The Society of Surgical Oncology, inc.
The American Society of Breast Surgeons.
Annals of Surgical Oncology

Log in | Register

The Memorial Sloan Kettering Cancer Center Nomogram is More Accurate than the 2009 FIGO Staging System in the Prediction of Overall Survival in a German Endometrial Cancer Patient Cohort

Alexandra Huss, Gabriele Ihorst, Sylvia Timme-Bronsert, Annette Hasenburg, Martin K. Oehler, Maximilian Klar
Gynecologic Oncology
Volume 25, Issue 13 / December , 2018



Despite the complexity of endometrial cancer (EC) tumor biology, treatment decisions are still mainly based on the post-surgical International Federation of Gynecology and Obstetrics (FIGO) stage. Prediction models considering more prognostic factors may represent a better risk assessment than FIGO stage alone. We tested the Memorial Sloan Kettering Cancer Center (MSKCC) nomogram for the prediction of overall survival (OS) in a German EC population.


Overall, 454 EC patients (322 type I and 132 type II) who received primary surgical treatment at our department between 1991 and 2011 were included in the analysis with a dataset of 68 covariates. Predicted OS was calculated using the online MSKCC nomogram and compared with the observed survival in our population. To estimate the discriminatory power, the concordance probabilities were calculated using the concordance probability estimate (CPE). Receiver operating characteristic curves were created and the area under the curve (AUC) values compared between predicted and actual OS.


After a mean follow-up of 183 months, 211 patients were reported dead (47%). Mean OS for all stages was 101 months (standard deviation 66.7 months). The 2009 FIGO system showed an AUC value of 0.6 and a CPE of 0.63, while the 3-year OS prediction of the MSKCC nomogram showed an AUC value of 0.8 and a CPE of 0.77.


This external validation of the MSKCC nomogram showed better discrimination and calibration values than the conventional FIGO classification system. The nomogram was externally validated and can serve as a tool for better risk-adapted treatment decisions and patient stratification, e.g. in clinical trials.

Add a comment

0 comment(s)


Follow the journal on Twitter and Facebook

Help to expand the reach of the journal to support the research and practice needs of surgical oncologists and their patients.