Core needle biopsy washout biomarkers combined with ultrasound for preoperative axillary risk stratification in breast cancer: a retrospective nomogram study with temporal validation

Background: Preoperative assessment of axillary lymph node (ALN) involvement informs breast cancer surgical planning. Axillary ultrasound (US) is routine but may miss occult disease. We evaluated whether vascular endothelial growth factor-C (VEGF-C) and cytokeratin 19 (CK19) in primary-tumor core needle biopsy (CNB) washout fluid improved US-based risk assessment.

Methods: This single-center retrospective temporal-validation study screened 493 women with primary breast cancer who underwent preoperative axillary US, primary-lesion CNB, and postoperative nodal pathology from June 2023 to July 2025. Eligible cases were split by calendar time into development and temporal-validation cohorts. CNB washout VEGF-C and CK19 were measured by enzyme-linked immunosorbent assay (ELISA). A four-variable logistic model was fitted in development and applied unchanged to temporal validation.

Results: The eligible analysis set included 380 patients: 278 in development and 102 in temporal validation. Pathological ALN metastasis was present in 114 (41.0%) and 42 (41.2%) patients, respectively. The model included tumor size, positive axillary US status, log₂(VEGF-C+1), and log₂(CK19+1). Development-cohort area under the receiver operating characteristic curve (AUC) was 0.861 [95% confidence interval (CI): 0.813–0.905], and temporal-validation AUC was 0.799 (95% CI: 0.706–0.880). Temporal-validation calibration intercept and slope were 0.183 and 0.869. At the development-derived Youden cutoff of 0.445, temporal-validation sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were 64.3%, 86.7%, 77.1%, and 77.6%.

Conclusions: CNB washout VEGF-C and CK19 complemented axillary US and tumor size for preoperative ALN risk stratification and retained moderate discrimination in later same-center patients. Multicenter external validation is required before clinical use.