Diet, genetic factors, and the risk of gastroesophageal reflux disease, Barrett's esophagus and esophageal adenocarcinoma

<i>Introduction</i>: Diet has been recognized as the most pivotal modifiable lifestyle factor in the development of Gastroesophageal Reflux Disease (GERD), Barrett's Esophagus (BE) and Esophageal Adenocarcinoma (EA). However, the associations between dietary ingredients, patterns, and diet-gene interactions with the risk of GERD, BE and EA remain unclear. <i>Methods</i>: The prospective cohort study included 502 412 participants from the UK Biobank. Dietary ingredients, patterns, and the associations with the incidence of GERD, BE and EA were investigated using Cox proportional hazard models. Additionally, we calculated a healthy dietary score based on eight primary dietary factors. Furthermore, polygenic risk scores were developed to explore potential diet-gene interactions. <i>Results</i>: Over an average follow-up of 12.5 years, we identified 29 564 incident cases of GERD, 4081 cases of BE and 539 cases of EA. Frequent tea intake was consistently associated with an increased risk of GERD (HR = 1.04, CI: 1.03, 1.05), BE (HR = 1.08, CI: 1.06, 1.11) and EA (HR = 1.07, CI: 1.01, 1.14), while higher dietary fiber consumption was inversely associated with the risks (HR = 0.92 for GERD, CI: 0.90, 0.94; HR = 0.78 for BE, CI: 0.73, 0.84; HR = 0.81 for EA, CI: 0.68, 0.97). In the dietary pattern analysis, the prudent pattern was linked to a lower risk of GERD, BE and EA (HR = 0.95 for GERD, CI: 0.91, 0.98; HR = 0.84 for BE, CI: 0.76, 0.92; HR = 0.70 for EA, CI: 0.53, 0.93). Significant additive and multiplicative interactions were observed between diet and genetic risk for BE (RERI = 0.32, CI: 0.11, 0.52; <i>P</i>_{multiplicative interaction} = 0.039) and EA (RERI = 0.84, CI: 0.34, 1.33; <i>P</i>_{multiplicative interaction} = 0.038). <i>Discussion</i>: Our study highlights the role of dietary exposure in the etiology of GERD, BE and EA. Healthy dietary interventions may be beneficial, especially for populations with high genetic risk.