Researchers are learning to predict before treatment how well patients are likely to respond to the new leukemia drug, Gleevec, and why patients relapse. The information suggests new ways to make the drug work better, according to reports in the Feb. 9 issue of The Lancet.

"Our work suggests Gleevec may be more effective when combined with similar drugs that attack different targets, and with other treatments," said Justus Duyster, MD, of the Technical University of Munich in Germany. Duyster was head of the study, which found why some patients relapse.

Gleevec, made by Novartis, is one of an emerging new class of "molecular target" cancer drugs meant to be more effective and less toxic because they aim only at targets on — or linked to — cancerous cells, largely leaving normal cells alone.

It works by attaching to — and thus disabling — a rogue enzyme that causes some kinds of white blood cells to reproduce very quickly, becoming leukemia. The disease treated by Gleevec is called chronic myelogenous leukemia (CML). It is a form of leukemia that strikes the infection-fighting white blood cells.

Many patients can do well for a long time on Gleevec if the drug is started when the disease is first found. But in those who start the treatment when they have advanced leukemia, it often stops working after a few weeks or months.

Earlier reports showed that in some patients where Gleevec stopped working, there were changes in the enzyme that Gleevec attacks. These changes caused the shape of the enzyme to change, so the drug could not fit into the new shape and block the enzyme from working. With the enzyme working again, the leukemia got worse.

Duyster's group looked at the enzyme before and after Gleevec treatment in leukemia cells of eight patients who relapsed, and found five new forms of the enzyme. All stopped the drug from working.

What probably happens in patients who relapse, he said, is that at first Gleevec kills off most of the leukemia cells, because most have the usual form of the enzyme. But any leukemia cells with a mutated form of the enzyme aren't killed by the drug, and continue making more copies of themselves.

Duyster's report shows it's crucial to knock out all leukemia cells as quickly as possible, to keep any from producing changed enzymes, said James D. Griffin, MD, of Dana-Farber Cancer Institute in Boston, in a comment in the same issue of The Lancet.

He agrees with Duyster that the best way to do that in some patients may be using Gleevec with other drugs that attack leukemia in other ways.

In the same issue of The Lancet, researchers led by Wolf-K Hofmann, MD, of University Hospital in Frankfurt/Main, Germany, looked at 2,500 genes in patients with the same enzyme-driven form of leukemia, both before Gleevec was given and afterward.

Comparing the activity of the genes in patients who did well on the drug to those who did not, they found 95 genes that may predict before treatment who will do well on Gleevec alone, and who won't. This knowledge could give doctors a chance to tailor treatment to each patient, according to Hofmann.

The two reports together "give a glimpse of the future," noted Griffin. An American Cancer Society (ACS) expert on new cancer therapies agrees.

"Gleevec has ushered in a phenomenal era of cancer drug therapy — it's more effective with fewer side effects, because it aims at a target present only in the kinds of leukemia cells it treats ," said T.J. Koerner, PhD, director of research information management for the ACS.

The new studies pinpoint the specific cause of Gleevec resistance, a great help in predicting what kinds of treatments will likely work best with Gleevec. Knowing the cause can also help prevent resistance, Koerner added.

"This work will help us continue to progress into a future where more patients have greater chances of long-term quality of life," Koerner concluded.