OBJECTIVE To examine the risk of prostate cancer associated with foods and nutrients, including individual fatty acids and carotenoids. METHODS Population-based case-control study of 858 men aged <70 years at diagnosis with histologically confirmed prostate cancer of Gleason Grade 5 or greater, and 905 age-frequency-matched men, selected at random from the electoral rolls. Dietary intakes were assessed with a 121-item food frequency questionnaire. RESULTS Inverse associations with prostate cancer were observed for (Odds ratio, OR, 95% confidence intervals, 95% CI for tertile III compared with tertile I) allium vegetables 0.7, 0.5-0.9; p trend 0.01, tomato-based foods 0.8, 0.6-1.0; p trend 0.03 and total vegetables 0.7, 0.5-1.0; p trend 0.04. Margarine intake was positively associated with prostate cancer 1.3, 1.0-1.7; p trend 0.04. The only statistically significant associations observed with nutrients were weak inverse associations for palmitoleic acid ( p trend 0.04), fatty acid 17:1 ( p trend 0.04), and 20:5 n-6 ( p trend 0.05); and a non-significant trend for oleic acid ( p trend 0.09). Neither total, nor beverage-specific, intake of alcohol was associated with risk. CONCLUSIONS Based on these findings, diets rich in olive oil (a source of oleic acid), tomatoes and allium vegetables might reduce the risk of prostate cancer.