Discordant results in 18F-FDG PET/CT and ultrasound-based assessment for axillary lymph node metastasis detection: A large retrospective analysis in 560 patients with breast cancer.

Abstract

PURPOSE\nUltrasound is the recommended modality to assess axillary lymph node involvement in breast cancer; nevertheless, 18F-fluorodeoxyglucose (18F-FDG) integrated positron emission tomography/computed tomography (PET/CT) diagnostic efficiency, to identify suspicious lesions, is also considered. We aim to report discrepancies in ultrasound and 18F-FDG PET/CT results.\n\n\nMETHODS\nThis single-centered retrospective analysis selected consecutive patients with invasive ductal biopsy-proven breast cancer, for whom divergent 18F-FDG PET/CT and axillary ultrasound imaging (and/or core needle biopsy if available) had been performed, and described clinical, histological, imaging, and surgery data.\n\n\nRESULTS\nThis retrospective study included 560 patients and identified discordant results between 18F-FDG PET/CT and ultrasound (suspicious 18F-FDG PET/CT and normal ultrasound imaging and/or core needle biopsy) in 20 (4%) patients. Axillary lymph node involvement was confirmed in 17 (85%) out of these 20 patients. Further, the lymph nodes were smaller than one centimeter in 12 (60%) patients, macrometastasic involvement (involvement >2\xa0mm) was detected in 13 (65%) patients, and more than 3 macrometastases were detected in 6 (30%) patients. All patients had an aggressive breast cancer. The sentinel node biopsy performed in 9 (45%) patients allowed to reveal lymph node involvement, even in cases of macrometastatic involvement.\n\n\nCONCLUSION\nDiscordant results were issued from normal ultrasound imaging and/or core needle biopsy, and suspicious 18F-FDG PET/CT revealed that 18F-FDG PET/CT may overcome axillary ultrasound limits in the specific case of aggressive breast cancers, especially for axillary lymph nodes smaller than 1 centimeter. Sentinel node biopsy remains a valuable aid, even in patients with macrometastatic involvement.