54. Genome-wide copy number abnormalities in lipomatous neoplasms detected with chromosomal microarray: Is gain of MDM2 necessary for 12q amplification characterization and prognostication?

Abstract

Lipomatous neoplasms (LN) are a heterogeneous group of adipocytic tumors, including benign lipomas, low-grade atypical lipomatous tumors (ALT), and high-grade liposarcomas [sub-grouped into de-differentiated liposarcomas (DDLS) evolved from ALT and pleomorphic liposarcomas]. Using histological (H&E) approaches, ALT are often difficult to distinguish from benign lipomas, while DDLS are challenging to differentiate from other high-grade sarcomas. Given this diagnostic challenge, molecular signatures can aid in the characterization/prognostication of these tumors and have shown MDM2 and/or CDK4 amplification in ALT and DDLS tumors, with MDM2 being recognized as the likely “driver” gene, based on fluorescence in situ hybridization (FISH) studies. We describe an atypical lipomatous tumor having features of spindle cell lipoma. A genome-wide copy number microarray (MA) assessment showed amplification on 3q (6-10 copies; 4 contiguous regions) and 12q (6-11 copies; 7 contiguous regions), with the 12q amplification regions being juxtaposed to, but not encompassing, the MDM2 gene. A FISH test for this specimen, using a commercial probe, yielded an “MDM2” positive result (albeit with smaller signals). This “discrepant” finding is not unexpected since the 209 kb FISH probe encompasses loci other than MDM2, including the CPM gene (which is amplified in the MA). To our knowledge, this is the first documented case of 12q amplification in a LN that does not encompass the MDM2 gene. Moreover, in our experience (n= 44 tumors), 100% of the 12q (MDM2/CPM) amplified LN have shown additional imbalances using MA methods [including, but not limited to, amplifications of 1q (60%); CDK4 and/or HMGA2 (58%), & HMGA2 (25%)]. In summary, our data underscore the advantages of using MA to evaluate LN. Moreover, this rare tumor having 12q15 amplification that does not encompass the MDM2 gene may provide new insight as to the role of MDM2 (versus possibly CPM) as a “driver” for ALT/DDLS tumor formation/progression.