Uterine prognostic indicators in low-grade endometrial cancers: anatomical and immunohistochemical findings discovered using an “end-result-driven” statistical approach.

Abstract

Objectives : It is useful to know which low-grade endometrial adenocarcinomas will likely display evidence of lymph node metastases or extrauterine disease either at hysterectomy or on follow up. In order to model this situation, we used an end-result-driven statistical assessment of pathology findings and related them to known case outcomes with the objective of separating cases with “good” from those with “poor” outcomes, and to determine measurable anatomical/ immunohistochemical factors associated with “good” or “poor” prognosis. Methods : We assessed anatomical, immunohistochemical and clinical data of 150 cancer hysterectomies. 119 women had no lymph node metastasis at the time of cancer hysterectomy and no evidence of recurrent disease on 12- 89 months follow-up. 31 women had either positive lymph node dissections at the time of cancer hysterectomy (N = 11) or disease persistence/recurrence over the same follow up interval. Results : We found 5 significant features (p<0.05) that separated “good” from “poor” clinical behavior: (1) Subclonal p53 immunostaining (combination of normal with one or more abnormal IHC-staining patterns, a here-to-fore unrecognized marker of “good-prognosis”), MELF-glands, >70% uterine wall invasion, lymphovascular space invasion (LVI), and stromal tumor-invasion of the uterine cervix. The latter four factors are historically established indicators of “poor-prognosis”. When applied as sorting-premises to our data set, these 5 indicators provided a 52%-to-48% split between potentially “good” and “poor” case outcomes, with all positive lymph node dissections and patient deaths relegated to “poor-prognosis” cases. Conclusions : We conclude that subclonal p53 immunostaining (most likely an indicator of POLε or MMR-deleted cancers), MELF-glands, >70% uterine wall invasion, LVI, and tumoral invasion of the uterine cervix have significant prognostic value in determining lymph node involvement and clinical case outcome. We argue that these five findings should be routinely listed in the pathology report and that p53 immunohistochemical staining should routinely be used as an adjunct test in the pathological assessment of low-grade endometrial adenocarcinomas.