Background: A nomogram to predict disease-specific mortality (DSM) following surgery for soft tissue sarcoma (STS) has been developed by the Memorial Sloan Kettering Cancer Center (MSKCC). The goal of this study was to validate this nomogram by assessing discrimination and calibration at the population level using a national cancer database.
Methods: Retrospective review of the Surveillance, Epidemiology, and End Results cancer registries identified patients undergoing surgery for STS from 1988 to 2011. Data for patient age, tumor size, tumor grade, histologic subtype, sex, primary tumor location, and tumor depth were entered into the nomogram calculator for each patient. Discrimination was quantified using a concordance index. Calibration was assessed by comparing quintiles of nomogram-predicted probabilities of disease-specific mortality (DSM) with American Joint Committee on Cancer (AJCC) stage DSM.
Results: Overall, 9237 patients were identified with complete information needed for the nomogram. With a mean follow-up of 45 months, the concordance index for nomogram-predicted DSM with actual DSM was 0.74 for the entire cohort. For low- and high-grade tumors, this was 0.71 and 0.66, respectively. Kaplan-Meier curves showed better calibration for nomogram-predicted DSM when compared with AJCC staging.
Conclusions: Our results validate the use of the MSKCC STS nomogram in the general population, with better predictive ability than AJCC staging. However, a concordance index of 0.74 suggests that further improvement in prognostication is needed, perhaps with biological markers or additional clinical variables.