Purpose: The main objective of this study was to determine the role of [(18)F]-2-fluoro-2-deoxy-D-glucose positron emission tomography (FDG PET) in the selection of patients with breast cancer as candidates for sentinel node biopsy (SNB) after neoadjuvant therapy.
Methods: Forty-four patients with primary breast cancer clinically classified as cT2, cT3 or cT4(a-c) cN0-N2 or cN3 M0 and with a baseline FDG PET scan positive both in the site of primary tumour and axillary lymph nodes underwent neoadjuvant therapy and then a second FDG PET scan. In the case of axillary FDG PET uptake, patients underwent axillary lymph node dissection (ALND). If the second FDG PET scan was negative for axillary involvement, SNB was performed in order to evaluate axillary lymph node status. Only in the case of SN positivity did total ALND follow.
Results: Specificity and positive predictive value of FDG PET for detection of axillary lymph node metastases after neoadjuvant therapy were as high as 83% (95% confidence interval: 51-97%) and 85% (95% confidence interval: 54-97%), respectively, whereas sensitivity, negative predictive value and diagnostic accuracy were inadequate for a correct staging (34, 32 and 48%, respectively).
Conclusion: The poor sensitivity of FDG PET in detecting axillary lymph node metastases makes SNB mandatory in cases of a negative scan. The relatively high positive predictive value seems to suggest a role of FDG PET in selecting patients who, after neoadjuvant therapy, are candidates for ALND, avoiding SNB. However, this issue requires confirmation in a larger series of patients.