Devising treatments to combat every single mutation would be a huge task—but that may not be necessary, experts say. It seems that many of the aberrant genes operate together to create trouble in a smaller number of distinct cellular pathways—the molecular processes that make a cancer cell so deadly. "It has to grow faster, recruit a blood supply, and move around the body," notes Stephen Elledge, a geneticist at Harvard Medical School. These common pathways are the likely drug targets. Think of it this way: Multiple cars are taking separate routes to the same garage, and instead of trying to stop each car on the road, you block the driveway. The pathways are shared by cancers in different organs; the new lung cancer study, for example, revealed mutations also fingered in cancer of the retina and colon, among others, suggesting that drugs can be used against a panoply of tumors.
Within just a few years, the technology may be available to sequence an individual tumor's genes and find out exactly what pathways need blocking, says Raymond DuBois, provost of M.D. Anderson Cancer Center in Houston and president of the American Association for Cancer Research. Theoretically, that means drugs could be much better matched to patients. That kind of matching is only occasionally available now. Beyond Herceptin, Gleevec has proved very effective at blocking signals that encourage cell replication in certain leukemia patients with a specific chromosomal defect. Lung cancer drugs Tarceva and Iressa target the cell-growth receptor EGFR; for reasons not yet fully understood but most likely involving mutations in EGFR, they work better in some patients than in others.
Clearly, converting this mass of new genetic information into perfectly tailored therapies isn't a quick and dirty exercise. For one thing, finding all the mutations linked to a certain cancer isn't enough. "You have to separate the wheat from the chaff—some of those changes are driving the car, and others are just passengers," says Slamon. How and why are genes turned on and off? And how do the signaling pathways affect and influence each other? Even when a target is identified as important and a drug made to block it, chances are the tumor has another way to reach the same end—or will find one as it continues to grow and mutate, which explains why people develop resistance to targeted therapies that work at first. "Even as patients are responding, the tumors are trying to figure out a way to escape the block," says Neil Gibson, vice president of oncology research at Pfizer Inc.
What this all suggests, many say, is that just as HIV is now managed by a drug cocktail, cancer may one day be handled as a chronic disease controlled by a changing mix of several drugs at once in sequence or combination. The race is on to expand the arsenal. According to a recent Datamonitor report, 10 new targeted drugs have entered the marketplace since 2005, and researchers are studying existing therapies to see if they're more effective together. More are in the pipeline; Genentech and Roche, for example, are working on pertuzumab, which prevents the HER2 receptor from communicating with related receptors. There won't be one magic bullet, says Elledge. But "I absolutely believe we will be able to do this."
New Avenues of Exploration
While most doctors now talk in terms of managing cancer rather than curing it outright, there is one avenue of research—cancer stem cells—that might offer a permanent solution. Just as healthy adult stem cells in the body's organs produce cells for renewal and repair, the thinking goes, a small fraction of the cells in a single tumor manufacture tumor cells like little factories. Chemotherapy and radiation may seem to make a person cancer free when they actually leave the factories untouched to crank up production again later. Researchers have reported finding stem cells in leukemias and myelomas, as well as in breast, brain, pancreatic, and other tumors. Whether the cells are true stem cells or simply cells that have mutated to possess the stem-cell-like power of self-renewal is up for debate; either way, those properties are likely controlled by unique pathways that might be singled out and targeted.