Development and internal validation of a nomogram based on HER2 status for predicting pathological complete response...

Neoadjuvant chemotherapy (NAC) is the pivotal therapy for triple-negative breast cancer (TNBC), with pathological complete response (pCR) serving as a critical prognostic indicator of NAC efficacy. The clinical tools stratifying TNBC patients by human epidermal growth factor receptor 2 (HER2) subtypes are currently lacking. This study aims to develop a nomogram based on HER2 status for predicting the NAC efficacy and prognosis in TNBC patients.

A retrospective analysis was performed on the clinical data of 122 patients with primary TNBC admitted to the Gansu Provincial Maternity and Child Health Hospital from January 2015 to December 2023 by collecting patients' inpatient medical records and outpatient follow-up records. According to the HER2 expression status, the patients were divided into HER2-zero group and HER2-low group. The disease-free survival (DFS) was compared by the Kaplan-Meier curves with log-rank test. The nomogram predicting pCR was established by a multivariate Logistic regression analysis model. Internal validation was conducted using bootstrapping with 1,000 samples to assess the robustness of the prediction model. The predictive performance of the nomogram was comprehensively evaluated using receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA).

A total of 122 patients with TNBC were finally enrolled. The median age at diagnosis was 50 years. Of these, 54.1% were premenopausal, 64.8% presented with lymph node‑positive disease, 69.7% exhibited high Ki-67 expression (>60%), 61.5% were p53‑positive, and 28.7% were androgen receptor (AR)‑positive. Median follow‑up duration was 48 months. The pCR was more notable when we compared HER2-zero group to HER2-low group (P=0.048). HER2-zero group also exhibited a more improved DFS when we compared to HER2-low group (P=0.043). Subgroup analysis results showed that there was statistically significant difference in DFS between P53-positive and P53-negative patients in HER2-zero group (P=0.02). Regardless of the HER2 expression status, the expressions of AR, P53 and Ki-67 were not associated with DFS in all other subgroups (P>0.05). The final independent predictors incorporated into the nomogram were HER2 status, AR, P53, and Ki-67 (all P<0.05). The area under the ROC curve (AUC) of the nomogram for predicting the pCR was 0.78. Calibration curve demonstrated that the predicted probabilities of the pCR generated by the nomogram was in good agreement with the actual observed probabilities (Hosmer-Lemeshow test, P=0.87). DCA showed that when the threshold probability ranged from 20% to 100%, the nomogram provided a favorable net benefit than either the "treat all" or "treat none" strategies.

HER2‑zero TNBC shows higher pCR rate and better DFS than HER2‑low TNBC. The nomogram based on HER2 status shows promising preliminary predictive performance in predicting pCR. Given the lack of external validation and single‑center design, this model remains exploratory and hypothesis‑generating.