Dynamic contrast-enhanced MRI based radiomics model for predicting the complete pathological response to neoadjuvant chemotherapy of breast cancer
10.3760/cma.j.issn.1005-1201.2019.09.004
- VernacularTitle: 基于动态对比增强MRI的影像组学模型预测乳腺癌新辅助化疗病理完全缓解的价值
- Author:
Zhiqi YANG
1
;
Xiaofeng CHEN
;
Jiada YANG
;
Weixiong FAN
;
Xiangguang CHEN
Author Information
1. Department of Radiology, Meizhou People′s Hospital, Meizhou 514031, China
- Publication Type:Journal Article
- Keywords:
Breast neoplasms;
Magnetic resonance imaging;
Radiomics;
Neoadjuvant chemotherapy
- From:
Chinese Journal of Radiology
2019;53(9):733-736
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the value of dynamic contrast-enhanced MRI (DCE-MRI) based radiomics model in predicting the pathological complete response (pCR) to neoadjuvant chemotherapy (NAC) of breast cancer.
Methods:In this retrospective study, 91 patients who had received NAC and had pathological response results were collected in Meizhou people′s hospital from January 2016 to August 2018. A primary cohort consisted of 63 patients and an independent validation cohort consisted of 28 patients. The patients were divided into pCR group of 23 cases and non-pathological complete response (Non-pCR) group of 68 cases. All the patients underwent dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) before NAC. A list of radiomics features were extracted using the A.K software and the corresponding radiomics signature was constructed. Logistic regression was used to develop the prediction model. The predictive ability of the model was tested by using the area under the curve (AUC) of ROC analysis.
Results:The discrimination performance of radiomics signature yielded a AUC of 0.750 in the primary dataset and a AUC of 0.789 in the validation dataset. The model that incorporated estrogen receptor (ER), progesterone receptor (PR) and radiomics features was developed, and had an AUC of 0.859 in the primary dataset and an AUC of 0.905 in the validation dataset.
Conclusion:The radiomics predictive model, which integrated with the DCE-MRI based radiomics signature, ER and PR, can be used as a promising and applicable adjunct approach for predicting the pCR to NAC of breast cancer.
