Acute Myeloid Leukemia: Turned on Genes Provide Keys to Treatment

BY MARK SHWARTZ

WINTER 2003 - Forty years ago, a child diagnosed with cancer had only a one in four chance of surviving. Today, the overall survival rate is 80 percent. Despite this dramatic turnaround, some pediatric cancers remain stubbornly difficult to treat.

Packard researchers are searching for the best combination of therapies for patients such as 4-year-old Ariana Riccio, who is battling Acute Myeloid Leukemia (AML), a rare form of leukemia that is difficult to treat. Pediatric Oncologist Gary Dahl, M.D., is using gene chip technology to analyze how a child's genetic profile can indicate which course of treatment will be most effective.

In an effort to find novel therapies and cures, oncologists at Packard Children's Hospital have begun zeroing in on the underlying genetic causes of cancer. Gary V. Dahl, M.D., a professor of pediatrics at Stanford, has devoted more than two decades of research to the fight against acute myeloid leukemia (AML) -- a rare form of leukemia that attacks white blood cells in the bone marrow, then spreads to vital organs such as the brain and spleen.

The standard treatment for AML includes blood transfusions, months of chemotherapy, and bone marrow transplants -- a difficult regimen that's only effective for some children.

"On average, half the patients diagnosed with AML die within two years," Dahl says. "At Packard, we see between four and 10 new cases annually."

Scientists first uncovered a genetic link to AML in the 1980s with pioneering studies of human chromosomes - large molecules of DNA that carry all of our genes.

These early studies found that 25 percent of AML patients had chromosomal abnormalities in their leukemia cells. "We now know that children who have these abnormalities actually do better than others with AML," Dahl says. "The question is why."

To find out, he and his colleagues have been conducting a long-term genetic study of more than 400 AML patients who are 18 or younger. Instead of examining large chromosomes under the microscope, however, the research team is using gene chip technology developed at Stanford to analyze all of the genes in all 400 patients -- that's approximately 30,000 genes per person.

"The amount of information we're producing is phenomenal," Dahl says. "These gene chips can analyze roughly 22,000 genes at once, so we can compare one person's leukemia cells with another person's and see which genes are turned on (expressed) and which are turned off."

The researchers will use these data to determine if there is a direct relationship between gene expression and a patient's response to various treatments.

"Perhaps we'll be able to identify which patients should get therapy earlier," Dahl says. "Or maybe we'll find out that patients with a certain genetic profile shouldn't be put at risk for a bone marrow transplant because they're going to do well with chemotherapy alone. Before the gene chip, you simply couldn't do this kind of research."

As part of the study, Dahl and his co-workers are looking at a specific gene known as FLT3. While certain leukemia cell chromosomal abnormalities can improve a patient's chance of survival, an abnormal FLT3 gene can have the opposite effect.

"We know that some 20 percent of AML patients have FLT3 mutations, and that patients with a mutation have poorer outcomes than people who don't have it," Dahl says, noting that the cure for AML may ultimately lie in solving these genetic mysteries.

"Leukemia is a population explosion of millions of baby cancer cells that just keep growing," he says. "People are looking for drugs that will interfere with the cancer-causing genes and allow the cells to mature and die a normal death, which is what normal cells do. When I started, only 3 percent of AML patients were cured, so I can see with my own eyes how the application of new drugs has made that difference."

Another technology that has revolutionized AML diagnostics is flow cytometry -- a technique developed by Stanford geneticist Leonard A. Herzenberg, Ph.D., more than three decades ago. The latest version, called super flow cytometry, allows researchers to identify a single cancerous cell among 100,000 healthy ones -- compared to only one in 100 that can be visually identified with a microscope.

"My job is to get rid of this disease," Dahl concludes. "How do I do it? One way is to treat the patients and cure them. Another way is to understand leukemia so you can prevent it.

"As an investigator, I'm absolutely dependent on bioinformatics, mathematicians and the developers of new hardware and software. There's a tremendous amount of cooperation that has to go on."