(P 420) CELLULARITY DEGREES IN PRIMARY MYXOID LIPOSARCOMA: ASSOCIATION WITH TUMOR IMMUNE INFILTRATE AND PATIENT OUTCOMES

Objective: Tumor cellularity stratifies the risk for patients with myxoid liposarcoma (MLPS). However, identifying the highest-risk patients remains challenging. This study aims to i) describe the morphological characteristics in terms of cellularity of myxoid liposarcoma (MLPS), ii) characterize the spatial distribution of the tumor immune infiltrate in MLPS according to the varying degrees of cellularity, and iii) investigate their association with patient outcomes.

Methods: MLPS morphology for patients who underwent surgery for an untreated primary tumor at our institution (2002-2013) was re-evaluated considering classic, round cell, and transitional component on the basis of cellular density (< 5%, 5-95%, >95%, respectively) . The transitional component was categorized as A, B, C according to progressively higher cellularity (mettere le % del lavoro pe A, B, C, altrimenti non si capisce) and lower presence of tumor myxoid stroma within each tumor. In order to describe the spatial distribution of immune cells in the tumor immune microenvironment (TIME), the area showing the highest infiltrate (HI), highest cellularity (HC), the lowest cellularity (LC) and the tumor border on healthy (TBh) or tumor (TBt) sides were sampled. The following markers were studied with immunohistochemistry and digital pathology: CD3, CD8, PD-1, and CD163.

Results: Thirty-five patients were included. All tumors showed the presence of cellularity >5% at histopathological analysis. Classic, transitional and round cell components were detected in variable percentages in 91.4%, 43.9-62.9%, and 41.4% of tumors, respectively. In all cases, more than one component was always present. After a median follow-up of 73 [interquartile range (IQR) 34-101 months], six patients developed disease relapse. These patients were more likely to have either transitional C Type density, or Round cell components, or both (P=0.014). TIME showed different composition according to the spatial distribution of analyzed immune cells sub-populations within each tumor, having HC and TB the highest enrichment of immune cells. Infiltrating immune cells were more frequently detected in tumors with Type C and Round cell type in the tumor areas HC (P=0.032), LC (P=0.019), and TBh (P=0.001).

Conclusion: Patients with a primary MLPS featuring a round cell component in conjunction with a transitional cell component type C faced a higher risk of disease progression and showed higher infiltrating T cells and tumor-associated macrophages. Independent validation of this classification is needed to support this new morphological assessment of MLPS for patient-risk stratification.