Dietary Flavonoid Intake and Esophageal Cancer Risk in the European Prospective Investigation into Cancer and Nutrition Cohort

Esther Vermeulen, Raul Zamora-Ros, Eric J Duell, Leila Luján-Barroso, Heiner Boeing, Krasimira Aleksandrova, H Bas Bueno-de-Mesquita, Augustin Scalbert, Isabelle Romieu, Veronika Fedirko, Marina Touillaud, Guy Fagherazzi, Florence Perquier, Esther Molina-Montes, Maria-Dolores Chirlaque, Marcial Vicente Argüelles, Pilar Amiano, Aurelio Barricarte, Valeria Pala, Amalia Mattiello, Calogero Saieva, Rosario Tumino, Fulvio Ricceri, Antonia Trichopoulou, Effie Vasilopoulou, Gianna Ziara, Francesca L Crowe, Kay-Thee Khaw, Nicholas J Wareham, Annekatrin Lukanova, Verena A Grote, Anne Tjønneland, Jytte Halkjær, Lea Bredsdorff, Kim Overvad, Peter D Siersema, Petra H M Peeters, Anne M May, Elisabete Weiderpass, Guri Skeie, Anette Hjartåker, Rikard Landberg, Ingegerd Johansson, Emily Sonestedt, Ulrika Ericson, Elio Riboli, Carlos A González

We prospectively investigated dietary flavonoid intake and esophageal cancer risk in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort. The study included 477,312 adult subjects from 10 European countries. At baseline, country-specific validated dietary questionnaires were used. During a mean follow-up of 11 years (1992-2010), there were 341 incident esophageal cancer cases, of which 142 were esophageal adenocarcinoma (EAC), 176 were esophageal squamous cell carcinoma (ESCC), and 23 were other types of esophageal cancer. In crude models, a doubling in total dietary flavonoid intake was inversely associated with esophageal cancer risk (hazard ratio (HR) (log₂) = 0.87, 95% confidence interval (CI): 0.78, 0.98) but not in multivariable models (HR (log₂) = 0.97, 95% CI: 0.86, 1.10). After covariate adjustment, no statistically significant association was found between any flavonoid subclass and esophageal cancer, EAC, or ESCC. However, among current smokers, flavonols were statistically significantly associated with a reduced esophageal cancer risk (HR (log₂) = 0.72, 95% CI: 0.56, 0.94), whereas total flavonoids, flavanols, and flavan-3-ol monomers tended to be inversely associated with esophageal cancer risk. No associations were found in either never or former smokers. These findings suggest that dietary flavonoid intake was not associated with overall esophageal cancer, EAC, or ESCC risk, although total flavonoids and some flavonoid subclasses, particularly flavonols, may reduce the esophageal cancer risk among current smokers.