94 – PREOPERATIVE PET-CT IS NOT USEFUL FOR IDENTIFYING DE-DIFFERENTIATION OF WELL-DIFFERENTIATED LIPOSARCOMAS

Abstract

Purpose: To determine whether combined modality Positron Emission Tomography and Computed Tomography (PET-CT) imaging can pre-operatively identify de-differentiated areas within well-differentiated liposarcomas/ atypical lipomatous neoplasms. Well-differentiated lipo-sarcomas show a reasonably homogeneous lesion with fat signal characteristics on MRI and are managed with surgical excision or regular observation. They can recur locally, but never metastasize. Up to 5% of well-differentiated liposarcomas will contain a de-differentiated component that is not apparent on MRI. When present, this de-differentiated component carries a much worse prognosis and requires more aggressive local management. Currently this is only identified after surgical resection. Pre-operative identification of a de-differentiated component within the lesion by PET-CT would allow for better treatment planning. However, PET-CT is an expensive investigation and has not been reported in this application before, although it has been used for imaging of soft tissue tumours.

Method: We have prospectively enrolled 40 subjects into this study. Eligible subjects presented with a > 8cm and deep well-differentiated liposarcoma in the extremities or retroperitoneum, which was confirmed by MRI. Subjects underwent a pre-operative PET-CT scan to look for areas of de-differentiation within the lesion. The PET-CT results were compared to the final pathological analysis of the surgical specimen. When necessary for diagnostic purposes, cytogenetic analysis was also completed.

Results: Thirty-one of the 40 subjects enrolled have had PET-CT scan and surgery. Seven subjects are waiting for PET-CT and surgery by December 2009. Two subjects were withdrawn from the study: one because the PET-CT scan could not be scheduled prior to surgery, and one because the subject ultimately declined surgery. Of the 31 lesions excised, 2 contained de-differentiated areas within the lesion. The PET-CT detected a small area of low to moderate FDG uptake (thigh lesion, max Standard Uptake Value (SUV) 3.6) for one of these patients; however no uptake on PET-CT was seen for the other (retroperitoneal lesion, max SUV 1.7).

Conclusion: PET-CT does not appear to reliably detect de-differentiated areas within well-differentiated lipo-sarcomas. Our data does not support the use of PET-CT for this purpose as it may be misleading and wasteful of resources.