Serum cholesterol and cancer risk: an epidemiologic perspective

This review has examined the evidence surrounding two questions: (a) Is having low serum cholesterol associated with increased risk of cancer? (b) Does reducing serum cholesterol increase the occurrence of cancer? Some elevated risk of cancer for males with low serum cholesterol levels has been noted: the median of the studies examined is consistent with a 30% increased risk. The answer for females is less clear. The median of the studies examined suggests no more than a 5-10% increased risk associated with having low serum cholesterol. However, the risk seems to depend strongly on whether females have a central or peripheral body fat pattern (54). The cancers most consistently associated with low serum cholesterol levels are those of the colon and lung in males, the cervix and breast (but only for females under 50 years of age) in females, and leukemia in both sexes. In contrast, high cholesterol levels have been linked with an increase in brain cancer. While immunologic, genetic, and dietary explanations have been offered to explain the association, it is difficult to support the idea that low serum cholesterol causes cancer in any direct manner. First, the findings themselves tend to be generally weak and somewhat inconsistent. Second, the strong influence of fat distribution in women suggests that a metabolic/hormonal basis underlies the association. One would not expect the results to differ by body fat pattern if the relationship were a causal one. Finally, if there were a direct causal role, one would expect populations with low serum cholesterol levels to have higher cancer rates. In China, counties with the lowest average plasma cholesterol levels have the lowest cancer rates (78). While this observation is open to a number of interpretations, it does not support the idea that low serum cholesterol is a tumor initiator. In aggregate, the trials of lipid-lowering interventions reviewed here show an increase in cancer occurrence (primarily mortality) of approximately 24% in the cholesterol-lowered groups. However, the post-trial experience has shown a comparative deficit of cancer occurrence in the experimental groups. Recent evidence indicates that products in the cholesterol biosynthetic pathway affect DNA replication and cell proliferation. These findings suggests a mechanism by which cholesterol lowering might accelerate the development of tumors already initiated. The data that have been reviewed in no way suggest that treatment of hypercholesterolemia should not be pursued. They do suggest the presence of a relatively small subpopulation in whom reduction of plasma cholesterol may lead to increased occurrence of cancer.(ABSTRACT TRUNCATED AT 400 WORDS)