Abstract A017: Lipidomic biomarkers of usual alcohol use and their association with pancreatic ductal adenocarcinoma risk

Abstract

Abstract There is suggestive epidemiologic evidence for an association between alcohol use and pancreatic ductal adenocarcinoma (PDAC). Previous studies reported alcohol use to be associated with high-density lipoprotein and fatty acyl ethyl esters. Lipidomic metabolites may improve assessment of self-reported alcohol use and provide mechanistic insights for the alcohol-PDAC association. We aimed to identify lipidomic signatures of self-reported alcohol use and evaluate their associations with PDAC risk using comprehensive lipidomic measures from nested case-control studies within the Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study and the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial cohorts (716 case-control sets). Controls were matched to cases (1:1) by age, sex, race, date of blood draw and cohort. Serum samples and self-reported alcohol use were collected at baseline up to 24 years before PDAC diagnosis. Absolute concentrations of 611 lipid species across 15 classes were measured on the Complex Lipid Panel (Lipidizer Sciex SelexIon-5500 QTRAP platform) and grouped into 161 class-specific fatty acids (FA) and 28 total FAs. We conducted cross-sectional analyses using linear regression to identify associations between log-transformed alcohol and 1-standard deviation (SD) increase in log-lipid concentrations among controls within each cohort. Conditional logistic regression and fixed-effects meta-analysis were used to calculate PDAC odds ratios (OR) per 1-SD increase in the log-concentration identified alcohol-lipidomic metabolites in common in both cohorts. All models were adjusted for age, smoking, BMI, height, diabetes, education, and diet quality. Alcohol was associated with 36 species, 13 class-specific FAs and 3 total FAs at Bonferroni thresholds <8.18 × 10−5, <3.11 × 10−4, and <1.79 × 10−3, respectively. These belonged to ceramides, cholesterylesters, diacylglycerols, hexosylceramides, lactosylceramides, lysophosphatidylcholines, phosphatidylcholines, phospatidylethanolamines, phosphatidylinositols, and triacylglycerols lipid classes, and mostly holding FA15:0 (pentadecanoic acid), FA17:0 (margaric acid) and FA18:2 (linoleic acid). Nine triacylglycerols (TAG(49:3-FA18:2), TAG(51:3-FA18:2), TAG(49:2-FA18:2), TAG(49:3-FA15:0), TAG(51:3-FA15:0), TAG(51:4-FA18:3), TAG(51:2-FA18:2), TAG(51:5-FA18:3), TAG(51:2-FA15:0)) and one phosphatidylcholine (PC(15:0/18:2)) lipid species and one total FA (FA15:0) showed significant inverse associations with PDAC with ORs ranging from 0.82 to 0.86 (FDR<0.10). There was no association between class-specific FAs and PDAC. Our results support the hypothesis that alcohol use is associated with serum lipid species from 10 lipid classes, mostly with FA15:0, FA17:0 and FA18:2. Nine triacylglycerols, one phosphatidylcholine and total FA15:0 showed inverse prospective associations with PDAC, the latter having been shown to decrease inflammation and regulate glucose metabolism in previous studies. These findings may help understand the inconsistent association between alcohol use and PDAC. Citation Format: Sabine Naudin, Paloma Mitra, Ting Zhang, Demetrius Albanes, Steven C. Moore, Rachael Stolzenberg-Solomon. Lipidomic biomarkers of usual alcohol use and their association with pancreatic ductal adenocarcinoma risk [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Pancreatic Cancer; 2023 Sep 27-30; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(2 Suppl):Abstract nr A017.