Pathophysiology

It has long been recognized that there is a strong environmental component to colorectal cancer. Dietary and other modifiable risk factors are estimated to account for up to 90% of colorectal cancers. Epidemiologic studies have confirmed that a western lifestyle including physical inactivity and high dietary fat and energy intake beyond expenditure, both of which lead to obesity, are important risk factors. The relationship between obesity and colon cancer risk is well studied, however the specific mechanistic pathways that lead to carcinogenesis are not conclusive. The association between the metabolic syndrome and colorectal cancer mortality and type 2 diabetes mellitus with colorectal cancer risk lend support to obesity-induced insulin resistance and hyperinsulinemia as a core pathway. Hyperinsulinemia results in a decrease in IGF-binding protein (IGFBP) and high levels of circulating insulin-like growth factor I (IGF-I). Activation of IGF receptors on colonocytes inhibit apoptosis, enhance epithelial cell proliferation and is associated with the development, progression, and metastatic potential of colon cancer. TNF-α, constitutively expressed in adipocytes, attenuates insulin signaling and causes insulin resistance by various mechanisms including inhibition of tyrosine phosphorylation, suppression of cytokine signaling protein 3 synthesis, and stimulation of adipocyte lipolysis leading to hypertriglyceridemia. In addition, obesity is associated with abnormalities in the bioavailability of plasma androgens and estrogens and increases levels of circulating sex hormones as a result of enhanced IGF-1 activity in the liver inhibiting hepatic sex hormone-binding globulin synthesis. Studies in male aromatase knockout mice (ArKO) have shown that a lack of estrogen results in the development of a metabolic syndrome characterized by excess adiposity, hepatic steatosis, and insulin resistance. The increase in androgen to estrogen ratio promotes visceral fat accumulation in the rodent model and leads to death of dopaminergic neurons in the hypothalamic acuate nucleus, which is pivotal to regulation of energy. These effects may account for the heightened risk of colorectal cancer observed in men and premenopausal women compared with postmenopausal women especially if they also take hormone replacement therapy (HRT). Leptin, a cytokine-like peptide, is produced in adipose tissue. It is involved in energy balance, regulation of food intake, and nutrient absorption in enterocytes. Leptin targets the ObRb receptor on colonocytes, leading to expression of transcription factors, cell proliferation, production of proinflammatory cytokines, and oxidative stress. Elevated levels of leptin have been associated with colon cancer although the association is inconsistent. Recent work demonstrates that leptin induces IGF-mediated pathway gene expression in colonocytes with an APC mutation suggesting a potential link between leptin and the increased risk of colon cancer associated with obesity. Leptin triggers an inflammatory response in tumor tissue through direct stimulation of colonocytes, inducing recruitment of cytotoxic T cells into the tumor microenvironment. This tumoral immune response, may be associated with the better prognosis associated with over expression of leptin receptor in patients with colon cancer.

Epidemiologic and randomized controlled trials have shown an inverse association between the risk of colorectal neoplasia and plasma concentrations or intake of calcium and/or vitamin D. Preliminary work on calcium intake, obesity, and insulin resistance in a cross-sectional study found that mean calcium intake was lower in obese adolescents than those of normal weight. Calcium intake was inversely associated with body trunk fat, insulin, and the homeostasis model assessment of insulin resistance (HOMA-IR) in the obese group. Girls in the highest quartile of calcium intake had decreased adiposity and insulin resistance. Greater adiposity has also been associated with lower blood levels of vitamin D and 25-hydroxyvitamin D. In a cross-sectional study of 381 healthy men and women from a randomized trial of calcium and vitamin D supplementation to prevent bone loss, individuals with the highest percentage body fat (>40%) had 20% lower 25-hydroxyvitamin D concentrations than those in the lowest quartile (<28% fat). This study suggests that vitamin D is sequestered in fat tissue, reducing its entry into the circulation and thus may be a factor in the obesity-related risk of neoplasia.

