The nuclear DNA content of prostate cancer specimens, both needle biopsies and aspiration biopsy specimens as well as transurethral resection (TUR) chips and radical prostatectomy specimens, can now be reliably measured by standardized methods of flow and static image cytometry. For prostate carcinomas of every clinical stage (A1-D2), DNA diploid tumors have a better prognosis than tumors of a similar stage and grade which are non-diploid. Of particular importance to this symposium is the fact that DNA diploid stage D1 and D2 tumors treated early by androgen deprivation generally have a remarkably good prognosis. In contrast, those patients with DNA non-diploid tumors progress early despite androgen deprivation. Such a result suggests that DNA ploidy can be used to identify prostate cancers which are potentially sensitive to hormonal manipulation. Additional investigations from several groups indicate that early stage prostate malignant lesions, for example stages A1, A2, B1, and B2, are generally DNA diploid (about 75%). Swedish data suggest a steady progression of prostate cancer from early diploid to tetraploid, to non-tetraploid aneuploid, to multiple stemline aneuploid tumors with time and advancing stage. Taken together, these data suggest that the earliest detectable prostate carcinomas should be overwhelmingly DNA diploid. A large majority of these patients with early tumors should be candidates for "chemoprevention" by pharmacologic methods which reduce the effective androgen stimulation of prostate tumor cells.