Postoperative nomogram for survival of patients with retroperitoneal sarcoma treated with curative intent.

BACKGROUND Current American Joint Committee on Cancer retroperitoneal sarcoma (RPS) staging is not representative of patients with RPS specifically and has limited discriminative power. Our objective was to develop a RPS disease-specific nomogram capable of stratifying patients based on probability of overall survival (OS) after resection. PATIENTS AND METHODS In all, 1118 RPS patients were evaluated at our institution (1996-2006). Patients with resectable, nonmetastatic disease were selected (n = 343) and baseline, treatment and outcome variables were retrieved. A nomogram was created and its performance was evaluated by calculating its discrimination (concordance index) and calibration and by subsequent internal validation. RESULTS Median follow-up and OS were 50 and 59 months, respectively. Independent predictors of OS were included in the nomogram: age (> or = 65), tumor size (> or = 15 cm), type of presentation (primary versus recurrent), multifocality, completeness of resection and histology. The concordance index was 0.73 [95% confidence interval (CI) 0.71-0.75] and the calibration was excellent, with all observed outcomes within the 95% CI of each predicted survival probability. CONCLUSIONS A RPS-specific postoperative nomogram was developed. It improves RPS staging by allowing a more dynamic and robust disease-specific risk stratification. This prognostic tool can help in patient counseling and for selection of high-risk patients that may benefit from adjuvant therapies or inclusion into clinical trials.