Abstract
Purpose: With the continuous improvement in the efficacy of neoadjuvant therapy (NAT), a significant proportion of breast cancer patients initially diagnosed with pathologically confirmed axillary lymph node metastasis (pN+) may achieve ypN0 status (no residual nodal metastasis) following NAT. This study aims to develop a predictive model for estimating the probability of achieving ypN0 status after NAT, thereby assisting surgeons in making optimal decisions regarding axillary management strategies. Methods: This retrospective study enrolled 671 patients diagnosed with pN+ at Tianjin Medical University Cancer Institute and Hospital between December 2018 and December 2022, all of whom completed NAT followed by surgical intervention. The cohort comprised 428 HER2-positive and 243 TNBC patients. Clinicopathological and ultrasound imaging data were systematically collected. Patients were stratified into training and validation sets at a 7:3 ratio based on admission dates. Univariate analysis was initially performed on the training set to identify potential factors associated with achieving ypN0 status post-NAT. Variables demonstrating statistical significance were subsequently incorporated into a multivariate logistic regression analysis to determine independent predictors. A predictive nomogram was then constructed using these independent factors via R software for visual interpretation of ypN0 probability. The predictive performance of the model was ultimately evaluated by generating receiver operating characteristic (ROC) curves to assess discriminative ability and calibration curves to quantify prediction accuracy, with further validation performed using the independent validation cohort. Results: In HER2 positive breast cancer patients, those exhibiting histological grade III, HER2 IHC 3+ expression, absence of lymphovascular invasion, clinical N1 stage, prominent and hypervascular tumor CDFI signal pre-NAT, and achievement of breast pathological complete response (bpCR) following NAT were significantly more likely to achieve ypN0 status. Conversely, among TNBC patients, independent predictors of post-NAT ypN0 achievement included histological grade III, taxane-platinum combination regimens, bpCR, dot-linear signals in axillary lymph nodes on post-NAT ultrasound, and minimal transverse diameter of node on final post-NAT ultrasound evaluation. Conclusions: This study established distinct predictive models for HER2-positive and TNBC cohorts with initial pN+ status to estimate the probability of achieving ypN0 following NAT. Both models demonstrated robust predictive performance through rigorous validation, providing clinicians with quantitative tools to optimize axillary management strategies and facilitate precision-based individualized treatment planning.