Obesity, insulin, and colon cancer.

Colon cancer is the second most common cause of death from malignancy for men and women combined in the United States. The incidence rates of colon cancer are high in essentially all economically developed countries. The importance of environmental factors is demonstrated by the substantial geographic variation in incidence of colon cancer, the consistent rises in incidence within populations undergoing economic development or “Westernization”, and the striking increases in the incidence of colon cancer in groups that have migrated from low to high incidence areas. Although several factors related to the Western lifestyle may contribute to cause colon cancer, many studies, including prospective studies, have found that obesity, usually assessed by body mass index (BMI), is associated with an increased risk of this cancer (1). This association appears to be stronger for men than for women, but most studies do support BMI as a risk factor for women. More recent evidence also suggests that a tendency for the central distribution of adipose (visceral adiposity) increases risk independently of BMI (1). The tendency for central adiposity in men could account in part for their stronger association. Whereas obesity increases risk, over 50 studies in diverse populations show that the most physically active individuals have approximately a 50 percent reduction in incidence of colon cancer (2). Although the evidence for a role of obesity and exercise is strong, the establishment of a mechanism would enhance the acceptability of exercise and weight maintenance as critical for colon cancer prevention. The consistent findings for obesity, central obesity, and physical inactivity as risk factors for colon cancer have generated the hypothesis that some component of the metabolic syndrome, characterized by insulin resistance, increases risk of colon cancer (1, 3). In addition to the similarity of risk factors for insulin resistance or hyperinsulinemia to those for colon cancer, this hypothesis has been supported by studies that have found individuals with type 2 diabetes mellitus to be at higher risk of colon cancer (4–6). Throughout the development of type 2 diabetes, individuals are exposed to high insulin levels.

The proposed underlying mechanism has typically been associated with the growth promoting properties of insulin. Insulin has several important functional relations with the insulin-like growth factor (IGF) axis, and variation in circulating insulin concentration is likely to be important in determining IGF-1 bioactivity. The concentration of growth hormone receptors in hepatocytes, which determines the liver's response to secrete IGF-1 upon growth hormone stimulation, is strongly regulated by insulin levels (7). Fasting decreases insulin production and concentration, which reduces growth hormone receptors in the liver, which lowers IGF-1 levels. Perhaps the more important role of insulin in non-fasting conditions is exerted through reducing levels of IGF binding proteins; specifically, high insulin levels inhibit hepatic secretion of IGFBP-1 and possibly IGFBP-2, which increases free or bioavailable IGF-1 levels (8, 9). Interestingly obesity is not associated with total IGF-1 levels, but does appear to be associated with free IGF-1 levels, probably due to this insulin-reducing effect on IGF binding proteins. Thus, hyperinsulinemia may stimulate cell proliferation and inhibit apoptosis through its effects on the IGF-1 axis.

The most direct evidence for the hypothesis would come from plasma or serum based studies which examine circulating insulin or related markers to colon cancer risk. Although the data are relatively limited at this point, they are generally supportive of the hypothesis that high levels of circulating insulin increases risk of colon cancer (10–12). In addition, in six studies of plasma IGF-1 and IGFBP-3, colorectal cancer (or its precursor adenoma) (13–18), most showed modest, mostly non-significant, increases in risk, with the exception of one study (17), which showed a stronger 2-3 fold increased risk. However, after IGF-1 levels were adjusted for IGFBP-3 levels, the association with IGF-1 generally increased, while a significant inverse association was observed for IGFBP-3 when adjusted for IGF-1 level. Thus, emerging evidence supports independent roles of hyperinulinemia and IGF-1 as risk factors for colon cancer.

Although insulin resistance has been proposed as the one of the underlying mechanisms, some aspects of this hypothesis require clarification. For example, either the elevated availability of energy substrates (triglycerides/glucose) (3) or the compensatory hyperinsulinemia have been emphasized as being critical (1). Some evidence suggests that hyperinsulinemia may be most relevant. First, the growth promoting properties of insulin are more directly related to the concentration of insulin itself than the underlying insulin resistance. For example, insulin concentrations directly influence concentrations of IGFBP1 (inversely). Secondly, direct markers of insulin secretion (e.g. C-peptide, 2-hour insulin) appear to be stronger predictors of colon cancer than markers of insulin resistance (e.g. fasting insulin) in at least some studies (10, 11), though further study is required. Thirdly, in at least one study, a substantially elevated risk of colon cancer was observed in women who had been diagnosed with type 2 diabetes mellitus in the relatively recent past, but this risk became attenuated 15 years after the diagnosis (4); this attenuation in risk is consistent with hypoinsulinemia at late stages of diabetes mellitus. Finally, in a recent study (19), compared to individuals with normal glucose tolerance, those with impaired glucose tolerance (but not diabetes) had a two-fold increased rate of total mortality, of which almost half of this increase resulted from cancer deaths, particularly from colon cancer. In contrast, diabetics had a three- to four-fold higher rate of total mortality, mostly due to cardiovascular consequences, but their cancer mortality was not appreciably increased. Hyperinsulinemia is generally more severe in those with impaired glucose tolerance but not diabetes, and diabetes occurs with worsening ability of the pancreas beta cells to produce insulin.

In conclusion, an increasing body of literature supports the hypothesis that insulin resistance and hyperinsulinemia increase the risk of colon cancer. This evidence further supports the recommendations for weight maintenance and exercise as preventive measures against colon cancer, in addition to a variety of other medical conditions.