Previous studies of diet and prostate cancer have focused on advanced disease and have suggested a positive association with saturated fat intake. We report a study assessing the relationship between diet and preclinical prostate cancer. A total of 215 men with preclinical prostate cancer and 593 controls with no evidence of cancer participated in a case-control study conducted in Quebec City between October 1990 and May 1993. The study population comprised two groups: men treated surgically for benign prostatic hypertrophy and participants in a prostate cancer screening program. Trained nutritionists interviewed the participants on their usual diet using a diet history questionnaire. Odds ratios for prostate cancer associated with quartiles of dietary intake and P values for trend were estimated by logistic regression while controlling for age, education, group, and family history of prostate cancer. A positive association was observed between total energy intake and preclinical prostate cancer (p = 0.004). The odds ratios for prostate cancer increased with each quartile of energy intake: 1.00, 1.77, 1.90, and 2.67. After adjustment for energy, nutrients were not associated with prostate cancer. This study provides some evidence that total energy intake is related to preclinical prostate cancer and suggests that diet could be involved earlier than thought in the occurrence of prostate cancer.