Dietary flavonoid patterns and prostate cancer: evidence from a Mexican population-based case-control study.

ABSTRACT

Flavonoids are a broad group of bioactive compounds with anticarcinogenic effects on the prostate that have been scarcely evaluated in Latin American populations. Our objective was to evaluate the association between dietary patterns of flavonoid intake and prostate cancer (PC) in a population-based case-control study carried out in Mexico City. Based on a semi-quantitative FFQ with a frame reference of 3 years before diagnosis or interview, we used an updated database for estimating the daily intake (mg/d) of flavones, flavonols and flavanols for 395 confirmed incident PC cases and 797 population controls matched by age (± 5 years). Histological PC differentiation was evaluated using the Gleason score at diagnosis. Flavonoid dietary intake patterns (FDIP) were determined through principal component analysis, and their association with PC was estimated using logistic regression models. Three FDIP were identified gallate pattern (GP) characterised by (-)-epicatechin-3-O-gallate, (-)-epigallocatechin-3-O-gallate and (+)-gallocatechin; luteolin pattern (LP) characterised by luteolin and (-)-epigallocatechin-3-O-gallate; and a mixed pattern (MP) that included (+)-catechin, (-)-epicatechin and quercetin. A higher GP (ORT3 v.T1 = 0·47; 95 % CI 0·33, 0·66) and LP intake (ORT3 v. T1 = 0·39; 95 % CI 0·27, 0·59) were associated with a decreased PC likelihood. In contrast, a higher MP intake (ORT3 v. T1 = 2·32; 95 % CI 1·67, 3·23) increased PC likelihood. The possible differential and synergistic anticarcinogenic role of flavonoid compounds in PC deserves further study.