By Julie Steenhuysen CHICAGO, Sept 13 (Reuters) - Cancer cells often have a way of outsmarting new targeted drug therapies, but U.S. researchers said on Thursday a combination of targeted drugs could shut down a tumor's backup plan, resulting in much more effective treatments. A number of these so-called targeted cancer drugs -- such as Roche's <ROG.VX> Tarceva and Novartis' <NOVN.VX> Gleevec -- work by blocking the activity of various protein switches that tell the cell to grow. They are known as receptor tyrosine kinases or RTKs. "They essentially allow the cell to communicate with the external world to sense growth factors that could maintain the survival of a cancer cell," said Dr. Ronald DePinho of the Dana-Farber Cancer Institute and Harvard Medical School, whose study appears in the journal Science. These protein switches are on the surface of all cells, and they go haywire in a number of cancers. Drugs that target a single switch have transformed the treatment of some patients with certain cancers -- for instance, Gleevec and chronic myelogenous leukemia. But they only work in a small percentage of people. And certain tumors, including the aggressive brain cancer glioblastoma multiforme, respond poorly to such drugs. DePinho and colleagues now believe they know why. His team studied 20 different batches of glioblastoma cells in the lab and found that many growth switches were flipped on at once. In 19 of the 20 cell lines, three or more were switched on. They tested tumor samples from newly diagnosed cancer patients and got a similar result. "We found there was a multitude of activated receptor tyrosine kinases," DePinho said in a telephone interview. "When you would extinguish one with a specific targeted agent, the other ones would simply step in." 'BROADLY APPLICABLE' When they tried Gleevec, known generically as imatinib, it had little effect on the cells. But when they combined it with two other drugs -- Tarceva, known generically as erlotinib, and Pfizer's<PFE.N> SU-11274 -- the growth signal was shut down and the cancer cells died. "It's a very important observation scientifically and it has immediate clinical implications," DePinho said. "This is broadly applicable. This paradigm is true for virtually all solid tumors that we've looked at." He and colleagues hope to start testing combinations of these targeted therapies in patients. A person's tumors would be profiled first to determine which signals are active, and then doctors would pick a drug combination that would work best. DePinho said it would take some time to get these therapies to cancer patients because the drugs used in combination might turn out to be toxic. He agreed that a cocktail of targeted drugs would be costly. Tarceva -- approved to treat lung and pancreatic cancer -- costs around $3,000 for a 30-day supply. But he said one of the reasons for the drugs' current high cost is the high failure rate of drug development. "If we can use science ... to design better clinical trials, the costs will be a lot less," DePinho said.
