Abstract
OBJECTIVE: To investigate the potential factors of false-negative sentinel lymph node biopsies (SLNB) for breast cancer and the clinical decision-making value of axillary lymph node dissection (ALND). METHODS: We retrospectively selected 312 patients with breast cancer who underwent SLNB in our hospital from May 2021 to July 2024. These subjects were divided into a false negative group and a true negative group according to the SLNB results. Influencing factors were analyzed by Logistic regression. A total of 300 breast cancer patients who underwent ALND at the same time were selected as the axillary group. Breast cancer patients who underwent SLNB were selected as the sentinel group. The pain levels, changes in shoulder joint function, and the incidence of complications were compared between the two groups. RESULT: Among 312 patients, 189 negative and 123 positive sentinel lymph nodes (SLNs) were detected by SLNB, and 177 negative and 135 positive SLNs were detected by ALND, with the false negative rate of 17.99% (34/189). Tumor size, vascular invasion, and number of SLNs were independent risk factors for false-negative SLNB in breast cancer (P < 0.05). The incidence of complications in the sentinel group was 9.62%, which was much lower than 32.00% of the axillary group (P < 0.05). Further molecular subtype subgroup analysis showed that the false negative rate of SLNs in patients with triple-negative breast cancer (26.47%, 9/34) was significantly higher than that in Luminal type (15.74%, 17/108) and HER2 overexpression type (16.67%, 4/24). A simple risk scoring system was constructed by integrating independent risk factors (tumor size ≥ 2 cm, SLNs ≤ 3, and no vascular invasion) based on the results of multivariate regression. This model was verified by the ROC curve, with an AUC of 0.82 (95% CI: 0.75-0.89), demonstrating good predictive efficacy. CONCLUSION: Tumor size, vascular invasion, and the number of SLNs detected were independent risk factors for false negatives in SLNB for breast cancer. The risk scoring system could complement the trial protocols such as ACOSOGZ0011 and AMAROS. This provided quantifiable tools for clinical decision-making.