The UK incidence of prostate cancer has been increasing in men aged < 60 years. Migrant studies and global and secular variation in incidence suggest that modifiable factors, including a high-fat diet, may contribute to prostate cancer risk. The aim of the present study was to investigate the role of dietary fat intake and its derivatives on early-onset prostate cancer risk. During 1999–2004, a population-based case–control study with 512 cases and 838 controls was conducted. Cases were diagnosed with prostate cancer when ≤ 60 years. Controls were sourced from UK GP practice registers. A self-administered FFQ collected data on typical past diet. A nutritional database was used to calculate daily fat intake. A positive, statistically significant risk estimate for the highest v. lowest quintile of intake of total fat, SFA, MUFA and PUFA was observed when adjusted for confounding variables: OR 2·53 (95 % CI 1·72, 3·74), OR 2·49 (95 % CI 1·69, 3·66), OR 2·69 (95 % CI 1·82, 3·96) and OR 2·34 (95 % CI 1·59, 3·46), respectively, with all P for trend < 0·001. In conclusion, there was a positive statistically significant association between prostate cancer risk and energy-adjusted intake of total fat and fat subtypes. These results potentially identify a modifiable risk factor for early-onset prostate cancer.