Carcinosarcoma of the Uterus: A Study From the Surveillance Epidemiology and End Result (SEER) Database

Uterine cancer (UC) is one of the leading gynecologic neoplastic disorders in the United States (US), of which over 80% are endometrioid adenocarcinomas (EA). In contrast to EA, carcinosarcoma (CS) of the uterus is a sporadic and highly malignant tumor, phylogenetically containing both epithelial and mesenchymal histologic elements. This study sought to analyze demographic, pathological retrospectively, and survival characteristics of a large cohort of CS patients compared to EA patients to identify prognostic factors and treatment approaches that may improve the current clinical management of CS patients.

Demographic and clinical data were abstracted from 88,530 patients diagnosed with uterine malignancy from the Surveillance, Epidemiology, and End Results (SEER) database for 38 years (1973-2010). Extracted variables were analyzed using the Chi-square test, paired t-test, and multivariate analysis, while Kaplan-Meier functions were used to compare survival between groups. Statistical analyses were performed with IBM Statistical Product and Service Solutions (SPSS^©), version 20.2 (IBM Corp., Armonk, NY).

A total of 3,706 cases of CS comprised 38.2% of uterine sarcomas (n=9,702), and 4.1% of uterine cancers overall (n=88,530). EA made up 88.6% (n=78,481) of all uterine cancers. CS patients presented later in life (68.3±11.5 years) than EA (61.9±12.5 years). 65.2% of CS and 77.8% of EA occurred in Caucasians. The incidence (per million) of EA was higher in Caucasians compared to African-Americans (AA) (41% vs. 26.8%), while the incidence of CS was higher among AA than Caucasians (4% vs. 1.9%, p<0.001). 33.4% of CS was poorly differentiated at presentation, compared to 13.1% of EA. 27.8% of CS patients presented with a distant disease compared to only 4.7% of EA patients. 29.9% of AA patients with CS presented with metastatic disease, compared to 28.2% of Caucasian patients (p<0.001). Mean survival for CS patients (6.6±0.2 years) was significantly lower than that of EA patients (17.7±0.7 years, p<0.001), and AA CS patients had significantly lower survival than Caucasians CS patients (4.5±0.4 years vs. 7.1±0.3 years, p<0.001). CS patients treated with combined surgery and radiotherapy had the highest survival (9.4±0.5 years, p<0.001), while EA patients treated with surgery alone had the highest survival (20.4±1.2 years, p<0.001). Survival among AA CS patients treated with combination therapy was significantly inferior compared to Caucasians (6.5±0.6 years vs. 9.8±0.5 years, p<0.001). Multivariate analysis identified CS histology (odds ratio [OR] 1.9, CI=1.7-2.1), AA race (OR 1.3, CI=1.2-1.4), age over 40 (OR 3.4, CI=2.9-4.1), undifferentiated grade (OR 3.0, CI=2.6-3.4), and distant metastases (OR 6.2, CI=5.8-6.8) as independently associated with increased mortality (p<0.005). The use of radiotherapy in CS patients was independently associated with decreased mortality (OR 0.1, CI=0.02-0.6, p<0.005).

Uterine CS is a highly malignant tumor with a significantly worse prognosis than EA. AA has a considerably higher CS incidence compared to EA. Moreover, AA CS had higher tumor grades, higher rates of metastatic disease, and experienced significantly lower overall survival compared to Caucasians despite receiving similar therapy. Primary radiotherapy or combination radiotherapy confers a survival advantage to AA uterine CS patients.