is one of the few real breakthroughs in cancer. It works so well because it blocks the very first mutation involved in those cancers, says Gleevec's developer, Brian Druker, a professor at the Oregon Health and Science University Knight Cancer Institute who says Ley's techniques are incredibly promising.

Researchers already have developed a drug that blocks one of the well-known mutations in the woman's leukemia cells, called FLT-3. That drug has had limited success, however, because the FLT-3 mutation happens relatively late in the cancer process, after a cell has already become abnormal, Druker says.

"It's like trying to put out a wildfire after it already covers acres and acres," says Druker, who wasn't involved in Ley's study.

Ideally, he says, researchers will develop more drugs that block the earliest changes in cancer.