A Gloomy Assessment of the War on Cancer

Years of intense research to develop breakthrough cancer treatments have largely failed to make a dent in cancer death rates in the United States, according to a new report. The study, published in tomorrow's New England Journal of Medicine, shows that despite the War on Cancer, cancer death rates were 6% higher in 1994 than in 1970. Recent strides against the disease have come largely from improved prevention and detection rather than treatment, the authors contend, and as a result, they call for a shift in priorities in the nation's cancer research program.

The study, led by John C. Bailar III, an epidemiologist at the University of Chicago, and his Chicago colleague, biostatistician Heather Gornik, analyzed cancer death numbers from 1970 to 1994 from the National Center for Health Statistics. They determined age-adjusted mortality rates for each type of cancer according to year, race, and sex. Mortality rates were adjusted to make up for changes in the overall size of the population and changes in age distribution over the 25-year study period.

Like previous studies, Bailar and Gornik found that overall cancer death rates rose steadily until the early 1990s, but dropped by 1% from 1991 to 1994. Last November, a team from the University of Alabama, Birmingham, reported a 3.1% drop from 1990 to 1995. Nevertheless, Bailar notes that the overall cancer mortality rate today continues to be higher than in 1970. The numbers also revealed that the prevalence of certain types of cancer is shifting. For example, the prevalence of melanomas and cancers of the brain and prostate is rising, while cancer of the stomach is falling. Meanwhile, better detection and prevention methods are causing a drop in colon and rectal cancer mortality rates. At the same time, the authors point out that a few treatments have been remarkably successful. There have been marked improvements, for example, in survival rates from cancers such as Hodgkin's disease, which tends to affect children and young adults. But because the number of such cancers is small, the overall effect is minimal.

The findings underscore the fact that "a great deal of the research that has gone into improving cancer treatments does not seem to be paying off," says Clark Heath, an epidemiologist at the American Cancer Society in Atlanta. Bailar and Gornik therefore propose that more research dollars be spent on prevention and detection methods, even, if necessary, at the expense of treatment research. Currently, for every $5 spent on treatment research, $1 goes to prevention, Bailar says. More research money is needed to study topics such as the cancerous effects of dietary choices and the cancer-causing effects of water and air pollution, says Bailar.

"We are not attacking cancer treatment," Bailar says. Such treatments cure almost half of all cancer patients and "can do a great deal for those who can't be cured," in terms of helping them live longer and more comfortably. "But the 50% cure rate is not good enough, because it's failed to bring the cancer death rate down and provides a strong argument in favor of a shift in research focus." Not all agree with that assessment, however. "No one is satisfied with a 50% mortality rate," says Samuel Broder, former director of the National Cancer Institute. But he adds that it would be a "serious error to be overly pessimistic" about future prospects for treatment.