Genetic screening research IDs most aggressive adult leukemia strains

Stanford Report, April 28, 2004

Genetic screening research IDs most aggressive adult leukemia strains
A new screening tool may enable cancer patients to get customized treatment

By AMY ADAMS

In an effort to help doctors better identify those cancer patients in need of the most aggressive therapy, researchers at the School of Medicine have uncovered 133 genes that point to the most dangerous strains of adult acute myeloid leukemia, or AML.

Their research, published in a recent issue of The New England Journal of Medicine, may help doctors pick the best treatment for their AML patients.

AML is the most common acute form of leukemia in adults.

In devising a treatment plan, a doctor relies on indicators such as how the cancer cells look under a microscope as well as the patient's age and disease history in an effort to determine how aggressive the cancer is likely to be.

But for some patients, those indicators aren't enough to place them in a high-risk or low-risk group, and doctors may be left guessing which treatments to try.

"This intermediate-risk group poses the biggest problem when it comes to assigning treatment," said Jonathan Pollack, MD, PhD, assistant professor of pathology, who led the study.

For AML, Pollack said it's particularly important to know how to treat the cancer because aggressive forms often relapse after chemotherapy. Treatments such as stem cell transplants, on the other hand, are widely considered to be too risky for patients who may have less aggressive forms of the disease.

In the study, Pollack's colleagues in Germany collected tumor samples from 116 adults with AML, including 45 with an intermediate-risk form of the disease.

Lars Bullinger, MD, then a postdoctoral scholar in Pollack's lab, used a technique called microarray analysis to take a snapshot of the activity level of 26,260 genes in each sample. From this, the researchers found two distinct patterns of which genes were turned on or off.

These patterns correlated with how long the intermediate-risk patients lived after diagnosis. Those in the first group lived on average about twice as long as those in the second group.

"You don't see a difference between these cancers under a microscope, but when you look at gene expression they are quite distinct," Pollack said.

On its own, this information could help doctors identify patients who require aggressive treatment, Pollack said. What's more, some of the genes that are activated differently between the two groups were not previously known to be involved in AML. Studying these genes and their proteins could lead to new treatments for the disease, Pollack said.

The Stanford researchers took the study a step further. Once the cancers were divided into the two groups, they whittled the total number of genes to the 133 that showed the most difference in activation between the groups.

This handful of genes effectively discriminates between patients who have the most and the least aggressive cancers, and is a more manageable number for screening in a doctor's office.

This study joins a growing body of research relying on microarray analysis to distinguish between patients with otherwise similar cancers.

The technique has been successful in cases involving breast, lung and liver tumors at helping doctors understand the disease and to identify the most appropriate treatments.
"This kind of study can benefit patients in the short term, especially in matching treatments to individuals," Pollack said.
That's in contrast to developing new drugs, a process that can take up to 15 years to reach patients.

In order for gene activity to become widely used for choosing treatments, Pollack said the results first must be independently replicated and researchers must find the smallest number of genes that can still discriminate between the various treatment groups.

Robert Tibshirani, PhD, professor of health research and policy, and graduate student Eric Bair conducted the statistical analysis for this work and are co-authors of the paper.

Blood screen may thwart some radiation side effects (4/21/04)