The kinetic parameters of dynamic contrast-enhanced MRI with ultrafast imaging in breast cancer patients receiving neoadjuvant chemotherapy: Prediction of pathologic complete response and correlation with histologic microvessel density

Early prediction of pathologic complete response (pCR) after neoadjuvant chemotherapy (NAC) in breast cancer patients can help forecast prognosis and guide decisions on adjuvant therapy. This study aimed to determine whether the kinetic parameters of dynamic contrast-enhanced MRI (DCE-MRI) with ultrafast imaging can predict pCR following NAC in breast cancer patients and whether these parameters are correlated with histologic microvessel density (MVD). In this retrospective study, 61 breast cancer patients who underwent NAC and surgery between August 2020 and 2022 were analyzed. Ultrafast and conventional DCE-MRI features, along with pathologic results, were compared between the pCR and non-pCR groups. Regression analysis was conducted to identify predictive factors for pCR. Additionally, MRI kinetic parameters were correlated with histologic MVD. Of the 61 patients, 17 (27.9%) achieved pCR. The pCR group exhibited a larger delayed washout component (P = .002) and a smaller angiovolume (P = .02) compared to the non-pCR group; however, these factors lost significance when accounting for tumor size, lymph node status, and molecular subtypes. In a subgroup analysis based on molecular subtype, a low initial enhancement value (≤362.5%) and angiovolume (≤10.3 cc) predicted pCR in human epidermal growth factor receptor 2-enriched breast cancer, with an area under the curve of 0.833. The maximum slope on ultrafast MRI was higher in the high MVD group compared to the low MVD group (P = .049). Human epidermal growth factor receptor 2-enriched breast cancer with low vascularity on DCE-MRI is more likely to achieve pCR, although MRI kinetic parameters were not independent predictors of pCR in all breast cancer subtypes. The maximum slope on ultrafast MRI was the only kinetic parameter that correlated with histologic MVD. Larger studies focused on molecular subtypes are warranted.