Background: Well differentiated liposarcomas (WDLPS) and dedifferentiated liposarcomas (DDLPS) have been shown to have supernumerary chromosomes with amplified sequences of the MDM2 and CDK4 genes. MDM2 and CDK4 protein overexpression have also been identified in these tumours.
Aim: To investigate whether immunohistochemistry (IHC) for MDM2 and CDK4 can be used to diagnose WDLPS and DDLPS.
Methods: IHC for MDM2/CDK4 was carried out on a series of 129 paraffin-embedded lipomatous and non-lipomatous soft tissue tumours. The cases were divided into four groups: WDLPS (n = 19), DDLPS (n = 10), benign adipocytic tumours (BAT) (n = 17), and other mesenquimal tumours (OMT) (n = 83). IHC results were compared in each group and the diagnostic efficacy of the test in identifying WDLPS and DDLPS among the other soft tissue tumours was determined. A percentage of tumour cell positivity was evaluated to better characterise the pattern of tumour immunostaining.
Results: Sensitivity and specificity of positive MDM2 and CDK4 immunostainings to identify WDLPS among BAT was 100% and 58.8%, and 68.4% and 88.2%, respectively. When distinguishing DDLPS from OMT, sensitivity and specificity of MDM2 and CDK4 were 90% and 65%, and 70% and 96.3%, respectively. The highest specificity was achieved when a case was considered positive with strong and diffuse immunoreactivity in more than 30% of the neoplastic cells (94.1% and 100%, and 77.1% and 98.8%, respectively).
Conclusion: Detection of MDM2/CDK4 protein overexpression by IHC can be used by pathologists to diagnose WDLPS and DDLPS. Considering a strong and diffuse immunostaining pattern in most of the neoplastic cells achieves the best results in identifying these tumours.
Statistics from Altmetric.com
Funding This study was supported by the National Council of Technological and Scientific Development (CNPq) of Brazil.
Competing interests None.
Ethics approval Ethics approval was obtained.
Additional images available online at http://picasaweb.google.com/aleixo.pedro
Provenance and peer review Not commissioned; externally peer reviewed.