Prediction of lymphatic dissemination in endometrioid endometrial cancer: Comparison of three risk-stratification models in a single-institution cohort.

OBJECTIVES

To compare the performance characteristics of 3 risk-stratification models, referred to as Mayo, Helsinki and Milwaukee models, in predicting lymphatic dissemination in endometrial cancer.

METHODS

A total of 1052 patients with stage I-III endometrioid endometrial cancer were included in the study. The areas under curve were compared with the receiver operating characteristic curve area comparison test. Chi-square and Fisher exact test were used for comparing categorical variables. The Kaplan-Meier method and multivariable Cox regression models were used for survival analyses. The median follow-up time was 55months (range 1-108).

RESULTS

Areas under curve were 0.781, 0.830 and 0.829 for the Mayo, Helsinki (P=0.285 vs. Mayo) and Milwaukee (P=0.292 vs. Mayo) models, respectively, in predicting lymphatic dissemination. The rates of false negatives and false positives were similar for all models. The lymphadenectomy rate decreased in the order of Mayo model (71.5%)>Helsinki model (62.4%)>Milwaukee model (48.8%). In patients with stage I cancer, disease specific survival was better for those who satisfied low-risk criteria according to any of the models. In patients with stage II-III cancer, this difference in survival was significant only for the Milwaukee model. Both lymphatic dissemination and high-risk tumor features as per the risk models were independent predictors of survival.

CONCLUSIONS

The studied models had a similar accuracy in predicting lymphatic dissemination in endometrial cancer. Lymphadenectomy rate was lowest for the Milwaukee model. Survival analyses suggest that variables included in the models predict patient outcome independently of tumor stage.