Correlation of 18F-FDG metabolic and volumetric parameters with clinicopathological factors of breast cancer

Introduction: Aim: To investigate the relationship between ¹⁸F-FDG-PET semi-quantitative metabolic and volumetric parameters, clinicopathological features (CPFs) and molecular subtypes in breast cancer (BC).

Methods: Patients and Methods: Initial staging PET/CT was retrospectively reviewed in patients with pathologically proven breast cancer. Tumour FDG uptake was visually evaluated and was semi-quantitatively assessed by drowning a 3-D volume of interest (VOI) over the perimeter of the primary lesion to obtain; maximum and mean standardized uptake values (SUVmax& SUVmean), metabolic tumour volume (MTV), total lesion glycolysis (TLG), tumour liver ratio (TLR) and tumour to mediastinal blood pool ratio (TMR). These parameters were compared with CPFs and molecular subtypes of BC.

Results: Results: One-hundred Fifty-five female patients were recruited in this study; median age was 50 years (range; 19-88 years). The median values of all PET parameters were significantly higher in stage III& IV, ER-negative patients, PR negative patients, patients with positive axillary LNs, distant metastases and in cases with ductal carcinoma ( insignificant with MTV and TLG). Apart from MTV, all the parameters were significantly higher in the younger age group (<45 years), grade III tumours, triple-negative subtype, HER2neu positive and high Ki-67 levels. All the median values of the PET parameters were significantly positively correlated with the size of the primary lesion (High correlation with MTV and TLG; r=0.804 and 0.839 respectively, P<.001) and the level of Ki-67 expression. Apart from MTV, all the parameters were negatively correlated with the age of the patients and BMI. In ROC analysis, TLG had the largest AUC and was better than SUVmax in predicting the metabolic T, N and M stage (AUC; 0.737, 0.707 and 0.763 respectively with p<.001 for all).

Conclusions: Conclusion: The median values of the metabolic and volumetric PET parameters were significantly positively correlated with the size of the primary lesion and the level of Ki-67 expression. Apart from MTV, all the parameters were negatively correlated with the age of the patients and BMI. TLG had the largest AUC and was better than SUVmax in predicting the metabolic T, N and M stages.