Abstract
OBJECTIVE: This study aimed to identify prognostic factors affecting progression-free survival (PFS) in patients with retroperitoneal liposarcoma (RPLS) after R0 resection. METHODS: A retrospective analysis was conducted on 183 RPLS patients who underwent R0 resection, evaluating general information, clinicopathological data, laboratory parameters, and follow-up outcomes. Based on follow-up outcomes, patients were categorized into progression-free survival (PFS) group (PFS, n=121) and disease progression group (DP, n=62). The general information, clinicopathological data, and laboratory parameters of the two groups were systematically compared, with statistically significant factors subsequently incorporated into the Cox multivariate regression analysis. Significant prognostic factors identified through Cox analysis were further evaluated using Kaplan-Meier (K-M) survival analysis, serving as the foundation for constructing the predictive model. The model's performance was rigorously assessed for 1-year, 3-year, and overall PFS prediction. RESULTS: Multivariate analysis identified age (HR=1.034, 95% CI=1.011-1.057, P=0.003), histologic subtype (Well-differentiated liposarcoma as reference. Dedifferentiated liposarcoma: HR=0.130, 95% CI=0.029-0.578, P=0.007. Myxoid/round cell liposarcoma: HR=0.190, 95% CI=0.084-1.635, P=0.190. Pleomorphic liposarcoma: HR=0.176, 95% CI=0.036-0.865, P=0.032. Mixed-type liposarcoma: HR=0.793, 95% CI=0.157-4.008, P=0.799), tumor stage (HR=0.440, 95% CI=0.257-0.755, P=0.003), and tumor differentiation grade (HR=0.395, 95% CI=0.236-0.661, P<0.001) as independent risk factors. The predictive models demonstrated excellent discriminative ability: 1-year model (AUC=0.944, NB=0.05-0.80), 3-year model (AUC=0.861, NB=0.05-0.73), the overall mode (AUC=0.903, NB=0.03-0.90). CONCLUSION: Advanced age, DDLPS, PLS, stage III-IV disease, and poor tumor differentiation were identified as independent predictors of shorter PFS in RPLS patients following R0 resection.