Abstract 2823: Dietary glycemic index, glycemic load and insulin index, in relation to risk of prostate cancer

Abstract

Abstract Background: The insulin and insulin-like growth factor (IGF) axis has been shown to play a role in prostate cancer tumorigenesis. Circulating IGF-1 and fasting insulin as well as insulin resistance have been associated with increased risk of total or advanced prostate cancer in some epidemiological studies. Through alterations by diet the insulin and IGF axis may be potentially modifiable. The dietary insulin response depends largely but not entirely on the quality and quantity of dietary carbohydrates. The glycemic index (GI) is a measure of the propensity of carbohydrate containing foods to raise blood glucose levels postprandially. The more recently developed food insulin index is a comparable measure which is based on the diet's effect on insulin secretion, also taking into account foods with low or no carbohydrate content. We hypothesized that high dietary glycemic index, glycemic load (GL) and insulin index (II) are associated with increased risk of prostate cancer. Material and Methods: We followed 47,779 Health Professionals Follow-up Study participants between 1986 and 2006 for prostate cancer incidence. Dietary intake was assessed by food frequency questionnaire (FFQ) in 1986 and updated every 4 years. Dietary GL was calculated by multiplying GI by the carbohydrate content and consumption frequency of each FFQ item and then summing over all the items. Values for the food insulin index of the FFQ items were provided by the University of Sydney, Australia. The dietary II was calculated as a function of food insulin index, energy content and consumption frequency of all FFQ items. Multivariate adjusted hazard ratios (HR) and 95% confidence intervals (CI) were estimated using Cox proportional hazards regression. Results: Between 1986 and 2006, we confirmed 5,112 prostate cancer cases, including 860 advanced/fatal (regionally invasive or metastatic disease at diagnosis, or metastases or death from prostate cancer during follow-up) cases. Dietary GI, GL and II were not associated with risk of total or advanced prostate cancer. For total prostate cancer, the multivariate adjusted HRs comparing the highest versus lowest quintile were 1.01 (CI 0.92-1.12, p-trend=0.26) for GI, 1.07 (CI 0.97-1.19, p-trend=0.19) for GL and 1.05 (CI 0.94-1.17, p-trend=0.15) for II. For advanced prostate cancer, the HRs were 1.05 (CI 0.75-1.47, p-trend=0.76) for GI, 0.99 (CI 0.70-1.40, p-trend=0.64) for GL, and 0.99 (CI 0.68-1.44, p-trend=0.92) for II. Results did not change substantially when the cohort was restricted to men with PSA screening. Stratification by BMI, diabetes and age at diagnosis, or by tumor grade, did not suggest any heterogeneity (all p-values for interaction were >0.11). Conclusion: These results suggest that long-term exposure to a diet with a high insulin response as reflected by high dietary glycemic index, glycemic load and insulin index does not affect prostate cancer incidence or progression. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2823.