Evaluation of pathological complete response after neoadjuvant systemic treatment of invasive breast cancer using diffusion-weighted imaging compared with dynamic contrast-enhanced based kinetic analysis

Abstract

PurposeThis study compared the performance of diffusion-weighted imaging (DWI) to dynamic contrast-enhanced (DCE)-MRI for diagnosing pathological complete response (pCR) before surgery. MethodOverall, 133 lesions from 133 patients who underwent pre-surgical MRI evaluation after neoadjuvant systemic treatment were included. Two readers blinded to the pathological diagnosis evaluated the images. MR images were obtained using a routine protocol sequence that included DWI and DCE-MRI. DWI of the target lesion was scored using a three-point scale. Kinetic patterns of lesions on DCE-MRI were scored using a four-point scale. The capacities of DWI and kinetic parameters for discriminating pCR and non-pCR were assessed via receiver operating characteristic (ROC) analysis. ResultsFor DWI scores, ROC analysis showed areas under the ROC curve (AUCs) of 0.84 (95% confidence interval: 0.77–0.90) and 0.85 (0.77–0.90) for readers 1 and 2, respectively; corresponding AUCs of kinetic scores were 0.89 (0.82–0.94) and 0.88 (0.81–0.93). Among the triple-negative subtype, the AUCs of DWI scores were 0.84 (0.70–0.93) and 0.88 (0.75–0.96) for readers 1 and 2, respectively; corresponding AUCs of kinetic scores were 0.94 (0.83–0.99) and 0.93 (0.82–0.99). Among the luminal subtype, the AUCs of DWI scores were 0.85 (0.71–0.94) and 0.82 (0.68–0.92) for readers 1 and 2, respectively; corresponding AUCs of kinetic scores were 0.82 (0.68–0.92) and 0.72 (0.56–0.85). ConclusionsOur DWI-based visual score and kinetic score showed similar diagnostic performances. Both DWI and kinetic scores tended to perform better in predicting pCR for the triple-negative subtype.