OBJECTIVE: High-risk prostate cancer patients are at risk for subclinical disease and micro-metastasis at the time of treatment. Nowadays, tight margins reduce dose to periprostatic areas compared to earlier techniques. We investigated whether rectangular fields were associated with fewer failures compared to conformal fields (with lower extraprostatic dose). METHODS: We selected 164 high-risk patients from the trial population of 266 T1-T4N0M0 patients, randomized between rectangular (n=79) and conformal fields (n=85). Prescribed dose was 66 Gy to the prostate and seminal vesicles plus 15 mm margin. We compared clinical failure rates (in- and excluding local failures), between both arms. Dose differences around the prostate were calculated based on an inter-patient mapping method. RESULTS: Median follow-up was 34 months. There were 9 clinical failures in the rectangular arm versus 24 in the conformal arm (p=0.012). Number of failures outside the prostate was 7 and 19, respectively (p=0.025). We observed average dose differences of 5-35 Gy between the arms in the regions around the prostate. CONCLUSIONS: We found a significantly lower risk of early tumor progression for patients treated with rectangular fields. Treatment failure can probably in part be prevented by irradiation of areas suspected of subclinical disease in high-risk prostate cancer.