CAN MODERATE ALCOHOL CONSUMPTION DECREASE PANCREATIC DUCTAL ADENOCARCINOMA (PDAC) RISK? A SYSTEMATIC REVIEW AND META-ANALYSIS

Introduction PDAC is increasing in incidence and has the worst survival of any malignancy. Lifestyle choices to reduce risk, based on robust evidence, are required. There are plausible biological mechanisms for how a moderate intake of alcohol (UK recommendations of ≤14 units/week) could prevent PDAC. These are alcohol: mitigating against pathological effects of KRAS genetic mutations, decreasing excess mitogenic insulin secretion, and the effects of antioxidants in alcoholic drinks. These mechanisms must be supported by robust epidemiological data to infer causality. The aim of this work is to systematically review prospective cohort studies, selected according to predefined criteria, to assess if a moderate alcohol consumption is inversely associated with PDAC risk. Cohort studies minimise biases inherent in other observational methodologies. Methods Medline, PubMed and EMBASE electronic databases were searched for prospective cohort studies investigating a moderate alcohol consumption and PDAC up to January 2019. The inclusion criteria were: >100 PDAC cases, definition of moderate consumption within UK guidelines, clear description of the comparator group, reporting effect sizes as risk/hazard ratios (RR/HR) and adjustment for covariates including: age, sex, smoking, body mass index and diabetes mellitus. Pooled effect sizes were calculated using an inverse variance random-effects model. Results Four prospective cohort studies meeting inclusion criteria were identified (9,090 PDAC cases, 57.5% female, mean age at diagnosis was 61.1 years). The cohorts were in the Netherlands, UK, pan-European and the US. The pooled effect size showed no association between a moderate alcohol consumption and PDAC risk (HR = 0.97, 95% CI; 0.91–1.03, I2 for heterogeneity 0.0%). There was no evidence of publication bias (Egger’s p = 0.404). Three cohorts investigated moderate intake of different beverages, reporting no association with either: wine (HR = 0.95, 95% CI; 0.83–1.09, I2 = 0.0%) beer (HR = 0.96, 95% CI; 0.79–1.17, I2 = 0.0%) or liquor (HR = 1.18, 95% CI, 0.89–1.57, I2 = 71.7%). There was no biological gradient within the moderate range. Conclusions The plausible biological mechanisms for a protective effect of alcohol are not supported by the epidemiological work. Drinking alcohol moderately cannot be recommended to decrease PDAC risk in the general population. Further research in a large prospective cohort which investigates risk in specific groups such as smokers and according to body mass index is required to see if alcohol may be beneficial in high risk groups for PDAC.