An Audit of Lipomatous Neoplasms: Correlation with MDM2 Amplification by Fluorescence In Situ Hybridisation
TG Kashima, DD Halai, R Tirabosco, S Hing, H Ye, AM Flanagan. Royal National Orthopaedic Hospital NHS Trust, London, United Kingdom; UCL, London, United Kingdom; University of Cambridge, Cambridge, United Kingdom
Background: MDM2 amplification has been reported as being a sensitive means of distinguishing lipomas from atypical lipomatous tumours/well differentiated liposarcomas (ALT/WDL). The aim of this study was to discover if employing MDM2 FISH for distinguishing between these 2 tumour types would be valuable in clinical practice.
Design: The microscopy of 139 benign/low grade and 25 high grade lipomatous neoplasms were reviewed by 3 pathologists (blinded to the clinical information) and the results correlated with the presence and absence of MDM2 amplification (MDM2 & CEP12 probes provided by Zytovision). 50 cells with at least two CEP12 signals were counted in each case. Cases were considered to be positive for MDM2 amplification if the MDM2:CEP12 ratio was equal to or more than 2.
Results: FISH was found to be informative in all cases. 59 of the 164 cases had MDM2 amplification. There was 100% agreement between the pathologists when making diagnoses of hibernoma (n=10), lipoblastoma (n=5), spindle cell lipoma (n=14) and pleomorphic liposarcoma (n=10): none of these tumours harboured MDM2 amplification. 13 of 15 cases of the dedifferentiated component of ALT/WDL revealed MDM2 amplification. Of the remaining 110 cases (conventional lipomas and ALT/WDL) 46 revealed an MDM2 amplification and 67% of these were unequivocally considered histologically to represent ALT/WDL by all pathologists. The tendancy was to overcall these mature lipomatous tumours. The tumours that were undercalled contained only a few scattered atypical cells whereas the majority of the small tumour cells in these neoplasms revealed MDM2 amplification. The average size of tumours reassessed as lipoma or ALT/WDL on the basis of MDM2 amplification was 16.4 cm (range 4-31cm) and 15.7 cm (3.5-29cm) respectively (p>0.05), indicating tumour size is unhelpful in distinguishing between these tumour types. 23 of the 25 tumours that recurred harboured an MDM2 amplification and the average size of the tumours that recurred was 18cm (range 12-24.5cm).
Conclusions: We found that using FISH to identify MDM2 amplification significantly improved distinguishing lipomas from ATL/WDL. Furthermore, correlation of MDM2 copy number with histological findings subsequently improved diagnostic accuracy.
Category: Bone & Soft Tissue
Monday, March 9, 2009 1:00 PM
Poster Session II # 4, Monday Afternoon