Predictors and outcomes of recurrent retroperitoneal liposarcoma with multiple tumors

Background: Retroperitoneal liposarcoma (RLS) is a rare but severe disease. Repeated postoperative recurrence with multiple tumors is a therapeutic dilemma. The clinical outcomes and survival predictors of recurrent RLS with multiple tumors remain to be explored. Methods: Patients with recurrent RLS were retrospectively analyzed. Univariate and multivariate analysis was performed to find independent prognostic factors that were correlated with Overall survival (OS) or progression-free survival (PFS). Factors significant in univariate analysis were further included into multivariate Cox proportional hazards regression model. The nomogram model was built to predict the survival status of patients. Variables that were significant in multivariable analysis were added to the internally validated nomogram models. The analysis of OS and PFS was performed by Kaplan-Meier analysis and log-rank test. Results: A total of 113 recurrent RLS patients with multiple tumors were enrolled in the study. The 1-, 3-, and 5-years OS (PFS) rates were 70.7% (76.1%), 35.9% (76.1%), and 30.9% (76.1%), respectively. Univariate and multivariate analyses showed that number of surgeries, resection methods, tumor size, status of pathological differentiation, pathological subtypes, and recurrence patterns were important prognostic factors for OS or PFS (each p < 0.05). Nomogram models were established to efficiently predict the prognostic status of patients. Patients with the local recurrence (LR) pattern had a poor prognosis and would derive no survival benefit from combined organ resection and R0/R1 resection (each p < 0.05). Conclusion: RLS patients recurrence with multiple tumors had a poor prognosis. Those patients should be followed up more frequently after surgery. The strategies of aggressive resection may not improve the survival of patients with LR pattern in the retroperitoneum. Prognostic factors in the efficient nomogram models should be considered in the individualized clinical management of recurrent RLS with multiple tumors.