Lead Investigator: Daniel Von Hoff
Turning the Tables on Pancreatic Tumors
by Dan Ferber It’s one of the deadliest cancers around and more than 100 anticancer drugs have failed to stop it. But now, pancreatic cancer appears vulnerable, because new information about the genetics of cancer cells is leading researchers to create novel drug strategies, according to results presented April 2, at the 91st annual conference of the American Association of Cancer Research.
Of the 29,800 people who are projected to develop pancreatic cancer in 2000 in the United States, all but about 800 will die of the disease, often just months after they’re diagnosed. That makes the disease the single most lethal form of cancer, says Daniel Von Hoff, director of the Arizona Cancer Center in Tucson. Decades of drug-discovery effort and testing have yielded only one drug that has shown results for most patients, without any side effects such as allergic reactions. However, he says the effectiveness of that drug, 5-fluorouracil, has been debated.
With pancreatic cancer, “the creed has been, ‘let the patient pass away and hope the period of suffering is not too long,” he says, adding: “There’s a lot of people who don’t want to believe this creed.”
To identify new drugs to treat pancreatic cancer, he says, researchers are trying to turn the tables on cancer cells, by taking advantage of their propensity to become genetically unstable and accumulate mutations. For example, cancer cells in patients treated with 5-fluorouracil evolve to resist the drug by overexpressing its molecular target, an enzyme called thymidylate synthase (ST). So researchers have screened for and found a new agent that is activated by the enzyme, and will therefore target cancer cells preferentially.
To test the new drug, Von Hoff and his colleagues used a screen called the human tumor-cloning assay, which allows them to identify those agents that kill pancreatic cancer cells cultured from pancreatic tumors that have been removed surgically. “The results were very impressive,” Von Hoff says. The researchers are planning to test the drug, called NB1011, in a phase I clinical trial.
To identify new drug targets, Von Hoff and other researchers are taking advantage of rapidly accumulating knowledge about the specific genetic alterations in pancreatic cancer cells. For example, 14% of pancreatic tumors are deficient in an enzyme called MLH1/MSH2 that fixes small mistakes during DNA replication.
By creating a strain of yeast deficient for that enzyme and screening for drugs that kill it, the Arizona team identified a drug called MGI-114 that kills pancreatic cancer cells that are also deficient in that enzyme. The Arizona team has begun a clinical trial to test the safety of the compound in pancreatic cancer patients.
The researchers also used similar yeast strains to identify the molecular target of a drug used to treat the disease for years, with unimpressive results. Cultured cancer cells and yeast cells with a defective DNA-copying enzyme are susceptible to gemcitibine, but cells with intact enzyme are not. This might explain why only 18 percent of pancreatic cancer patients respond to gemcitibine, Von Hoff says.
Genetics can also identify entirely new and unexpected drug targets, Von Hoff says. Sometimes, mutations in cancer cells eliminate the function of a protein, and it is difficult to design a therapy to replace a missing protein. So researchers are using yeast to find unexpected new drug targets. For example, certain mutations weaken the ability of yeast to repair its DNA, but don’t kill the cell. Those same mutations make cancer cells unstable. By screening for mutations that kill the mutant yeast but not normal yeast, researchers can spot new targets for anticancer drugs.
With the exploding amount of genetic information about pancreatic tumors, Von Hoff is optimistic that new drugs can be developed to treat this deadly cancer. “The molecular genetic approach has given us substantial therapy leads,” he says.