首页|联合ADC组学标签和激素受体预测乳腺癌新辅助化疗病理完全化解

联合ADC组学标签和激素受体预测乳腺癌新辅助化疗病理完全化解

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
  • 维普
目的 探讨使用联合ADC图组学标签和激素受体模型来预测乳腺癌新辅助化疗(NC)病理完全缓解的价值.方法 共收集了 165例NC前行DWI检查女性乳腺癌患者资料,年龄在28-70岁之间.基于ADC共提取了 396个组学特征并进行特征筛选.比较病理完全缓解和病理部分缓解组间临床特征的差异.将有差异临床特征和最优影像组学特征纳入logistic回归以建立模型.使用ROC曲线和决策曲线来评估模型的效能.结果 在训练集和测试集中,激素受体表达状态在病理完全缓解和病理部分缓解组间的差异均具有统计学意义(P＜0.05).基于ADC图组学标签模型在训练集和测试集中预测乳腺癌NC病理完全缓解的曲线下面积(AUC)分别为0.785和0.639.基于联合ADC图组学标签和激素受体建立的联合模型在训练集和测试集中预测乳腺癌NC病理完全缓解的AUC分别为0.904和0.789.联合模型的临床获益高于基于ADC图组学标签模型.结论 联合ADC图学标签和激素受体建立的联合模型对预测乳腺癌NC病理完全缓解具有较好的价值.
Combined Apparent Diffusion Coefficient Maps Radiomics Signature and Hormone Receptors Predict Pathological Complete Response of Neoadjuvant Chemotherapy in Breast Cancer
Objective To investigate the value of a combined apparent diffusion coefficient(ADC)mapsbased radiomics signature and hormone receptors model in predicting pathological complete response to neoadjuvant chemotherapy(NC)in breast cancer.Methods Data was collected from 165 female breast cancer patients,aged 28 to 70 years,who underwent NC prior to diffusion-weighted imaging examination.A total of 396 radiomics features were extracted from the ADC maps followed byfeature selection.The clinical features of the pathological complete response(PCR)group and pathological partial response(PPR)group were compared.Significant features and optimal radiomics features were then included in the logistic regression to establish the models.The performance of the model was evaluated using the ROC curve and decision analysis curve.Results There was significant difference in hormone receptors between the two groups both in the training and testing set.The performance of ADC maps maps-based radiomics model yield an AUC of 0.785 in the training set and an AUC of 0.639 in the testing set.The proposed model that combined(ADC)maps-based radiomics signature and hormone receptors yielded a maximum AUC of 0.904 in the training set and an AUC of 0.789 in the testing set.The decision curve showed that the benefit of the combined model was higher than that of the(ADC)maps-based radiomics model.Conclusion The combined model,which integrated with(ADC)maps-based radiomics signature and hormone receptors,may serve as a potential markers to successfully predict complete pathological response to NC.

Breast CancerApparent Diffusion CoefficientHormone ReceptorRadiomicsNeoadjuvant Chemotherapy

李小苑、杨志企、陈湘光、陈寿让、温伟华、杨宇扬、戴卓智、陈小凤

展开 >

梅州市人民医院乳腺外科(广东梅州 514031)

广东省临床分子诊断与抗体药物工程技术研究中心(广东梅州 514031)

梅州市人民医院影像科(广东梅州 514031)

汕头市中心医院影像科(广东汕头 515031)

展开 >

乳腺癌 表观扩散系数 激素受体 影像组学 新辅助化疗

梅州市人民医院培育项目梅州市人民医院培育项目梅州市社会发展科技计划项目梅州市社会发展科技计划项目广东省医学科研基金项目国家自然科学基金

PY-C2020016PY-C20200222020B1052022B46B202128082101985

2024

10.3969/j.issn.1672-5131.2024.01.028

中国CT和MRI杂志
北京大学深圳临床医学院 北京大学第一医院

中国CT和MRI杂志

CSTPCD
影响因子：1.578
ISSN：1672-5131
年,卷(期)：2024.22(1)
  • 6