1) To investigate the relationship between dietary factors related to the glycemic effect and pancreatic cancer. Specifically, we will examine the associations between glycemic index, glycemic load, total available carbohydrates, and specific carbohydrates contributing to glycemic load (i.e. starch, sucrose, fructose) and the risk of pancreatic cancer in the PLCO trial. Total available carbohydrates will be used rather than total carbohydrates because resistant starch and fiber do not contribute to glycemic index and may overestimate glycemic load in this study (1). Because participants in this study have overall healthier lifestyles than the general population, we expect to find null or weakly positive associations in non-stratified analyses. 2) To examine whether insulin resistance or dietary fat intake modifies the associations between glycemic index/load and carbohydrate intake and the risk of pancreatic cancer. Because insulin and glucose measurements are not available, overweight and physical inactivity will be used as surrogates for insulin resistance in this study. We expect that each measure of glycemic effect has a stronger positive association with pancreatic cancer among overweight and/or physically inactive individuals. Sensitivity analyses among non-diabetic participants will be conducted due to possible dietary changes following diabetes diagnosis. The analyses will also be stratified according to percentage of dietary fat intake.

Kevin Dodd
Andrew Flood
Richard Hayes
Helen Mullholland
Jeanine Genkinger
Rachael Stolzenberg-Solomon
Regina Ziegler