The Effect of Model-Based Bolus Arrival Time on Quantitative DCE Parameters and Prediction of Breast Cancer Therapy Response: A Preliminary Study

Motivation: Uncertainty in contrast bolus arrival time (BAT) can cause bias in estimated parameters from pharmacokinetic (PK) analysis of DCE-MRI data. Goal(s): Evaluate effects of BAT on Ktrans and its predictive performance for breast cancer (BC) response to neoadjuvant chemotherapy (NAC). Approach: Eleven BC patients treated with NAC underwent longitudinal DCE-MRI, which were fitted with the Tofts model and Shutter-Speed model using a fixed BAT and voxel-wise model-based BATs. Results: Tumor Ktrans differed significantly (P < 0.05) between the two BAT methods, with improved Ktrans predictive performance using model-based BAT. Impact: Use of model-based BAT at voxel level for DCE-MRI PK modeling may potentially improve accuracy of estimated Ktrans and its predictive performance for NAC response.