The most common strategy for designing new cancer drugs is to identify a target – a malfunctioning gene or protein – and screen large collections of compounds to see if any can block it. But what if the target is unknown or can't be reached by existing compounds?
Dana-Farber's Kimberly Stegmaier, MD, has devised a solution that holds promise for a variety of cancers. It relies on genetic fingerprints, distinctive patterns of gene activity within cells, as biomarkers of potential drugs' effectiveness.
A focus of her work is acute myeloid leukemia (AML), a condition where chemotherapy provides a cure for less than half of patients. AML cells are a classic example of arrested development. Stuck in a kind of eternal youth, they fail to take on characteristics of mature white blood cells known as leukocytes. Their genetic fingerprints don't resemble those of normal cells.
"A therapy for the disease would force AML cells to 'differentiate,' or take on the specialized role of mature leukocytes," Stegmaier remarks. "A sign of that change would be that the cells' gene fingerprints shift from the AML pattern to the leukocyte pattern."
Stegmaier and her colleagues screened some 10,000 chemicals to see if any caused this shift. One of several that did is a compound known to block a protein called EGFR (for epidermal growth factor receptor). But EGFR couldn't be the compound's target for the simple reason that AML cells don't have that protein.
Stegmaier reasoned that if the compound wasn't hitting EGFR, it was probably hitting a similar protein, one that, like EGFR, interferes with enzymes known as tyrosine kinases. The challenge now is to find which of these proteins is the actual target and use that information to identify potential therapies that can eventually be tested in patients – a task that Stegmaier and her colleagues are currently shouldering.
Spring/Summer 2009 Table of Contents
Dana-Farber Cancer Institute, 450 Brookline Avenue, Boston, MA 02215 | Call us toll-free:
(866) 408-DFCI (3324)