Sa1382 Dietary Saturated Fatty Acids Are Positively Associated With the Development of Pancreatic Cancer -Data From a UK Prospective Study (EPIC) Using 7-Day Food Diaries

Abstract

Introduction An increased dietary intake of saturated fatty acids could promote pancreatic carcinogenesis through increasing insulin resistance. Our aim was to investigate this association for the first time using nutrient information measured by food diaries in a cohort study. Methods A total of 23,658 men and women aged 40–74 years recruited between 1993–1997 into the EPIC-Norfolk study, UK, completed 7-day food diaries which recorded all foods, brands, recipes and portion sizes eaten. The diaries were interpreted by nutritionists using a computer programme which converted participant recorded text into nutrient values of: total saturated fatty acids intake, palmitic and stearic acids. The cohort was monitored until June 2010 to identify those participants who developed incident pancreatic cancer, confirmed by a review of medical notes. The HRs of developing cancer were estimated across quintiles of intake using Cox regression, adjusted for total energy intake and oleic acid intake, the latter which increases insulin sensitivity. Results During 17 years of follow-up, 88 participants were diagnosed with pancreatic cancer (55% women, median age at diagnosis = 73.4 years, range 52.1–88.8 years). There were no associations with either total or individual saturated fatty acids in the whole cohort or in women. However, in men, increasing quintiles of dietary saturated fat acids were positively associated with risk (highest vs. lowest quintile HR = 9.68, 95% CI: 1.72–54.54, P = 0.01) with a trend across quintiles (Trend HR = 1.76, 95% CI: 1.17–2.64). There were similar associations for palmitic acid (trend HR = 2.29, 95% CI: 1.41.3.71) and stearic acid (trend HR = 1.51, 95% CI:1.00–2.30). Conclusion The data support an aetiological role of dietary saturated fatty acids in pancreatic cancer in men, due to the large effect sizes, dose-responses, temporality and plausible biological mechanisms. These macronutrients should be measured in future aetiological studies and the reasons for differences between genders investigated. Disclosure of Interest None Declared.