Evidence of an Association Between Obesity and Colorectal Cancer

The positive association between BMI and colon cancer risk has been well studied although some inconsistencies exist, in particular for women. In the NIH-AARP Diet and Health Study on 307,708 men and 209,436 women followed for 4.5 years, BMI showed a strong linear correlation with colon, but not rectal, cancer risk in men and women. The association was seen in younger (50–66 years) but not older (67–71 years) women and was not modified by HRT. The National Health and Nutrition Examination Survey (NHANES) epidemiologic follow-up study of 13,420 individuals in the United States followed up for approximately 15 years showed that baseline BMI in the overweight and obese range is associated with the development of colon cancer in men and women. The hazards ratio were similar for men and women, 2.47 (95% CI 1.14–5.32), 3.72 (95% CI 1.68–8.22), and 2.79 (95% CI 1.22–6.35) for a BMI of 26 to 27.9 kg/m ² , 28 to 29.9 kg/m ² , and 30 kg/m ² or more, respectively, compared with those with a BMI less than 22 kg/m ² . Subscapular skin fold thickness (which may reflect greater central adiposity), but not triceps skin fold thickness, was associated with colon cancer incidence in men but not women, suggesting that the distribution of adiposity affects the RR in a gender-specific way. To examine the independent effects of the distribution of adiposity on colon cancer risk, follow-up data on 3802 individuals in the Framingham cohort with BMI and WC measures at 2 age periods was analyzed. A BMI of 30 kg/m ² or more was associated with an increased risk of colon cancer in those aged 30 to 54 years (RR 1.5, 95% CI 0.92–2.5); the RR in those 55–79 years was 2.4 (95% CI 1.5–3.9). The BMI effect was stronger in men than women and for proximal colon cancer. These risks diminished when WC was added to the model. Men and women with a large WC had a 70% increased risk of colon cancer and those with an extra large WC had a 90% increased risk. This effect was not attenuated by BMI suggesting that body fat distribution is an independent risk factor for colon cancer, proximal and distal, in these subjects. Sedentary men and women in the highest WC category had the greatest risk RR of 3.0 (95% CI 1.0–9.1) and 4.4 (95% CI 1.5–12.9), respectively.

The European Prospective Investigation into Cancer and Nutrition (EPIC) examined the association between different anthropometric features and risk of colon and rectal cancer in nearly 370,000 European men and women followed for 6 years. No anthropometric measure was related to rectal cancer risk. A 55% increased risk of colon cancer was observed in men in the highest versus lowest quintiles of BMI but a similar association was not evident in women. WC and WHR were associated with a nearly 50% increased RR of colon cancer in men and women. The association of WC and WHR and colon cancer risk observed among postmenopausal women was ameliorated in those who used HRT.

Two recent meta-analyses have estimated the strength of the association between measures of adiposity and colorectal cancer in men and women. In the first publication, the estimated RR of colorectal cancer was 1.19 (95% CI 1.11–1.29) comparing BMI of 30 kg/m ² or more with BMI less than 25 kg/m ² and 1.45 (95% CI 1.31–1.61) comparing highest to lowest level of WC. There was a stronger association in men 1.41 (95% CI 1.30–1.54) than women 1.08 (95% CI 0.98–1.18). For every 2 kg/m ² increase in BMI the risk of developing colorectal cancer increases 7%. Every 2 cm increase in WC is associated with a 4% increased risk of colorectal cancer. The risk was greatest for colon, rather than rectal, cancer and was 30% higher in obese men compared with obese women. In the second analysis, the risks of BMI, WC, and WHR on colorectal cancer were assessed. For each 5-unit increase in BMI, the risk of colon cancer increased in men by 30% (95% CI 1.25–1.35) and in women by 12% (95% CI 1.07–1.18). BMI was associated with rectal cancer in men but not women. Colon cancer risk increased with every 10 cm increase in WC in men by 33% (95% CI 1.19–1.49) and by 16% (95% CI 1.09–1.23) in women, and with increasing WHR per 0.1 unit in men by 1.43 (95% CI 1.19–1.71) and women by 1.20 (95% CI 1.08–1.33).

Although the association between obesity and colorectal cancer seems irrefutable, there is evidence that the association of BMI, percent body fat, and WC on digestive system mortality can be attenuated by higher levels of muscular strength. Overall and site-specific cancer death rates/100,000 person years decreased across incremental thirds of muscular strength in 8677 men. Furthermore, the association between obesity and cancer mortality did not persist after adjusting for muscular strength. The mechanisms that may explain the lower risk of cancer mortality seen in men with higher levels of muscular strength may be through enhancement of insulin sensitivity by physical activity and increased glucose uptake by skeletal muscle.

Evidence of an Association Between Obesity and Adenomas

In addition to the positive association between increasing adiposity and the risk of colorectal cancer, several cross-sectional or case-control studies have also shown a moderately increased risk of obesity for an increased prevalence of colorectal adenomas. Determinants of insulin secretion and insulin-like growth factors have been directly associated with risk for colorectal cancer but few studies have looked at the factors in relation to colorectal adenomas. The Nurses Health Study identified 380 cases with distal colorectal adenomas diagnosed between 1989 and 1998 and 380 controls among nondiabetic women in the cohort of 32,826 women. Individuals in the highest quartile of C-peptide concentrations had the highest risk of adenoma 1.63 (95% CI 1.01–2.66) compared with the bottom quartile. This suggests that hyperinsulinemia may play a role in the development of adenomas. A cohort study on nearly 8000 average-risk Japanese with a mean age of 47.5 years analyzed the effect of being overweight and obese by BMI on the prevalence of colorectal neoplasia at baseline colonoscopy, and incidence of neoplasia in 2568 of the subjects on follow-up colonoscopy 1 year later. Not surprisingly, the prevalence of adenoma at baseline colonoscopy increased proportionally from 15.5%, 20.6%, 22.7% to 24.2% according to quartile of baseline BMI in kg/m ² (<21.35, ≥21.35 to <23.199, ≥23.199 to <25.156, ≥25.156). The prevalence of multiple adenomas increased in subjects according to the BMI, but no association with large adenomas or adenomas with high-grade dysplasia was noted. The incidence of adenoma on follow-up examination increased proportionally with BMI from 12.9%, 15.7%, 18.3% to 19%. The incidence rates were lowest (9.3%), in subjects who experienced a weight reduction, 16.2% in those with a weight gain, and 17.1% in those with no change in weight ( P = .01 reduction vs no reduction groups).

