Diagnostic Value of Radiomics Analysis in Contrast-Enhanced Spectral Mammography for Identifying Triple-Negative Breast Cancer.

Yongxia Zhang,Fengjie Liu,Heng Ma,Ran Zhang,Jian Sun,Hao Shi,Han Zhang,Lei Song

doi:10.3389/fonc.2021.773196

Yongxia Zhang, Fengjie Liu + Show 6 more

Open Access

https://doi.org/10.3389/fonc.2021.773196

Copy DOI

Abstract

PurposeTo evaluate the value of radiomics analysis in contrast-enhanced spectral mammography (CESM) for the identification of triple-negative breast cancer (TNBC).MethodCESM images of 367 pathologically confirmed breast cancer patients (training set: 218, testing set: 149) were retrospectively analyzed. Cranial caudal (CC), mediolateral oblique (MLO), and combined models were built on the basis of the features extracted from subtracted images on CC, MLO, and the combination of CC and MLO, respectively, in the tumour region. The performance of the models was evaluated through receiver operating characteristic (ROC) curve analysis, the Hosmer-Lemeshow test, and decision curve analysis (DCA). The areas under ROC curves (AUCs) were compared through the DeLong test.ResultsThe combined CC and MLO model had the best AUC and sensitivity of 0.90 (95% confidence interval: 0.85–0.96) and 0.97, respectively. The Hosmer–Lemeshow test yielded a non-significant statistic with p-value of 0.59. The clinical usefulness of the combined CC and MLO model was confirmed if the threshold was between 0.02 and 0.81 in the DCA.ConclusionsMachine learning models based on subtracted images in CESM images were valuable for distinguishing TNBC and NTNBC. The model with the combined CC and MLO features had the best performance compared with models that used CC or MLO features alone.

Highlights

Triple-negative breast cancer (TNBC) accounts for 10–20% of all diagnosed breast cancers [1]
Two of the 8 radiomics features in the combined model were original GLRLM from the cranial caudal (CC); the other 6 features were from the mediolateral oblique (MLO) feature datasets
The radiomics features in the combined model were the same as part of radiomics features in the CC and MLO models

Summary

Introduction

Triple-negative breast cancer (TNBC) accounts for 10–20% of all diagnosed breast cancers [1]. Given the lack of the expression of human epidermal growth factor receptor-2 (HER-2) and estrogen and progesterone receptors, which can be used for targeted therapy, TNBC is difficult to treat and has a high recurrence and metastasis rate, and a low survival rate [2]. Immunohistochemistry, which analyzes part of the tumor tissue obtained by invasive biopsy or surgery, is commonly used for assessing the molecular subtype of breast cancer. Given the spatial and temporal heterogeneity of breast tumors [3], the accuracy of biopsy is limited. An alternative method is necessary to assess the molecular subtype of the breast cancer completely and non-invasively

Methods

Results

Conclusion