BioScience Trends. 2018;12(2):185-192. (DOI: 10.5582/bst.2018.01015)

Infiltration characteristics and influencing factors of retroperitoneal liposarcoma: Novel evidence for extended surgery and a tumor grading system.

Wang Z, Wu JH, Lv A, Li CP, Li ZW, Zhao M, Hao CY


This study sought to evaluate the infiltration tendency of retroperitoneal liposarcoma (RPLS) from a new pathological angle by exploring the infiltration characteristics, which could provide helpful information to facilitate surgical decision-making and prognosis prediction. Concurrently, we aim to identify significant indicators of infiltration. A total of 61 consecutive patients with RPLS at our institution were retrospectively analyzed. All patients received extended surgery. The tumor infiltration characteristics and influencing factors were studied based on the pathological diagnosis. Univariate and multivariate analyses of organ infiltration (OI) and surrounding fat infiltration (SFI) were performed. OI was found in 95 (28.5%) resected organs from 39 (60.7%) patients, and SFI was found in 119 (35.7%) resected organs from 47 (77%) patients. The tumor infiltrated the serosal layer of 13 organs (13/37, 35.1%), the muscularis layer of 18 organs (18/37, 48.6%) and the submucosa of 6 organs (6/37, 16.2%). The percentage of lipoblasts and the rates of necrosis and mitosis were all significantly higher in high-grade tumors (dedifferentiated, round cell, and pleomorphic). A high lipoblast percentage (≥ 20%) was the only independent risk factor for OI. A recurrent tumor and a high-grade tumor were independent risk factors for SFI. In conclusion, RPLS has a high infiltration tendency, such that it frequently infiltrates organs and surrounding fat tissue. Therefore, extended resection of the tumor and the adjacent organs is recommended. The percentage of lipoblasts was associated with the tumor grade and infiltration characteristics; thus, lipoblast percentage may become a new grading factor for RPLS.

KEYWORDS: Retroperitoneal sarcoma, liposarcoma, infiltration

Full Text: