A Different Kind of Cancer Risk
If researcher Lorne Brandes is right, a number of drugs are to cancer as gasoline is to fire
Suppose a drug taken by millions of people is found one day to accelerate the growth of cancerous cells--in lab mice and rats. The drug doesn't cause cancer. But it seems to stimulate cancerous cells that otherwise would usually repair themselves or be gobbled up by the body's immune system.
Now suppose the same drug helps a melancholy middle-aged executive abandon thoughts of suicide; helps a once despondent elderly woman find renewed joy in her grandchildren; helps a young college graduate once empty of hope envision a future full of promise.
Should candidates for the drug be told that taking it might multiply their chances of cancer? Surely everyone has the right to weigh the risks and benefits of any drug. But which is worse: a future clouded by bleak thoughts or by cancer? And what about people who are depressed because they have cancer?
The dilemma is anything but hypothetical. Canadian researcher Lorne Brandes has concluded that certain antidepressants do make cancerous tumors induced in lab rodents grow faster. "This is a very unhappy finding, but what am I supposed to do--deep-six it?" asks Brandes, senior investigator and medical oncologist with the Manitoba Cancer Treatment and Research Foundation in Winnipeg. Additional work by Brandes has produced similar conclusions about certain antihistamines. The drugs' manufacturers have told U.S. News that they stand behind the safety of their products.
Many scientists have long worried that beneficial drugs are kept off the market because the lab tests required by the Food and Drug Administration unrealistically brand them as carcinogens. In the tests, lab animals receive doses of the drugs many times greater than any human would take, then are watched over an entire lifetime to see whether cancer develops. Bruce Ames, a noted biochemist at the University of California at Berkeley and developer of a simple test-tube screening that quickly fingers many carcinogens, asserts that standard animal tests are nearly useless for predicting risks to humans. About half of all chemicals, says Ames, could cause cancer in rodents simply because the superhigh doses produce a constant level of irritation to animal cells--much as a lifetime of smoking does. In April 1992, a panel of experts convened to re-evaluate standard tests for assessing human health risks from chemicals, including the risk of cancer from pharmaceuticals. The panel urged that additional tests be investigated to reflect new understanding of the many steps to cancer.
Then there are the researchers, Brandes among them, who worry about drugs reaching the market that don't actually cause cancer but might promote it--something existing tests are not designed to catch. In Brandes's study, published in July 1992 in the journal Cancer Research, rodents were deliberately given cancer. One group of rodents was then injected with mouse-size equivalents of human doses of Prozac or Elavil. Rather than letting the animals live out a normal life span of 18 to 24 months, Brandes killed them after only a few weeks. Their tumors were two to three times heavier than tumors in the rodents not exposed to the drugs.
advertisement
