Abstract
1709
Objectives FDG PET/CT is a powerful tool for the detection of metastatic melanoma. However, studies of diagnostic accuracy at specific disease stages are limited by a lack of well-defined patient populations. This study aims to assess the performance of FDG PET/CT in patients with locally recurrent melanoma who have no evidence of distant metastasis.
Methods 435 scans performed on 193 patients with melanoma referred for FDG PET/CT between 12/2004 and 10/2008 were reviewed. 26 scans in patients with no pre-test evidence of distant metastases were identified in which locoregional recurrence was seen on PET/CT. The number of local metastases identified at PET/CT was compared to the number of lesions reported on physical exam. Distant metastases identified by PET/CT were recorded. All findings were validated by imaging or pathology follow-up.
Results 26 scans met inclusion criteria and demonstrated 38 locoregional metastases compared to physical exam which identified 33 locoregional metastases. 9/26 (35%) PET/CT scans identified additional locoregional lesions when compared to physical exam alone. 8 scans demonstrated the same number of lesions as physical exam and 6 scans identified fewer metastases compared to physical exam. 5/26 (19%) scans demonstrated distant metastases that were not known prior to PET/CT.
Conclusions When PET/CT demonstrates locally recurrent melanoma, it often identifies more locoregional metastases compared to physical exam, and frequently reveals occult distant metastases. PET/CT should be strongly considered prior to therapy in patients with locally recurrent melanoma and no evidence of distant metastases.
- © 2009 by Society of Nuclear Medicine