Abstract
Benign lipomas and well-differentiated liposarcomas share many histological and molecular features. Due to their similarities, patients with these lipomatous tumors are misdiagnosed up to 40% of the time following radiological detection, up to 17% of the time following histological examination, and in as many as 15% of cases following fluorescent in situ hybridization for chromosomal anomalies. Incorrect classification of these two tumor types leads to increased costs to the patient and delayed accurate diagnoses. In this study, we used genomics analysis to identify several genes whose mRNA expression patterns were significantly altered between lipomas and well-differentiated liposarcomas. We confirmed our findings at the protein level using a panel of 30 human lipomatous tumors, revealing that C4BPB, class II, major histocompatibility complex, CIITA, EPHB2, HOXB7, GLS2, RBBP5, and regulator of RGS2 protein levels were increased in well-differentiated liposarcomas compared to lipomas. We developed a multi-protein model of these markers to increase discriminatory ability, finding the combined expression model with CIITA and RGS2 provided a high ability (AUC=0.93) to differentiate between lipomas and well-differentiated liposarcomas with sensitivity at 83.3% and specificity at 90.9%.