ObjectiveTo investigate the factors influencing pathological complete response (pCR) after neoadjuvant chemotherapy (NACT) in patients with luminal breast cancer (LBC), and to construct and validate a nomogram-based predictive model. MethodsPatients with LBC who received NACT at the Affiliated Hospital of Southwest Medical University between January 2021 and February 2025 were retrospectively enrolled. Patients were randomly divided into training cohort (n=205) and validation cohort (n=87) by a ratio of 7∶3. Multivariate logistic regression analyses was performed in the training cohort, and a nomogram was developed based on the multivariate results. Model discrimination was evaluated using receiver operating characteristic (ROC) curves, calibration was assessed using calibration plots, and clinical utility was examined using decision curve analysis (DCA) in both cohorts. ResultsMultivariate logistic regression analysis in the training cohort showed that clinical tumor stage 4 [OR=0.018, 95%CI (0.001, 0.312), P=0.006], estrogen receptor expression>37.5% [OR=0.275, 95%CI (0.095, 0.798), P=0.018], and Ki-67 index>47.5% [OR=4.134, 95%CI (1.480, 11.544), P=0.007] were independent factors associated with pCR after NACT in LBC patients. A nomogram was constructed accordingly. The area under the ROC curve of the predictive model was 0.834 in the training cohort and 0.785 in the validation cohort. Calibration curves and Hosmer-Lemeshow tests demonstrated good predictive performance of the model in both cohorts (χ2=1.610, P=0.807; χ2=1.859, P=0.762). DCA indicated that the nomogram provided the greatest net benefit when the threshold probability ranged from 0% to 50% in both cohorts. ConclusionsClinical tumor stage, estrogen receptor expression level, and Ki-67 index were independent predictors of pCR after NACT in LBC patients. The nomogram constructed based on these factors showed good predictive performance in both the training and validation cohorts.
