Tumour 18 F-FDG Uptake on preoperative PET/CT may predict axillary lymph node metastasis in ER-positive/HER2-negative and HER2-positive breast cancer subtypes.

Abstract

To evaluate the association between tumour FDG uptake on preoperative PET/CT and axillary lymph node metastasis (ALNM) according to breast cancer subtype. The records of 671 patients with invasive breast cancer who underwent (18) F-FDG PET/CT and surgery were reviewed. Using immunohistochemistry, tumours were divided into three subtypes: oestrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative, HER2-positive, and triple-negative. Tumour FDG uptake, expressed as maximum standardized uptake value (SUVmax), and clinicopathological variables were analysed. ALNM was present in 187 of 461 ER-positive/HER2-negative, 54 of 97 HER2-positive, and 38 of 113 triple-negative tumours. On multivariate analysis, high tumour SUVmax (≥4.25) (P < 0.001), large tumour size (>2 cm) (P = 0.003) and presence of lymphovascular invasion (P < 0.001) were independent variables associated with ALNM. On subset analyses, tumour SUVmax maintained independent significance for predicting ALNM in ER-positive/HER2-negative (adjusted odds ratio: 3.277, P < 0.001) and HER2-positive tumours (adjusted odds ratio: 14.637, P = 0.004). No association was found for triple-negative tumours (P = 0.161). Tumour SUVmax may be an independent prognostic factor for ALNM in patients with invasive breast cancer, especially in ER-positive/HER2-negative and HER2-positive subtypes, but not in those with triple-negative subtype. • Tumour SUVmax could be an imaging biomarker for predicting ALNM • Tumour SUVmax predicting ALNM is effective in ER-positive/HER2-negative and HER2-positive subtypes • Tumour SUVmax predicting ALNM is inaccurate in triple-negative subtypes • Accurate prognostic prediction based on molecular subtype may facilitate individualized management.