Summary: The aim of this study was to compare the World Health Orgazization (WHO), the International Federation of Gynecology (FIGO), and nuclear grading systems and to evaluate their reproducibility, as well as their prognostic impact, in relation to clinical variables and DNA ploidy. The study comprised 266 women with endometrial carcinoma in clinical stages I–IV. Low-risk patients, i.e., patients with grade 1–2 tumors in clinical stage I, were generally not included. In univariate Cox analyses. WHO, FIGO, and especially nuclear grading (p < 0.001), as well as age, stage, and ploidy, were prognostic regarding survival. In the multivariate Cox analyses, WHO and FIGO grades yielded little further independent information beyond nuclear grade. When DNA ploidy was added to the analyses, nuclear grade lost most of its impact because aneuploidy was a powerful factor (p < 0.001) that covaried with nuclear grade. The WHO, FIGO, and nuclear grading systems were found to be subjective, with correlation coefficients of ∼0.60 between different pathologists. Age, stage, and aneuploidy were the most important predictors of outcome in the multivariate analyses. Nuclear grade was a strong prognostic factor in the univariate analyses and a more important indicator than the commonly used histopathological grading systems.
©1996International Society of Gynecological Pathologists