Major Dietary Factors and Prostate Cancer Risk: A Prospective Multicenter Case-Control Study

Abstract

The association between diet and prostate cancer (PC) risk, although suggestive, still remains largely elusive particularly in the Asian population. This study sought to further evaluate the possible effects of different dietary factors on risk of PC in Iran. Using data from a prospective hospital-based multicenter case-control study, dietary intakes of red meat, fat, garlic, and tomato/tomato products, as well as thorough demographic and medical characteristics, were determined in 194 cases with the newly diagnosed, clinicopathologically confirmed PC and 317 controls, without any malignant disease, admitted to the same network of hospitals. Odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were obtained after adjustment for major potential confounders, including age, body mass index, smoking, alcohol, education, occupation, family history of PC, and total dietary calories. Comparing the highest with the lowest tertile, a significant trend of increasing risk with more frequent consumption was found for dietary fat (OR: 1.79, 95% CI: 1.71–4.51), whereas inverse association was observed for tomato/tomato products (OR: 0.33, 95% CI: 0.16–0.65). A nonsignificant increase in PC risk was revealed for dietary red meat (OR: 1.69, 95% CI: 0.93–3.06). For garlic consumption, a borderline reduction in risk was observed (OR: 0.58, 95% CI: 0.32–1.01; P = 0.05). In conclusion, our study supports the hypothesis that total fat may increase PC risk and tomatoes/tomato products and garlic may protect patients against PC.

ACKNOWLEDGMENTS

The authors would like to thank the nursing, secretary, and administrative staff of the Urology Research Center, Sina Hospital, especially Mrs. G. Abdi for her excellent cooperation in the data entry. Patient recruitment has been the dedicated collaboration of the following contributors: Dr. M. Lotfi, Dr. MA. Amirzargar, Dr. H. Mazdak, Dr. A. Roshani, and Dr. A. Kheirollahi. We are also grateful to Ms. M. Tayebi for her helpful assistance in preparation of the manuscript and Ms. M. Chamari for her valuable collaboration in nutritional evaluation. This study was supported in part by a grant from the Tehran University of Medical Sciences (No. 2812).

Notes

¹The sum does not add up to the total because of some missing values.

²Family history of prostate cancer in 1st degree relatives.

^†Fisher's exact test.

^‡Chi-square test, * independent samples t test.

^aEstimated using multiple logistic regression model, adjusted for total dietary calories.

^bEstimated using multiple logistic regression model, adjusted for total dietary calories, age, body mass index, occupation, educational level, smoking, alcohol, and family history of PC.

^†Food intake was divided approximately into tertile-based percentile distributions among population control subjects.

^‡Referent group.