Home > News

WEDNESDAY, April 14 (HealthDayNews) -- Scientists have moved one step closer to figuring out why different people with acute myeloid leukemia (AML) respond differently to different therapies.

Right now, a litany of factors -- including age, white blood cell count and, most importantly, the genetics of the cancer cells -- are factored into treatment decisions. But this has not been enough to determine the best therapy for each patient, new research suggests.

"Even within the so-called genetic subtypes, some patients do well and some don't and the question has always been raised, if you give the same drugs to the same group of patients, why do some do well and some do poorly?" said Dr. Stephen J. Forman, director of the Hematologic Neoplasia Program at City of Hope Cancer Center in Duarte, Calif.

However, two studies in the April 15 issue of the New England Journal of Medicine report on a technique called "gene expression profiling" of AML specimens that better determines which patients should receive which therapies.

Currently, chromosomal abnormalities in the cancer cells are used to assign AML patients to low-, intermediate- or high-risk groups. Low-risk patients are treated with standard chemotherapy, while high-risk patients are treated additionally with bone marrow transplantation, which can have serious side effects. That leaves the intermediate group.

"It's the intermediate-risk group that clinicians are least sure what to do with," said Dr. Jonathan R. Pollack, senior author of one of the studies and an assistant professor of pathology at Stanford University School of Medicine. "The largest potential benefit is with that group of people, and it's quite a large group."

Physicians don't want to give an aggressive therapy with major side effects to someone who is not likely to respond. On the other hand, it may be worth it if the patient has a good chance of getting better.

"The biggest potential benefit [from gene expression profiling] for [intermediate-risk] patients is that we might be able to more accurately assign the appropriate treatment to individual patients," Pollack said. "If you can match the therapy to the given patient's disease, that patient is going to do better with less potential harm."

Pollack and his colleagues examined gene expression levels in 65 blood samples and 54 bone marrow specimens from 116 intermediate-risk adults with AML, taken before the patients began treatment. The researchers then divided the samples into subgroups according to their particular gene expression profile, and then cross-referenced these with the patients' relapse and survival information.

The researchers were eventually able to identify variations in 133 genes that distinguished groups of patients.

While the most immediate benefit could be in assigning patients to existing therapies, this information may also help in the development of new therapies.

"Even if one identifies a good drug target, the whole process [of developing a drug] can take 10 to 15 years," Pollack said. "The benefits from that will be in the much longer term compared to the potentially more immediate benefits of improved treatment stratification of existing drugs. That potential benefit could be seen in a year or two."

A second study, this one out of the Netherlands, looked at the gene expression profiles in blood and bone marrow samples from 285 AML patients.

On the basis of this information, the researchers were able to identify 16 different groups of AML patients based on their molecular "signature." The presence of certain mutations and the level of gene expression indicated which patients had a poor prognosis and which had a better one.

Essentially, these studies show that scientists are approaching a new level of precision in treating AML, experts said.

"Many years ago all we had was looking under the microscope, then we had the ability to look at leukemia cells as to what their phenotype was. The next step was cytogenetics [which looks at abnormalities in the chromosomes of cancer cells] and that really helped, and this becomes the next level, molecular genetics," Forman said.

"It isn't just true in AML. For every single human cancer, there will be molecular genetic types that distinguish and help distinguish why some people do well and some don't," he said.

The procedure is not yet ready for prime time,Forman said. However, "my guess is that in the same way that cytogenetics became what doctors expect, I think this will be, within the next couple of years, something that is there, that doctors will expect."

More information

For more on acute myeloid leukemia, visit the National Cancer Institute or the Leukemia & Lymphoma Society.

(SOURCES: Jonathan R. Pollack, M.D., Ph.D., assistant professor, pathology, Stanford University School of Medicine, Palo Alto, Calif.; Stephen J. Forman, M.D., director, Hematologic Neoplasia Program, City of Hope Cancer Center, Duarte, Calif.; April 15, 2004, New England Journal of Medicine)

Copyright © 2004 ScoutNews, LLC. All rights reserved.