Clinicopathologic Characteristics and Clinical Outcome of Localized Liposarcoma: A Single-Center Experience over 25 Years and Evaluation of PD-L1 Expression.

PURPOSE: For liposarcoma (LPS), clinical course and proper treatment strategies have not been well-established. Recently, immune-checkpoint inhibitors have shown potential efficacy in LPS. We aimed to describe the clinical course of LPS and evaluate the clinical impact of programmed death-ligand 1 (PD-L1). MATERIALS AND METHODS: We reviewed all consecutive patients (n=332) who underwent curative-intent surgery for localized LPS at Asan Medical Center between 1989 and 2017. PD-L1 testing was performed in well-differentiated and dedifferentiated LPS. RESULTS: The median age was 56 years with males comprising 60.8%. Abdomen-pelvis (47.6%) and well-differentiated (37.7%) were the most frequent primary site and histologic subtype, respectively. During a median follow-up of 81.2 months, recurrence was observed in 135 (40.7%), and 86.7% (117/135) were loco-regional. Well-differentiated subtype (hazard ratio [HR], 0.38), abdomen-pelvis origin (HR, 2.43), tumor size larger than 5 cm (HR, 1.83), positive resection margin (HR, 2.58), and postoperative radiotherapy (HR, 0.36) were significantly related with recurrence-free survival as well as visceral involvement (HR, 1.84) and multifocality (HR, 3.79) in abdomen-pelvis LPS. PD-L1 was positive in 31.5% (23/73) and 51.3% (39/76) of well-differentiated and dedifferentiated LPS, respectively, but had no impact on survival outcomes. CONCLUSION: Clinical course of LPS was heterogeneous according to histology and anatomic location. Clear resection margin was important to lower recurrence and postoperative radiotherapy might have additional benefit. A decent portion of well-differentiated and dedifferentiated LPS were positive for PD-L1, but its prognostic role was unclear. Further research is needed to determine clinical implications of PD-L1, especially for advanced-stage LPS with unmet needs for effective systemic treatment.