Predictive value of F-fluorodeoxyglucose uptake for axillary lymph node metastasis in operable breast cancer: impact of molecular subtypes.

OBJECTIVES

To evaluate the predictive value of standardized uptake value (SUV) in both primary tumors and axillary lymph nodes (ALNs) using FDG PET/CT for lymph node metastasis in breast cancer patients, and to assess the influence of molecular subtypes on this predictive performance.

METHODS

This retrospective study included 287 patients with invasive ductal carcinoma (IDC) who underwent FDG PET/CT prior to surgery between September 2016 and December 2019. The maximum standardized uptake value (SUV) of primary tumors (SUV-B) and ALNs (SUV-LN) were analyzed. Molecular subtypes were classified as hormone receptor-positive, HER2-positive, and triple-negative breast cancer (TNBC). Receiver operating characteristic (ROC) curve analysis was performed to assess and compare the diagnostic performance of SUV-B and SUV-LN for predicting ALN metastasis.

RESULTS

Among the 287 patients, 62 (21.6%) had confirmed ALN metastasis. The median SUV-LN was significantly higher in patients with metastasis compared to those without metastasis (1.5 vs. 0.9; P < 0.001). SUV-LN demonstrated good discriminative performance for ALN metastasis (AUC: 0.796), whereas SUV-B did not show significant predictive value (AUC: 0.536). The SUV_LN demonstrated significantly lower predictive performance for ALN metastasis in the hormone-positive group (AUC: 0.796) compared to the excellent discriminative performance in the HER2-positive (AUC: 0.923, P = 0.018) and TNBC (AUC: 0.940, P = 0.004) groups. Hormone receptor-positive tumors also exhibited lower FDG uptake in metastatic lymph nodes compared to HER2-positive and TNBC subtypes (P = 0.031).

CONCLUSION

FDG PET/CT SUV-LN effectively predicts ALN metastasis in HER2-positive and TNBC subtypes. Hormone receptor-positive breast cancers show lower FDG uptake in metastatic ALNs, reducing diagnostic accuracy. This finding may aid in selecting the most appropriate diagnostic modality based on tumor characteristics in the era of personalized medicine.