Non-Mass Enhancements on DCE-MRI: Development and Validation of a Radiomics-Based Signature for Breast Cancer...

Area of Science:

• Radiology
• Medical Imaging
• Oncology

Background:

• Distinguishing benign from malignant non-mass enhancement lesions (NMEs) on breast MRI is challenging.
• Accurate differentiation is crucial for appropriate patient management and treatment decisions.

Purpose of the Study:

• To evaluate the added value of a radiomics signature for differentiating benign from malignant NMEs on dynamic contrast-enhanced breast MRI (DCE-MRI).
• To develop and validate a combined model incorporating radiomics and clinical features.

Main Methods:

• A retrospective study involving 232 patients (247 NMEs) for model development and 72 patients (72 NMEs) for validation.
• Radiomic features were extracted from DCE-MRI images.
• Least Absolute Shrinkage and Selection Operator (LASSO) regression was used for feature selection and signature construction.
• Multivariable logistic regression analyses built three models: clinical, radiomics, and combined.
• Nomogram and decision curve analyses assessed the combined model's clinical utility.

Main Results:

• The routine MR model showed high sensitivity (0.942) but low specificity (0.589).
• The radiomics model with six features significantly correlated with malignancy (P<0.001).
• The combined model demonstrated improved specificity (0.839) and good discrimination, with sensitivity of 0.869.
• The nomogram achieved good discrimination in the validation cohort (sensitivity 0.820, specificity 0.864).

Conclusions:

• A radiomics signature significantly enhances the differentiation between benign and malignant NMEs on breast MRI.
• The developed radiomics nomogram, integrating radiomics signatures and time-intensity curve types, offers a valuable tool for clinical decision-making.
• This radiomics-based approach can improve the accuracy of diagnosing suspicious NMEs.