Longitudinal DCE MRI Vascular Textures: Radiologic and Biologic Insights for pCR Prediction in HER2-Negative Breast...

Area of Science:

• Radiology and Medical Imaging
• Oncology
• Biomedical Engineering

Background:

• Accurate prediction of pathologic complete response (pCR) is crucial for HER2-negative breast cancer treatment optimization.
• Dynamic contrast-enhanced MRI (DCE-MRI) offers insights into tumor vascularity, a key factor in treatment response.
• Existing models often rely on limited clinical or volumetric data, potentially missing subtle indicators of response.

Purpose of the Study:

• To develop and validate a novel prediction model for pCR in HER2-negative breast cancer.
• To leverage longitudinal changes in DCE-MRI-derived vascular textures for improved predictive accuracy.
• To compare the performance of the new vascular texture-based model against traditional functional tumor volume (FTV) methods.

Main Methods:

• Retrospective analysis of DCE-MRI data from I-SPY2 and ACRIN 6698 trials for model development and internal validation.
• External validation using an independent hospital cohort.
• Extraction of Image Biomarker Standardization Initiative-standardized vascular textures from FTV.
• Development of the DCE-MRI Vascularization-Based Response Tracking (DCE-VASC-TRACK) model incorporating texture changes, clinical factors, and FTV.
• Performance evaluation using Area Under the Receiver Operating Curve (AUC).

Main Results:

• The DCE-VASC-TRACK model demonstrated significant association between vascular texture changes (complexity, run-length variance) and pCR.
• In the external test cohort, DCE-VASC-TRACK achieved a higher AUC (0.86) compared to the FTV-based model (0.72).
• Vascular texture analysis revealed biological pathway enrichment in angiogenesis and TGF-beta signaling.

Conclusions:

• Incorporating midtreatment DCE-MRI vascular texture dynamics significantly enhances pCR prediction in HER2-negative breast cancer.
• The DCE-VASC-TRACK model offers superior predictive performance over models relying solely on clinical and FTV features.
• Vascular texture analysis provides deeper biological insights into treatment response mechanisms.