Differential association between dairy intake patterns and incident prostate cancer: a potential dairy matrix effect.

ABSTRACT

OBJECTIVE: To evaluate the association between dairy intake patterns and the risk of prostate cancer (PC), and its histological differentiation, among men from Mexico City. METHODS: We analyzed the information from 394 incident PC cases paired by age (± 5 years) with 794 population controls. According to the Gleason score at diagnosis, cases were classified as well- (≤ 6), moderately- (= 7), and poorly differentiated PC (≥ 8). Based on a semiquantitative-food frequency questionnaire and using energy-density approach, we estimated the energy-adjusted daily intake of whole milk, cheese (fresh, Oaxaca, and Manchego), cream, and yogurt. Through a principal component analysis, we identified three dairy intake patterns whole milk, cheese, and yogurt. The association between each dairy intake pattern and PC was evaluated from independent nonconditional logistic regression models. We also evaluated the mediator role of calcium and saturated fat intake. RESULTS: After adjustment, a high intake of whole milk pattern was associated with a 63% increased risk of PC (ORhigh vs low 1.63; 95% CI 1.17-2.25, p trend = 0.002); at expenses of moderately (ORhigh vs low 1.77; 95% CI 1.09-2.85, p trend = 0.015) and poorly differentiated PC (ORhigh vs low 1.75; 95% CI 1.05- 2.92, p trend = 0.031). The association was mainly mediated by calcium intake (proportion mediated = 1.17; p < 0.01). No associations were found between cream and yogurt intake patterns with risk of PC, and its histological grade. CONCLUSIONS: A differential association of dairy intake patterns with risk of PC, and the poorly differentiated PC, was identified. This association seems to be determined by different dairy matrices and it is mediated by calcium content. Longitudinal studies are needed to confirm these findings and be able to identify other potential mediators in the etiology of PC.