Their goal: Prove these medicines prevent cancer and perhaps even help cure it.
While there are good scientific reasons to think they are onto something, the experiments to settle it are not finished, and the optimists could be dead wrong. Nevertheless, cancer researchers and pharmaceutical executives entertain fantasies of a breakthrough role for an off-the-shelf medicine.
Among them is Dr. Philip Needleman, who has spent more than a decade — first at Washington University and now at Pharmacia, where he is research director — developing Celebrex as an arthritis medicine. Just three years on the market, it is the 10th biggest selling prescription drug in the United States.
But in his vision, that success is a mere warm-up. "People ask what gets me juiced," Needleman says. "It's the possibility that in five or 10 years, someone will say, 'Oh yes. Celebrex. That's also used in arthritis."'
Many in industry, government and medical schools seem to share his daydream. In a world where people are used to disappointment, hopes run high that Celebrex and its rivals can be redirected against the most feared disease of all.
The drugs block production of a chemical called COX-2, which triggers pain and inflammation and may also fuel the growth of cancer, where it is often found in abundance.
The first big test will be in preventing colon cancer, second only to lung cancer as a killer. Three big studies involving about 6,000 volunteers will see if Celebrex and Vioxx stop pre-cancerous growths in the colon. The results should be known within two years.
"If we could reduce the incidence of this disease by half, what an incredible contribution that would be," muses Dr. Monica Bertagnolli of Brigham and Women's Hospital in Boston, director of one of the studies. "That's what is driving all of this."
But that might be just the start. Experiments on lab animals strongly suggest the arthritis drugs could also help cure cancer, especially if combined with chemotherapy or radiation. Whether the approach will actually work is unclear, since testing Celebrex on cancer patients has just begun, but many research teams are joining in.
In fact, Dr. Andrew Dannenberg, director of cancer prevention at Cornell's Weill Medical College, estimates there may be as many as 100 separate cancer studies involving these drugs worldwide.
"When in the history of drug development has a drug moved from arthritis to cancer prevention and then been fast-tracked into cancer therapy?" he says. "It's completely unprecedented."
One of these studies will examine whether Celebrex shows any sign of warding off lung cancer in 20-year, pack-a-day smokers. Many others are testing the drugs in people with cancer of the breast, lung, esophagus, skin, prostate and bladder.
One reason for doctors' willingness to try the drugs is their apparent safety. Celebrex and Vioxx — known as COX-2 inhibitors — were designed to be easier on the stomach than aspirin and other inflammation fighters. Their lack of frequent side effects makes them unique in cancer, a field not known for gentle therapies.
"Everyone is looking to integrate COX-2 inhibitors into every aspect of cancer treatment, because we don't expect them to be toxic," says Dr. Adam Dicker of Jefferson Medical College in Philadelphia.
A major sponsor of this research is the National Cancer Institute, which oversees Bertagnolli's colon cancer prevention study plus many of the smaller treatment experiments with Celebrex.
Dr. Ernest Hawk, the agency's chief of gastrointestinal research, says the government is betting on the drug because of the overlapping lines of evidence — from epidemiology, animal experiments and more — that the approach should work.
"What stands out is the weight of the evidence," Hawk says. "I can't think of any class of drugs that have this much going for them, especially in colon cancer. It's the consistency of the story."
Some of the most intriguing evidence comes from big population studies. At least 25 of them involving about 2 million people have examined whether colon cancer is less common in people who take aspirin, which works similarly to the new arthritis drugs. Together they suggest that regular users are about 40 percent less likely than expected to have precancerous polyps or to die from colon cancer.
Results of the first carefully controlled experiment of aspirin in colon cancer will be made public at a cancer conference in April. Dr. John Baron of Dartmouth Medical School said it will show aspirin in fact does slow the return of precancerous polyps after their removal, although the benefit is modest.
"There may be a tendency for people to hope for a magic bullet and not get it," Baron cautions. Nevertheless, "the idea that aspirin is effective opens the door for the COX-2 agents," which are safer and may be more powerful, too, as experiments on lab animals seem to show.
Celebrex is already approved to help block polyp formation in people with a rare genetic condition who develop tens of thousands of polyps and invariably get colon cancer by their 40s. Bertagnolli's study and two others, sponsored by Merck, Pharmacia and Pfizer, will settle whether the drugs can do the same for garden-variety polyps.
Some believe Baron's results are a strong hint they can, although the researchers hope for a more powerful effect than Baron saw with aspirin. In the end, they say, aspirin may turn out to be the best choice for people at risk of both heart attacks and colon cancer; the COX-2 blockers may make more sense for those worried solely about cancer.
Still, safety is likely to be a concern for the new drugs, too, since even rare side effects could outweigh the benefits when taken lifelong by large numbers of healthy people at average risk of cancer.
Aspirin, ibuprofen and similar nonsteroidal anti-inflammatory drugs shut off production of prostaglandins, substances that cause inflammation but also have benefits. They do this by interfering with an enzyme called cyclooxygenase, or COX, which the body needs to make prostaglandins.
In the late 1980s, Needleman and others found there actually are two varieties of COX: COX-1 promotes normal body functions, making blood clot and protecting the lining of the stomach. COX-2 comes into play in response to injury and illness and causes pain and inflammation.
The discovery led to the design of drugs that stop inflammation in a more precise way — by blocking only COX-2 rather than both kinds of COX, as aspirin does. The result is medicines that relieve arthritis while having little effect, either good or bad, on the heart or the stomach.
An inkling that the same drugs might also ward off cancer emerged from rat and mouse experiments in the 1990s. In one, scientists looked for a way to change the fate of mutant rodents that typically grow 600 colon polyps. When they knocked out the gene that makes COX-2, they found the animals' polyp counts fell to 90.
In another, they grew mice that make vastly elevated amounts of COX-2 in their mammary glands. The result: They got cancer.
In tumor-prone animals, Celebrex and Vioxx can suppress or prevent cancer of the colon, prostate, intestines, breast, skin, lungs, bladder and tongue. Many other anti-inflammatory drugs can do this, too, but the COX-2 blockers seem to be the most potent, at least in animal experiments.
Translating such discoveries into treatments for people is always tricky. But scientists say there is plenty of circumstantial evidence that COX-2 does bad things in humans, as well. For instance, levels of the enzyme are unusually high in some pre-malignant growths, such as colon polyps, as well as in many kinds of cancer.
Scientists are trying to learn exactly why, but they believe the process probably starts harmlessly enough. The body makes COX-2 as a normal response to injury but sometimes doesn't know when to stop.
"In those cells that go on to become cancers, the ability to turn off production of COX-2 may be lost," says Dr. Steven Dubinett, UCLA's chief of lung cancer research.
Scientists theorize too much COX-2 may help cancers flourish in several ways: It promotes the growth of new blood vessels, which tumors need for nourishment. It makes them less likely to die on cue, as ordinary cells do when they become hurt or defective. And it protects the tumor from the usual surveillance of the immune system.
By using drugs like Celebrex to turn down COX-2 production, scientists believe they may take away one of cancer's shields, making it more prone to killing by chemotherapy and radiation treatment.
"Often new drugs have surprising effects not initially anticipated, and they may be good or bad," says Dartmouth's Baron. "This whole story illustrates this perfectly."
by Daniel Q. Haney