多模态机器学习预测乳腺癌新辅助化疗后前哨淋巴结状态
本研究纳入362例可手术早期乳腺癌患者,整合MRI、乳腺X线、灰阶超声影像组学及临床危险因素,构建并验证可解释的多模态机器学习模型。LightGBM模型在内部验证队列、外部测试队列1和队列2的AUC分别达0.95、0.93和0.91,在不同分子亚型中判别能力存在差异。
要点
- 多模态影像组学联合临床特征可较准确预测NAC后SLN状态,辅助腋窝手术决策。
- LightGBM模型主要受DCE-MRI球形度特征驱动,提示MRI形态学特征价值较高。
- 模型在不同分子亚型中AUC波动较大,临床应用需结合分子分型谨慎解读。
核心结论
- LightGBM模型对可手术早期乳腺癌NAC后SLN状态具有较好的术前预测能力,但需更大规模多中心前瞻性研究验证其临床效用。
研究设计与样本
- 回顾性研究,共纳入362例完成NAC并接受手术的可手术早期乳腺癌患者。
- 216例用于建模,146例分别纳入两个独立外部测试队列。
- 影像组学特征提取自MRI、乳腺X线和灰阶超声,并整合临床危险因素。
- 采用LightGBM、随机森林、逻辑回归和XGBoost四种算法构建预测模型。
- 以SHAP方法解释模型,并按乳腺癌分子亚型分层验证模型的临床稳健性与泛化能力。
关键影像表现/征象
- 模型主要由DCE-MRI来源的球形度特征驱动。
- 影像征象层面原文未报告具体信号、密度或回声特征的细节。
诊断标准与关键数值
- LightGBM模型整合25个特征,内部验证队列AUC为0.95,外部测试队列1为0.93,外部测试队列2为0.91。
- 按分子亚型分层:HER2过表达型AUC为0.81,Luminal B HER2阴性型为0.92,Luminal B HER2阳性型为0.98,三阴型为0.79。
- 原文未报告敏感度、特异度或具体分类阈值。
临床应用与阅片要点
- 阅片时可重点关注DCE-MRI上病灶的球形度特征,该特征对模型预测贡献最大。
- 报告乳腺MRI时注意描述肿块形态是否趋于球形,有助于提示NAC后SLN状态。
- 该模型在不同分子亚型中判别能力差异明显,用于临床决策前需明确患者分子分型。
- 外部验证队列样本量较小且亚型分布不均衡,解读结果时应持谨慎态度。
局限性
- 研究为回顾性设计。
- 外部测试队列规模较小。
- 乳腺癌分子亚型分布不均衡。
- 敏感度、特异度及具体分类阈值未报告。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
RATIONALE AND OBJECTIVES: Although accurate preoperative prediction of sentinel lymph node (SLN) status in operable early-stage breast cancer (EBC) after neoadjuvant chemotherapy (NAC) is crucial for optimal axillary surgery, it remains challenging with a single imaging modality. We aimed to develop and validate an interpretable multimodal machine-learning model to improve diagnostic accuracy.
MATERIALS AND METHODS: In total, 362 operable EBC patients who had completed NAC followed by surgery were enrolled, namely 216 patients in the primary cohort and 146 patients in two independent external test cohorts. Radiomics features derived from magnetic resonance imaging (MRI), mammography and grayscale ultrasound were integrated with clinical risk predictors to build predictive models using Light Gradient Boosting Machine (LightGBM), Random Forest, Logistic Regression, and eXtreme Gradient Boosting algorithms. The optimal model was further selected by balancing complexity and performance, interpreted using SHapley Additive exPlanations, and validated in independent external cohorts stratified by breast cancer molecular subtypes to assess its clinical robustness and generalizability.
RESULTS: The LightGBM model incorporating 25 features demonstrated robust performance, achieving AUCs of 0.95, 0.93, and 0.91 in the internal validation cohort, external test cohort 1 and external cohort 2, respectively. It was predominantly driven by DCE-MRI-derived sphericity features and maintained different discriminative ability across molecular subtypes, yielding AUCs of 0.81 for HER2-overexpressing, 0.92 for Luminal B HER2-negative, 0.98 for Luminal B HER2-positive, and 0.79 for triple-negative.
CONCLUSION: The LightGBM model demonstrated promising preoperative predictive ability for SLN status after NAC in operable EBC, but its clinical utility requires validation in larger, multicenter prospective studies, considering the retrospective design, small external cohorts, and subtype imbalance.
引用
- Jin, Yang, Liu, et al. A Machine-Learning Model Using Pre-treatment Multimodal Data to Predict Sentinel Lymph Node Status After Neoadjuvant Chemotherapy in Operable Early-stage Breast Cancer. Acad Radiol. 2026. PMID: 42469044. doi:10.1016/j.acra.2026.06.009
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。