Research Article: Development and internal validation of a radiomics-clinical combined model for predicting axillary pathological complete response in clinically node-positive breast cancer patients after neoadjuvant chemotherapy
Abstract:
Accurate prediction of axillary pathological complete response (apCR) after neoadjuvant chemotherapy (NAC) in clinically node-positive (cN+) breast cancer patients may guide surgical de-escalation from axillary lymph node dissection to sentinel lymph node biopsy. This study aimed to develop and internally validate a combined radiomics-clinical prediction model for apCR.
This single-center retrospective study enrolled 386 cN+ breast cancer patients (training, n = 270; validation, n = 116). Pre-NAC DCE-MRI radiomic features were extracted from primary tumors. LASSO regression selected eight features for Rad-score construction. Clinical predictors were identified via logistic regression. Model performance was evaluated using AUC, calibration metrics, and decision curve analysis.
The overall apCR rate was 43.5% (168/386). The combined model (tumor size, HER2 status, Ki-67, breast clinical complete response, and Rad-score) achieved a validation AUC of 0.703 (95% CI, 0.610–0.792). It significantly outperformed the radiomics-only model (?AUC = 0.094, P = 0.004) but not the clinical-only model (?AUC = 0.020, P = 0.713). The combined model showed a calibration slope of 0.811 and an intercept of 0.018, indicating moderate overfitting. Risk stratification showed monotonic gradients across tertiles (low 18.8%, intermediate 48.0%, high 58.8%). After bootstrap bias correction, the optimism-corrected training AUC of the combined model was 0.742.
The combined model demonstrated moderate discriminatory ability but did not add significant value over clinical predictors alone. The current misclassification rate precludes direct clinical application for surgical de-escalation. External multicenter validation is warranted.
Introduction:
Accurate prediction of axillary pathological complete response (apCR) after neoadjuvant chemotherapy (NAC) in clinically node-positive (cN+) breast cancer patients may guide surgical de-escalation from axillary lymph node dissection to sentinel lymph node biopsy. This study aimed to develop and internally validate a combined radiomics-clinical prediction model for apCR.
Read more