[摘 要] 目的：探究雌激素受体（ER）阳性乳腺癌患者新辅助内分泌治疗后病理完全缓解（pCR）的疗效影响因素，构建并验证 预测模型。方法：回顾性选取2020年1月至2024年1月河北医科大学第四医院收治的715例接受新辅助内分泌治疗及手术治疗 的ER阳性乳腺癌患者，根据术后病理结果分为pCR组（72例）和未达pCR组（643例）。按7∶3比例随机分为训练集和测试集，收 集患者临床病理资料及实验室指标。通过单因素和多因素Logistic回归分析筛选未达pCR的独立危险因素，并构建预测模型。 利用训练集和测试集的受试者工作特征（ROC）曲线、校准曲线及决策曲线分析（DCA）评估模型性能。结果：训练集与测试集患 者临床资料比较无显著差异。训练集中，pCR 组与未达pCR 组在肿瘤增殖标志蛋白Ki-67 指数、治疗前血清癌抗原15-3 （CA15-3）水平、治疗前雌激素受体1（ESR1）基因突变状态、治疗前miR-1-3p表达水平、治疗前性别决定区Y框蛋白9（SOX9）表 达水平、临床分期、淋巴结转移等方面存在差异（P < 0.05）。多因素Logistic回归分析显示，Ki-67指数、治疗前血清CA15-3水平、 治疗前ESR1基因突变状态、治疗前miR-1-3p表达水平、临床分期为ER阳性乳腺癌患者新辅助内分泌治疗后未达pCR的独立 影响因素（P < 0.05）。基于上述因素构建列线图预测模型，并通过ROC曲线、校准曲线及DCA验证模型具有较好的区分度、预测 准确性和临床实用性。结论：基于Ki-67指数、血清CA15-3水平、ESR1基因突变状态等因素构建的列线图预测模型，对ER阳性 乳腺癌患者新辅助内分泌治疗后未达到pCR风险具有良好预测效能，可为临床决策提供参考。

[Abstract] Objective: To investigate the factors influencing pathologic complete response (pCR) following neoadjuvant endocrine therapy in patients with estrogen receptor (ER)-positive breast cancer and to develop and validate a predictive model. Methods: A retrospective study was conducted on 715 patients with ER-positive breast cancer who received neoadjuvant endocrine therapy and surgical treatment at the Fourth Hospital of Hebei Medical University between January 2020 and January 2024. Based on postoperative pathological results, the patients were divided into a pCR group (72 cases) and a non-pCR group (643 cases). The cohort was randomly divided into a training set and a testing set in a 7∶3 ratio, and clinicopathological and laboratory data were collected. Independent risk factors for failure to achieve pCR were identified using univariate and multivariate logistic regression analyses, and a predictive model was constructed. Model performance was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA) for both the training and testing sets. Results: There were no significant differences in clinical data between patients in the training set and the test set. In the training set, there were significant differences between the pCR group and the non-pCR group in terms of the tumor proliferation marker protein Ki-67 index, pre-treatment serum cancer antigen 15-3 (CA15-3) levels, pre-treatment estrogen receptor 1 (ESR1) gene mutation status, pre-treatment miR-1-3p expression levels, pre-treatment SRY-box transcription factor 9 expression levels, clinical stage, and lymph node metastasis (P < · · 554 张华, 等. ER阳性乳腺癌患者新辅助内分泌治疗后疗效影响因素及预测模型构建 0.05). Multivariate logistic regression analysis revealed that the Ki-67 index, pre-treatment serum CA15-3 levels, pre-treatment ESR1 mutation status, pre-treatment miR-1-3p expression levels, and clinical stage were independent influencing factors associated with failure to achieve pCR/non-pCR following neoadjuvant endocrine therapy in patients with ER-positive breast cancer (P < 0.05). A nomogram prediction model was constructed based on these factors, and validation using ROC curves, calibration curves, and DCA confirmed that the model possesses good discriminatory power, predictive accuracy, and clinical utility. Conclusion: The nomogram prediction model incorporating factors such as the Ki-67 index, serum CA15-3 levels, and ESR1 gene mutation status demonstrates good predictive performance for the risk of failing to achieve pCR following neoadjuvant endocrine therapy in patients with ER-positive breast cancer, and can serve as a reference for clinical decision-making.

[基金项目] 2025年度河北省医学科学研究课题计划（20250102）