Outcome Prediction in Primary Resected Retroperitoneal Soft Tissue Sarcoma: Histology-Specific Overall Survival and Disease-Free Survival Nomograms Built on Major Sarcoma Center Data Sets

Abstract

Integration of numerous prognostic variables not included in the conventional staging of retroperitoneal soft tissue sarcomas (RPS) is essential in providing effective treatment. The purpose of this study was to build a specific nomogram for predicting postoperative overall survival (OS) and disease-free survival (DFS) in patients with primary RPS. Data registered in three institutional prospective sarcoma databases were used. We included patients with primary localized RPS resected between 1999 and 2009. Univariate (Kaplan and Meier plots) and multivariate (Cox model) analyses were carried out. The a priori chosen prognostic covariates were age, tumor size, grade, histologic subtype, multifocality, quality of surgery, and radiation therapy. External validation was performed by applying the nomograms to the patients of an external cohort. The model's discriminative ability was estimated by means of the bootstrap-corrected Harrell C statistic. In all, 523 patients were identified at the three institutions (developing set). At a median follow-up of 45 months (interquartile range, 22 to 72 months), 171 deaths were recorded. Five- and 7-year OS rates were 56.8% (95% CI, 51.4% to 62.6%) and 46.7% (95% CI, 39.9% to 54.6%. Two hundred twenty-one patients had disease recurrence. Five- and 7-year DFS rates were 39.4% (95% CI, 34.5% to 45.0%) and 35.7% (95% CI, 30.3% to 42.1%). The validation set consisted of 135 patients who were identified at the fourth institution for external validation. The bootstrap-corrected Harrell C statistics for OS and DFS were 0.74 and 0.71 in the developing set and 0.68 and 0.69 in the validating set. These nomograms accurately predict OS and DFS. They should be used for patient counseling in clinical practice and stratification in clinical trials.