Association between semiquantitative PET parameters and molecular subtypes of breast invasive ductal carcinoma

Abstract

Background: Molecular subtypes of breast cancer have been proposed since 2012. The correlation between various baseline [18F]fluorodeoxyglucose ([18F]FDG) uptake parameters, including total lesion glycolysis (TLG), and molecular subtypes of primary breast cancer lesions in patients with invasive ductal cancer will be investigated.

Methods: Staging [18F]FDG PET/CT for breast invasive ductal carcinoma were retrospectively evaluated. Breast lesions were examined for estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2) and proliferation index (Ki-67). Breast tumors were classified into five molecular subtypes: Luminal A, Luminal B-HER2(-), Luminal B-HER2(+), HER2(+) and Basal or Triple Negative cancers. The correlations between tumor characteristics and PET semiquantitative data of primary breast lesion (SUVmean, SUVmax, Mean tumor volume (MTV), TLG) were assessed. Specific Breast Uptake Ratio (SBUR) is used as a new quantification method of breast uptake to correct for physiological background activity.

Results: Fifty-eight patients were included. TLG was significantly higher in triple negative group when compared with luminal A (P<0.01). Significantly higher uptake was found in triple negative lesions when compared with luminal B-HER2(-) and luminal B-HER2(+) categories using SUVmax, SUVmean and TLG (all P<0.05). Conversely, no statistically significant difference for [18F]FDG uptake was observed between all other molecular subtypes. No value of SBUR in terms of correlation with histopathological parameters was demonstrated.

Conclusions: TLG was superior to SUVmax and SUVmean in differentiating between triple negative breast cancer lesions and all other molecular subtypes. SBUR was not different statistically between various molecular subtypes.