The positive association between obesity, weight gain, and adenoma risk was confirmed in 600 subjects who underwent colonoscopy as part of a prospective multi-ethnic cohort study called the Insulin Resistance Atherosclerosis Study (IRAS). Obesity, as determined by BMI at the time of colonoscopy, was associated with 2.16 (95% CI 1.13–4.14) risk of adenoma and was stronger in women 4.42 (95% CI 1.53–12.78) than in men 1.26 (95% CI 0.52–3.07). The risk of adenoma increased among participants who gained weight (>4 pounds) in the previous 5 years 2.30 (95% CI 1.25–4.22) and 10 years 2.12 (95% CI 1.25–3.62) compared with those who maintained their weight over the same period. A trend toward a protective benefit was noted in individuals who had a weight loss in the previous 10 years (odds ratio [OR] 0.38, 95% CI 0.13–1.11), but not those who reported a weight loss in the previous 5 years. In this study, the strongest associations with adenoma risk were observed when obesity was measured at the time of colonoscopy, suggesting that obesity may be a promoting factor in the growth of colorectal adenomas.

The Black Women’s Health Study followed 33,403 women older than 29 years to determine the risk of adenoma in African American women. The incidence rate ratio (IRR) comparing women with a current BMI of 35 kg/m ² or more to less than 25 kg/m ² was 1.35 (95% CI 1.12–1.62). The IRR, comparing the highest (≥0.87) to lowest (<0.71) quintiles of WHR was 1.26 (95% CI 1.04–1.54). The amount of weight gained since the age of 18 years was significantly associated with the risk of adenoma. Women gaining between 5 and 14 kg of weight had a risk of 1.34 (95% CI 0.99–1.51), those gaining 15–29 kg had a risk of 1.59 (95% CI 1.17–2.15), and those gaining 30 kg or more had a risk of 2.01 (95% CI 1.29–3.13).

Pooled data from 2465 individuals participating in 2 prospective randomized trials comparing the effect of a high versus low fiber diet and the effect of ursodeoxycholic acid on adenoma recurrence were used to assess the association between BMI and WC on adenoma recurrence. BMI was categorized as normal, overweight, or obese and WC (in inches) as small, medium, large, and extra large. Follow-up colonoscopy was done at a mean of 3.1 years after the clearing colonoscopy. The risk of recurrent advanced adenomas was increased in overweight men 1.60 (95% CI 1.09–2.33) and obese men 1.62 (95% CI 1.04–2.53) but this effect was not seen in women. WC was not associated with adenoma recurrence in the total population or in gender specific subgroups. In a larger analysis of 8 trials on 9167 subjects, Martinez and colleagues confirmed a BMI of 30 kg/m ² or more to be a significant independent risk factor for recurrent adenoma (1.23, 95% CI 1.08–1.41) but not advanced adenoma, compared with individuals with a BMI of less than 25 kg/m ² .

The effect of distribution of adiposity on colorectal neoplasia risk has been assessed in a few studies. The contribution of visceral adiposity to risk as measured by abdomino-pelvic computed tomography (CT) scan, was evaluated in 200 asymptomatic Korean subjects undergoing CT and colonoscopy for routine health evaluation. VAT and WC were significantly higher in those with colorectal neoplasia. The odds ratio of neoplasia was 4.07 (95% CI 1.01–16.43) for those with VAT over 136.61 cm ² over relative to those with a VAT of less than 67.23 cm ² . WC was not independently associated with colorectal neoplasia. Another Korean case-control study of 165 adenoma cases and 365 polyp-free controls determined that abdominal obesity (as measured at the top of the hip bone by a doctor, and defined by a waist circumference of ≥90 cm for men and ≥85 cm for women) was associated with an increased risk of colorectal adenoma. The association persisted when adjusted for BMI and the strength of the association was similar in men 2.77 (95% CI 1.54–5.00) and women 2.65 (95% CI 1.02–6.90). The prevalence of adenomas and advanced adenomas was 37% and 9% in the abdominally obese group versus 22% and 3% in the normal waist group, respectively.