Colorectal cancer: to screen or not to screen?

A simple, widely available blood test for colorectal cancer could save nearly 10,000 lives a year in the United States and thousands more worldwide, two new studies show-but only if more people use it.

Although the test is painless, many people find it repugnant because it requires them to supply a fecal sample for lab analysis. Still, doctors say, a little unpleasantness beats risking an untimely, agonizing death.

Colorectal cancer kills 55,000 people in the United States each year. The fecal occult blood test, often ordered by physicians for people over age 50, detects traces of blood shed by malignant bowel tissue. This enables doctors to diagnose and remove some tumors early, boosting a person's odds of sur- vival. Still, because the test detects just one of every four tumors, doctors have questioned whether mass screening would improve overall survival.

Until now, just one controlled trial had addressed the effectiveness of large-scale screening. In that study of 46,551 people, Jack S. Mandel and his colleagues at the University of Minnesota in Minneapolis reported that screen- ing reduces colorectal cancer mortality by one-third. But the study, published in the May 13, 1993 New England Journal of Medicine, had a drawback. The researchers used a highly sensitive test that misses fewer cases of colorectal cancer, but yields more false positives, than the one generally available. Doctors then had to perform the more expensive, more invasive colonoscopy to confirm the presence of cancer.

The new studies, done in Nottingham, England, and Funen, Denmark, assessed the blood test now in common use, the researchers report in the Nov. 30 Lancet.

Jack D. Hardcastle and his colleagues at the University Hospital of Queens Medical Centre in Nottingham tested 44,838 symptomfree men and women age 45 to 74 at least once between 1981 and 1991. The researchers compared the per- centage of deaths from colorectal cancer in this group with that in a group of 74,998 untested people. They found that 15 percent more people in the untested group died of the disease-and recommended that the United Kingdom consider routine screening for colorectal cancer.

In Funen, Ole Kronborg and his team at Odense University Hospital studied 30,966 people who had not been tested and 30,967 people who had. The volunteers ranged in age from 45 to 75. The researchers found that 18 percent more unscreened than screened people died of the disease. Kronborg also con- cludes that mass testing could be beneficial.

The American Cancer Society estimates that just one in four U.S. adults has had the test. To expand this pool, the nation should begin mass screening without waiting for "perfect tests," argue David Lieberman of the Oregon Health Sciences University in Portland and Marvin H. Sleisenger of the Univer- sity of California, San Francisco in a Lancet editorial. The cost of such testing, however, would work out to $200,000 for each life saved, they note.

Efforts to develop more accurate tests are under way. Bert Vogelstein of Johns Hopkins Medical Institutions in Baltimore is working on a way to detect an altered version of the ras gene, which is shed in the feces of people with colorectal cancer. "Its mutations are particularly easy to detect," he says.

David Shibata of Harvard Medical School in Boston and his colleagues report in the Dec. 5 New England Journal of Medicine that people whose tumors express the DCC (deleted in colon cancer) gene are more likely to survive than those whose tumors do not. Ultimately, they say, patients with better prospects might benefit most from aggressive therapy.