Serguienko A, et al. - Since the final differential diagnosis to differentiate between more aggressive and less aggressive form of liposarcoma is based on post-surgical histological examination and there is a lack of molecular biomarkers for early detection, therefore, researchers investigated genes that most precisely can predict dedifferentiated liposarcoma (DDLPS) differential diagnosis. They used atypical lipomatous tumor (= 8), well-differentiated liposarcoma (= 9) and DDLPS (= 20) samples to analyze 11 metabolic genes implicated in general and adipose tissue-specific metabolism. A three-gene signature based on PNPLA2, LIPE and PLIN1 was established. By employing this metabolic signature, a highly accurate differential diagnosis between WD (well-differentiated)- and DDLPS was achieved even in samples comprising lipid droplets, a marker of differentiation, which ensures its extreme suitability for use on biopsies. In 45% of DDLPS samples examined under TCGA project, the deletion of the PNPLA2 gene was noted, and a link was identified between the deletion and significantly lower PNPLA2 expression level. The likely implicated molecular mechanisms are unveiled as far as the pathogenesis of the disease is concerned, and a recent finding that deletion of PNPLA2 is a new factor in liposarcoma progression is supported.