At a WHO consensus conference on Early Diagnosis and Prognostic Parameters in Localized Prostate Cancer, a working group discussed the clinical utility of DNA measurements in stages T2 and T3 prostate carcinoma. Incidentally discovered prostate cancer of stage T1 was excluded. The members of the working group, representing various clinical and laboratory disciplines, discussed technical considerations of DNA measurements by flow and image analysis, pretreatment prediction of prognosis, and posttreatment clinical relevance. The group agreed to subdivide tumors into diploid, tetraploid and non-tetraploid aneuploid, expressing various degrees of aggressiveness and gave guidance for the definition of limits of these groups. The panel agreed that knowledge on DNA ploidy prior to treatment is of value in treatment decisions, particularly when surveillance is a treatment option. Aneuploid tumors can be expected to respond very poorly to either irradiation or endocrine therapy, and the presence of aneuploid tumor, either on pretreatment biopsies or in radical prostatectomy specimens, is an ominous sign. The identification of a group of patients with a uniformly poor prognosis should encourage medical oncologists and basic scientists to develop adequate treatment options for this particular group. The panel expressed a strong opinion that DNA ploidy should be uniformly studied in clinical trials, particularly in patients with localized prostate cancer.