Prediction of low-risk breast cancer using perfusion parameters and apparent diffusion coefficient

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Accurate identification of low-risk breast cancer is crucial for appropriate treatment selection.
• Distinguishing low-risk tumors from others can optimize patient management and reduce overtreatment.

Purpose of the Study:

• To evaluate differences in perfusion and diffusion parameters between low-risk and non-low-risk breast tumors.
• To develop a predictive model for low-risk breast cancer using imaging biomarkers.

Main Methods:

• Prospective enrollment of 87 patients with 91 tumors undergoing surgery.
• Definition of low-risk breast cancer: ER-positive, low histologic grade, low Ki67 (<14%), and negative lymph nodes.
• Acquisition of quantitative/semiquantitative perfusion parameters and apparent diffusion coefficient (ADC) using DCE-MRI and DWI.
• Statistical comparison using Fisher's exact test, Chi-square, and t-tests; logistic regression and ROC analysis for model development.

Main Results:

• Univariate analysis showed significant differences in wash-in and iAUCqualitative based on tumor characteristics (HG, ER, HER-2, Ki67, LVI).
• ADCdiff significantly differed by HG, HER-2, and Ki67 status.
• Multivariate analysis identified Ktrans, iAUCqualitative, and ADCdiff as significant predictors of low-risk tumors.
• Combined parameters yielded an AUC of 0.78, superior to individual parameters; ADCdiff correlated positively with wash-in and iAUCqualitative.

Conclusions:

• Perfusion (Ktrans, wash-in, iAUCqualitative) and diffusion (ADCdiff) parameters from DCE-MRI and DWI can effectively differentiate low-risk breast cancer.
• A predictive model incorporating these parameters shows promise as an imaging biomarker for guiding breast cancer treatment strategies.