Development and Validation of Tumor‐size‐stratified Prognostic Nomograms for Patients with Uterine Sarcoma: A SEER Database Analysis

Abstract Background Tumor‐size‐stratified analysis on the prognosis of uterine sarcoma is insufficient. This study aimed to establish the tumor‐size‐stratified nomograms to predict the 3‐ and 5‐year overall survival (OS) of patients with uterine sarcoma. Methods The data analyzed in this study were obtained from the Surveillance, Epidemiology, and End Results (SEER) database. We collected data from patients with uterine sarcoma diagnosed between 2004 and 2015. According to the median tumor size of 7.8 cm, the enrolled patients were divided into two tumor size (TS) groups: TS <7.8 cm and TS ≥7.8 cm. Patients in each group were randomly divided into the training and validation cohorts with a ratio of 7:3. Chi‐square test was used to compare differences between categorical variables. Multivariate Cox regression models were used to identify significant predictors. We calculated the concordance index (C‐index) and the area under the receiver operating characteristics curve (AUC) to validate the nomograms. Results Compared with TS <7.8 cm group, TS ≥7.8 cm group had more patients of 45–64 years group, higher black race prevalence, higher proportion of myometrium tumor, higher stage, and higher grade; In the TS <7.8 cm training cohort, six variables (age, race, marital status, tumor primary site, stage, and grade) were identified as significantly associated with OS in multivariate analysis. However in the TS ≥7.8 cm training cohort, only four variables (surgery on primary site, tumor size, stage, and grade) were significantly identified; The C‐index of two nomograms were 0.80 and 0.73 in training cohorts, respectively, and the AUC values for 3‐ and 5‐year OS predictions in training cohorts were all above 0.80. Similar results were observed in validation cohorts. Conclusions This study found that the significant prognostic factors were different between two tumor size groups of uterine sarcoma patients. The tumor‐size‐stratified nomograms, which we constructed and validated, might be useful to predict the probability of survival for patients with uterine sarcoma.