• Pietro Ferrari, Sabina Rinaldi, Mazda Jenab, Annekatrin Lukanova, Anja Olsen, Anne Tjønneland, Kim Overvad, Françoise Clavel-Chapelon, Guy Fagherazzi, Marina Touillaud, Rudolf Kaaks, Anne von Rüsten, Heiner Boeing, Antonia Trichopoulou, Pagona Lagiou, Vassiliki Benetou, Sara Grioni, Salvatore Panico, Giovanna Masala, Rosario Tumino, Silvia Polidoro, Marije F Bakker, Carla H van Gils, Martine M Ros, H Bas Bueno-de-Mesquita, Sanda Krum-Hansen, Dagrun Engeset, Guri Skeie, Amiano Pilar, Maria-José Sánchez, Genevieve Buckland, Eva Ardanaz, Dolores Chirlaque, Laudina Rodriguez, Ruth Travis, Tim Key, Kay-Tee Khaw, Nicholas J Wareham, Malin Sund, Per Lenner, Nadia Slimani, Teresa Norat, Dagfinn Aune, Elio Riboli, and Isabelle Romieu.
• International Agency for Research on Cancer, Lyon, France. ferrarip@iarc.fr
• Am. J. Clin. Nutr. 2013 Feb 1; 97 (2): 344-53.

BackgroundLimited scientific evidence has characterized the association between dietary fiber intake and risk of breast cancer (BC) by menopausal status and hormone receptor expression in tumors.ObjectiveWe investigated the relation between total dietary fiber and its main food sources (vegetables, fruit, cereals, and legumes) and BC risk by using data from the European Prospective Investigation into Cancer and Nutrition (EPIC).DesignA total of 11,576 invasive BC cases in 334,849 EPIC women mostly aged 35-70 y at baseline were identified over a median follow-up of 11.5 y. Dietary fiber was estimated from country-specific dietary questionnaires. Multivariable Cox proportional hazards regression models were used to quantify the association between dietary variables and BC risk with energy adjustment by using the residual method. Subgroup analyses were performed by menopausal status and estrogen receptor (ER) and progesterone receptor (PR) expression in tumors.ResultsBC risk was inversely associated with intakes of total dietary fiber [hazard ratio comparing fifth quintile to first quintile (HR(Q5-Q1)): 0.95; 95% CI: 0.89, 1.01; P-trend = 0.03] and fiber from vegetables (0.90; 0.84, 0.96; P-trend < 0.01) but not with fiber from fruit, cereals, or legumes. Overall, associations were homogeneous by menopausal status and ER and PR expression in tumors. For vegetable fiber, stronger associations were observed for estrogen receptor-negative and progesterone receptor-negative (HR(Q5-Q1):0.74; 95% CI: 0.59, 0.93; P-trend = 0.01) than for estrogen receptor-positive and progesterone receptor-positive tumors (0.92: 0.81, 1.03; P-trend = 0.05), with P-heterogeneity = 0.09.ConclusionDiets rich in dietary fiber and, particularly, fiber from vegetables may be associated with a small reduction in risk of BC, independently of menopausal status.

29 keywords...

hide…