Background: Diagnosing adipocytic tumors can be challenging because it is often difficult to morphologically distinguish between benign, intermediate and malignant adipocytic tumors, and other sarcomas that are histologically similar. Recently, a number of tumor-specific chromosome translocations and associated fusion genes have been identified in adipocytic tumors and atypical lipomatous tumors/well-differentiated liposarcomas (ALT/WDL), which have a supernumerary ring and/or giant chromosome marker with amplified sequences of the MDM2 and CDK4 genes. The purpose of this study was to investigate whether quantitative real-time polymerase chain reaction (PCR) could be used to amplify MDM2 and CDK4 from total RNA samples obtained from core-needle biopsy sections for the diagnosis of ALT/WDL.
Methods: A series of lipoma (n = 124) and ALT/WDL (n = 44) cases were analyzed for cytogenetic analysis and lipoma fusion genes, as well as for MDM2 and CDK4 expression by real-time PCR. Moreover, the expression of MDM2 and CDK4 in whole tissue sections was compared with that in core-needle biopsy sections of the same tumor in order to determine whether real-time PCR could be used to distinguish ALT/WDL from lipoma at the preoperative stage.
Results: In whole tissue sections, the medians for MDM2 and CDK4 expression in ALT/WDL were higher than those in the lipomas (P < 0.05). Moreover, karyotype subdivisions with rings and/or giant chromosomes had higher MDM2 and CDK4 expression levels compared to karyotypes with 12q13-15 rearrangements, other abnormal karyotypes, and normal karyotypes (P < 0.05). On the other hand, MDM2 and CDK4 expression levels in core-needle biopsy sections were similar to those in whole-tissue sections (MDM2: P = 0.6, CDK4: P = 0.8, Wilcoxon signed-rank test).
Conclusion: Quantitative real-time PCR of total RNA can be used to evaluate the MDM2 and CDK4 expression levels in core-needle biopsies and may be useful for distinguishing ALT/WDL from adipocytic tumors. Thus, total RNA from core-needle biopsy sections may have potential as a routine diagnostic tool for other tumors where gene overexpression is a feature of the tumor.