Dietary inflammatory index and incidence of breast cancer in the SUN project

Autores: Gardeazábal González, Itziar; Ruiz-Canela, Miguel (Autor de correspondencia); Sánchez Bayona, Rodrigo; Romanos Nanclares, Andrea; Aramendía Beitia, José Manuel; Shivappa, N.; Hebert, J. R.; Martínez González, Miguel Ángel; Toledo Atucha, Estefanía
Título de la revista: CLINICAL NUTRITION
ISSN: 0261-5614
Volumen: 38
Número: 54
Páginas: 2259 - 2268
Fecha de publicación: 2019
Lugar: WOS
Background & aims: Breast cancer (BC) is the most commonly diagnosed cancer, and diet is suspected to play a role in its development. Dietary factors may mediate this process through modulation of inflammation, though findings from previous studies have not been consistent. We aimed to longitudinally assess the association between the dietary inflammatory index (DII (R)), a frequently used method to assess the inflammatory potential of the diet, and incident BC. Methods: We included 10,713 middle-aged, Spanish female university graduates from the SUN cohort. DII (R) scores were derived from a validated 136-item food-frequency questionnaire, and it was based on scientific evidence on the relationship between diet and inflammatory biomarkers. Diagnosis of BC was reported by the participant or, if deceased, by the next of kin or identified from death certificates. Self-reports of BC were confirmed by revision of medical reports by an experienced oncologist. Cox proportional hazard models were used to estimate multivariable-adjusted hazard ratios (HR) and 95% confidence intervals (CI) for the association between quartiles of DII (R) and incident BC. Results: After 10.3 years of median follow-up, we identified 100 confirmed and 168 probable incident BC cases. The multivariable-adjusted HR for participants in the 4th quartile to the 1st quartile was 1.44 (95% CI 0.76-2.72; p-trend: 0.339) when confirmed cases were analyzed, and 1.20 (95% CI 0.72-1.99; p-trend: 0.757) for the probable cases. We neither observed statistically significant differences in regard to menopausal status. Conclusions: The apparent increase in risk between DII (R) scores and BC in our cohort was not statistically significant, which could be partly explained by the small number of observed cases. (C) 2018 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.