Real-World Management and Survival Outcomes in Patients with Primary or Recurrent Retroperitoneal Liposarcoma: A Single-Center Retrospective Cohort Study

Background/Objectives: Histological subtype and resection completeness are established prognostic factors in retroperitoneal liposarcoma, but evidence often derives from resected primary disease. We characterized management and post-referral survival in a referral cohort. Methods: We retrospectively analyzed 61 patients managed from May 2015 to September 2025. Overall survival (OS) was measured from first presentation at our center. Kaplan–Meier and exploratory Cox analyses were performed, allowing the association between tumor diameter and mortality to vary over time. Results: Forty-two patients had primary disease, and 19 had recurrent disease; 12 had well-differentiated liposarcoma (WDLPS), and 49 had dedifferentiated liposarcoma (DDLPS). Among 24 patients who underwent surgery, 21 had macroscopically complete (R0/R1) resection. Twenty-five patients died. No deaths occurred in WDLPS; median OS was 29.8 months in DDLPS (p = 0.008). In DDLPS, recurrent versus primary disease was not associated with OS (hazard ratio [HR], 0.99; 95% confidence interval [CI], 0.41–2.38). Eastern Cooperative Oncology Group (ECOG) performance status ≥ 2 (n = 6 in the entire cohort) was associated with poorer OS (adjusted HR, 8.03; 95% CI, 2.20–29.36), and the association between tumor diameter and mortality varied over time (adjusted interaction HR, 1.44; 95% CI, 1.07–1.95). Conclusions: Observed post-referral survival differed by histological subtype. Among patients with DDLPS, no survival difference between primary and recurrent disease was demonstrated, but the wide CI and referral-related survivor selection preclude an inference of equivalence. Associations involving performance status and time-varying tumor diameter were exploratory and require external validation.