Insulin–Insulin-Like Growth Factor Axis and Colon Cancer

There is now compelling evidence linking physical inactivity and obesity with an increased risk of developing colon cancer. Several studies suggest that this association between lifestyle factors and colon cancer is mediated by hyperinsulinemia and insulin resistance, via the insulin–insulin-like growth factor (IGF) axis. Lending further support to this hypothesis is the fact that elevated plasma glucose and type II diabetes are also recognized as risk factors for developing colon cancer. Activation of the insulin-IGF axis may not only be an important risk factor for developing colon cancer, but may also have important prognostic implications in patients diagnosed with the disease. This parallels what has been previously shown in breast cancer where elevated levels of plasma insulin or C-peptide, or the presence of the insulin resistance syndrome, can increase the risk of both breast cancer recurrence and death. Understanding the insulin-IGF pathway and its role in colon cancer is therefore of clinical importance. The insulin-IGF pathway consists of three ligands (insulin, IGF-1, and IGF-2), six receptors (insulin receptor [IR] alpha [fetal], IR beta [adult], IGF-1 receptor [IGF-1R], IGF-2R, hybrid IGF-1R/IR alpha, hybrid IGF-1R/IR beta), and up to seven binding proteins (IGFBP1-7; Fig 1). Until recently much of the research on this pathway focused on IGF-1, a peptide growth factor that is secreted primarily by the liver in response to pituitary growth hormone, or high caloric intake. In the circulation, IGF-1 is mostly bound to IGFBP3, which not only regulates IGF-1 bioavailability, but may also have independent effects on cell proliferation and apoptosis. IGF-1 binds and activates both insulin and IGF receptors that are expressed on a number of cells including colon cancer cells. IGF-1-mediated receptor activation leads to cell growth, proliferation, and inhibition of apoptosis. Initial interest in the role of IGF-1 in colon cancer stemmed from observations that there was an increased risk of colorectal cancer noted in patients with acromegaly, a condition characterized by elevated levels of both pituitary growth hormone and IGF-1. This potential link was further supported by several in vitro studies on human colon cancer cells, which showed that IGF-1 promoted cell proliferation, IGF-1 receptors were frequently overexpressed on colon cancer cells, and IGF-1 R blockade with a monoclonal antibody inhibited cell proliferation. Clinically, however, a link between IGF-1 and colon cancer is less clear. Two prospective epidemiologic studies have shown that higher plasma IGF-1 and lower plasma IGFBP3 are associated with an increased risk of developing colon cancer; however, these associations were only statistically significant after adjustment for each other. It is unclear whether IGF-1 and IGFBP3 are causal factors in colorectal cancer. Insulin, another member of this family, which is most widely known for its metabolic effects, may also play an important role in colorectal carcinogenesis and progression. Preclinical studies have shown that insulin not only promotes the growth and survival of colorectal cancer cells, but can also promote IGF-I biosynthesis, and enhance IGF-I bioavailability by inhibiting the production of IGFBPs 1, 2, and 3. Insulin also directly stimulates in vitro mitogenesis of cultured normal colorectal epithelial cells and tumor angiogenesis. Animals fed high-fat, high caloric diets have increased insulin levels, develop insulin resistance, and experience increased colon tumor growth when compared with those fed calorie-restricted diets. There is also evidence linking elevated plasma insulin and glucose with increased adenoma risk and decreased apoptosis of normal colonic mucosal cells. Taken together, these results indicate that both IGF-1 and insulin may play a key role in colon cancer carcinogenesis and that further exploration is needed. Wolpin et al, in their article in this issue of Journal of Clinical Oncology, have demonstrated that, among patients with curatively resected colorectal cancer, higher levels of prediagnosis C-peptide and lower levels of plasma IGFBP1 were associated with increased mortality. They performed a prospective observational study nested within two large US cohorts to evaluate the association between mortality and prediagnosis circulating C-peptide, IGFBP1, IGF-I and IGFBP3. The use of C-peptide as a surrogate marker for plasma insulin secretion is not ideal; though in the nonfasting state, it is probably a more accurate marker of insulin resistance than insulin. There is growing evidence that C-peptide can independently activate intracellular signaling pathways that could affect cell growth and proliferation, leading to the possibility that C-peptide has effects that are not insulin related. In Wolpin’s study, it is interesting to note that C-peptide levels were associated with overall mortality but not colorectal cancer-specific mortality, whereas IGFBP1 was strongly associated with both overall and colorectal cancer–specific death. Since, hepatic expression of the IGFBP1 gene is strongly regulated by insulin, IGFBP1 may actually be a better marker of insulin activity than nonfasting C-peptide. However, like C-peptide, IGFBP1 may also exert direct effects on cellular function and may play an independent role in carcinogenesis and JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 27 NUMBER 2 JANUARY 1