For several decades, clinical and histologic assessment of various phenotypic properties has provided a basis for treatment planning. However, it is recognized that, preoperatively, clinical assessment identifies only 20% of patients with advanced disease. Furthermore, the variability in intraoperative sampling, the subjectivity and limitations of histologic interpretation, and the variability in response to standardized treatment modalities represent concerns associated with the current treatment of endometrial carcinoma. Presumably, early dissemination, early recurrence, treatment refractoriness and, ultimately, compromised survival are reflections of the inherent biologic characteristics of the tumor. A reasonable assumption is that proscribed molecular events determine various behavioral characteristics of tumors that become manifested at the time of transformation rather than evolving as the tumor volume increases. Therefore, the identification of one or more of these quantifiable molecular variables that directly or indirectly assess tumor biology would assist clinicians in determining patient risk status and in selecting treatment options. As noted, DNA ploidy is an independent, broadly applicable, quantifiable predictor of progression-free survival in patients with endometrial cancer and, therefore, warrants designation as a major prognostic factor or therapeutic determinant. Aneuploidy implies the presence of an abnormal quantity of genomic material and imparts a progressively less favorable prognosis as the DNA index increases. These assayable aberrancies of cellular DNA content presumably reflect the more extreme alterations at the molecular level. Because neoplastic transformation is generally a multistep process, aberrations in several proto-oncogenes or tumor suppressor genes (or both) presumably must be realized before a clinical malignancy develops. A number of genes that encode for various regulatory proteins are overexpressed in endometrial cancer. Whether these aberrancies are fundamental to the pathogenesis of this disease process is unclear. Nevertheless, there appears to be an association between DNA ploidy and the overexpression of several regulatory genes, such as c-fms, K-ras, HER-2/neu, and p53. Although overexpression of these oncogenes and tumor suppressor genes harbor prognostic significance in endometrial cancer, the ploidy status of the tumor appears to represent the most cogent objective variable. As the etiopathogenesis of endometrial carcinoma becomes more discernible, one can envision a limited number of tissue-specific molecular-genetic indices characterizing the risk status of patients. Because the estimated number of deaths from endometrial cancer has doubled since 1987, reassessing of the therapeutic determinants for this disease process is important. The management objective for endometrial cancer by the turn of the century should be the identification of patients at high risk for advanced disease or post-treatment recurrences (or both) at the time of clinical declaration of symptoms and diagnosis. Such pretreatment identification would afford patients at high risk for advanced or recurrent disease access to physicians with special expertise and would facilitate the evaluation and application of new or modified therapeutic modalities. Equally important would be the identification of patients at low risk for untoward outcome, thereby avoiding the cost and morbidity of excessive therapeutic measures.