High cholesterol = high cancer risk?

High Cholesterol = High Cancer Risk?

Some epidemiologic studies haveshown an association between a high-fat diet and subsequent development of cancer, and this along with the heart desease link is why many doctors recommend that people lower their fat and cholesterol intake. But other studies have failed to show the cancer connection, and several have even found a link between low blood cholesterol levels and cancer.

Now, three new studies, two in Europeand one in the United States, have addressed the issue. They will no doubt add to the data base, but they won't necessarily answer the question of what to eat. The European studies strengthen the low-fat diet suggestion: They report an association between high blood cholesterol levels and an increased risk of colorectal cancer. The U.S. work shows no strong relationship between dietary fat and breast cancer.

While the European findings are inagreement with some previous studies, they run counter to others, including the Framingham (Mass.) heart study (SN: 1/24/81, p.55), that link low blood cholesterol levels to a higher incidence of colorectal cancer. In one of the newly reported studies, Sven A. Tornberg and his colleagues at the Karolinska Hospital in Stockholm, Sweden, analyzed data from more than 92,000 people over a 15-year period. Blood levels of cholesterol and cholesterol-laden lipoproteins were measured in the mid-1960s, and subsequent cancer was monitored.

The researchers found a statisticallysignificant correlation between high cholesterol and the development of rectal cancer in men: The cancer risk was 65 percent higher among the group with the top 20 percent of cholesterol levels compared with men in the bottom 20 percent. The data also pointed, although not as strongly, to a direct cancer-cholesterol relationship in men with colon cancer and in women with colon or rectal cancer.

Gerd Alexander Mannes and his colleaguesat the University of Munich in West Germany looked at people with colorectal adenomas, growths that are thought to be precancerous. When they compared blood cholesterol levels in these patients with the results obtained in the examinations of 842 people whose colons were checked for adenomas, they found "a small positive association." After adjusting for age and relative obesity, both of which increase cholesterol levels, the researchers determined that people who had the top 20 percent of cholesterol levels were twice as likely to have adenomas as those in the bottom 20 percent.

The association between high cholesteroland cancer could be an indirect one, both groups suggest in the Dec. 25 NEW ENGLAND JOURNAL OF MEDICINE. High blood cholesterol levels reflect a diet high in saturated fats, they say, and the higher cancer rate may be due to the production of carcinogens by fat metabolism.

The U.S. National Cancer Institute(NCI) recommends that Americans reduce their intake of saturated fat and cholesterol. Peter Greenwald, director of the division of cancer prevention at NCI, says the current reports "are very consistent with what we're saying." Cutting down on saturated fat and cholesterol "is good for both heart disease and cancer."

Greenwald says studies that showed alink between low cholesterol levels and colorectal cancer were not convincing in this respect. Many of them were designed to study heart disease incidence rather than cancer, he says, and the inverse relationship disappeared when data from people who developed cancer within two years were excluded. The disappearance could mean that the precancerous state had lowered their cholesterol levels and created a misleading connection, Greenwald observes.

But Paul Sorlie of the National Heart,Lung, and blood Institute, one of the researchers who analyzed the Framingham data, said ater a preliminary look at the reports that the Swedish study may have missed the low-cholesterol-cancer connection because the lowest cholesterol levels in the study group were higher than those in Framingham and elsewhere. "It's at the lower cholesterol levels that you see a higher risk," he says.

The inconsistency at the other end ofthe scale--why the Framingham data and other studies revealed no increased risk with increased cholesterol levels -- is at the moment unexplainable, he says. "I think it's still moot. It could be methodology, or it could be some difference between the populations, like lifestyle, genetics or nutrition."

The breast cancer study ends up in themiddle ground, with dietary fat intake showing no effect on the development of breast cancer. Harvard University researchers, led by Walter C. Willet, report in the Jan. 1 NEW ENGLAND JOURNAL OF MEDICINE on their study of nearly 90,000 female registered nurses who were questioned about their eating habits.

During a four-year prospective study,601 of them were diagnosed with breast cancer; an evaluation of the incidence among women in the top 20 percent of fat intake compared with the lowest 20 percent showed "no evidence of a positive relation." The finding could be due to a true lack of effect, or an effect could take longer than four years to show up, the authors suggest.