FIGO staging for carcinosarcoma: can the revised staging system predict overall survival?

Objectives: The purpose of this study is to detect differences in overall survival between the 1988 FIGO staging and current staging of uterine carcinosarcomas to determine if revised 2009 staging accurately predicts actual patient survival.

Methods: From 1988 until 2010, patients with uterine carcinosarcoma were retrospectively identified from tumor registry records. Patients were grouped in both broad stages (1-4) and all FIGO substages in order to detect differences. Time-dependent receiver operating characteristic curves (ROC) were generated to predict death before the end of the second year post-diagnosis for both the new and revised system. Kaplan Meier estimated median survival time was utilized to compare actual patient survival.

Results: Of 112 patients with carcinosarcoma, 37 patients (33%) had FIGO Stage I disease, 15 patients (13.4%) had Stage II disease, 36 patients (32%) were diagnosed as Stage III, and 24 patients (21.4%) had Stage IV disease. 106 of 112 (94.6%) patients underwent lymphadenectomy (pelvic +/- para-aortic). Four patients (3.6%) were downstaged when utilizing broad staging criteria: 2 patients were downstaged from Stage II to I, and 2 patients were downstaged from Stage III to Stage I and II respectively. When looking at substage, the area under the ROC was 0.67 for the former staging system, and 0.65 for the revised staging. Kaplan-Meier estimated median survival time post-diagnosis was 610 days (95% CI [478,930]).

Conclusion: Based upon our reclassification of 112 patients with uterine carcinosarcoma, the revised FIGO staging system does not predict survival more accurately than former staging. Carcinosarcoma has an overall poor prognosis and better indicators of survival are needed.