Serum cholesterol and the incidence of cancer in a large cohort.

The relation of serum cholesterol to the development of cancer remains unclear. We have examined the incidence of cancer in 160,135 men and women who were health plan members of the Northern California Kaiser Permanente Medical Care Program (KPMCP) and who had a multiphasic health examination in the period 1964 through 1972. In addition to all cancer, we examined individually the nine most common cancers in men and the 12 most common in women. Cancer diagnoses (N = 7477) recorded by the State of California Resource for Cancer Epidemiology Section and the health plan hospital discharge file were used to calculate incidence by quintile of serum cholesterol. Multivariate analysis adjusted for race, education, smoking status, alcohol consumption, body mass index and, in women, for age at menarche, parity, and menopausal status. No strong or consistent relation of low cholesterol to cancer incidence was found. Of the 21 cancers examined, only lymphoma in men and cervical cancer had significantly elevated risks at the lowest quintile of serum cholesterol compared with risks in the highest quintile. Cancer incidence in the first 2 years after the cholesterol measurement was consistently higher among persons whose cholesterol levels were in the lowest quintile. This prospective study did not find evidence that low cholesterol increased the risk of cancer but supports the idea that preclinical cancer in some way lowers serum cholesterol.