The identification of prognostic variables is an important aspect of managing and counseling women with endometrial adenocarcinoma. The surgical stage, age, cell type, depth of myometrial invasion, and histologic grade have all been previously demonstrated to be related to prognosis. Several reports have indicated that tumor ploidy as determined by flow cytometry with fresh or fixed cells removed from paraffin blocks of endometrial adenocarcinomas can contribute to the assessment of prognosis. To verify the significance of DNA content in endometrial adenocarcinoma, we conducted an historical cohort study on a subgroup of women from a Gynecologic Oncology Group (GOG) protocol of early clinical stage disease. Flow cytometry was performed at one facility on cells extracted from blocks obtained from several GOG member institutions. Blocks were submitted for 293 of 933 eligible patients. Ninety-two histograms were of good quality, with 55 interpreted as diploid and 37 as aneuploid. One hundred sixty-two histograms were technically suboptimal, of which 137 were considered probably diploid, 13 probably aneuploid, and 12 unacceptable due to high background noise. Of the commonly accepted prognostic variables, only depth of invasion was significantly related to the ploidy status. There was no discernable difference in survival between patients with diploid and patients with probable diploid and probable aneuploid tumor types. Incorporation of the flow cytometry data into a proportional hazards regression model adjusted for age and surgical stage revealed a significant increased risk of disease-related death (relative risk, 4.1; 95% confidence interval, 2.3 to 7.3) for patients with aneuploid tumor type as compared to patients with diploid tumor type. This study confirms the prognostic significance of ploidy determination by flow cytometry and also indicates some of the difficulties of retrospectively applying this technology to cooperative group studies.