The Clinical Study of Intratumoral and Peritumoral Radiomics Based on DCE-MRI for HER-2 Positive and Low Expression Prediction in Breast Cancer

Yiyan Shang ^1,2, Yunxia Wang ^1,2, Yaxin Guo ^2,3

⁰Department of Radiology, People's Hospital of Henan University, Zhengzhou, Henan, People's Republic of China.

Breast Cancer (dove Medical Press) +

|

December 20, 2024

View abstract on PubMed

Summary

Radiomics analysis of DCE-MRI can non-invasively predict human epidermal growth factor receptor-2 (HER-2) expression in breast cancer. Peritumoral radiomics showed strong predictive performance for HER-2 status, aiding preoperative assessment.

Area Of Science

Oncology
Radiology
Medical Imaging

Background

Core biopsies may miss tumor heterogeneity, impacting treatment decisions.
Radiomics offers a non-invasive approach to assess tumor characteristics beyond biopsy.
Accurate prediction of human epidermal growth factor receptor-2 (HER-2) status is crucial for breast cancer management.

Purpose Of The Study

To evaluate the clinical utility of intratumoral and peritumoral radiomics from DCE-MRI for predicting HER-2 expression in breast cancer.
To compare the performance of different radiomics models and clinical combined models for HER-2 status prediction.

Main Methods

Two tasks were performed: distinguishing HER-2 positive/negative and HER-2 low/zero expression.
Radiomics models (intratumoral, peritumoral, combined) were built using decision trees.
Clinical combined models incorporated clinicopathological features and radiomics scores using logistic regression.

Main Results

Estrogen receptor, progesterone receptor, Ki67, and MRI-reported lymph nodes showed significant differences across HER-2 expression groups.
The peritumoral radiomics model achieved the highest AUC (0.774 training, 0.727 testing) for HER-2 positive/negative prediction.
The intratumoral + peritumoral radiomics model showed the best performance (AUC 0.777 testing) for HER-2 low/zero prediction.

Conclusions

Radiomics features derived from DCE-MRI are valuable for preoperative prediction of HER-2 expression status.
Both radiomics and nomogram-based approaches aid in non-invasively assessing HER-2 status before surgery.

Keywords:

breast cancer dynamic contrast enhanced magnetic resonance imaging human epidermal growth factor receptor-2 intratumoral and peritumoral nomogram radiomics