定量MRI预测Ⅰ～Ⅲ期乳腺癌分子亚型和预后的可行性研究OA北大核心CSTPCD

Objective:To explore the feasibility and value of intravoxel incoherent motion(IVIM)and dynamic contrast en-hanced MRI(DCE-MRI)in evaluating molecular subtypes and prognosis of stage Ⅰ～Ⅲ breast cancer.Methods:From January 2020 to March 2022,data of 144 patients with breast cancer confirmed by surgical pathology in the Affiliated Hospital of Jining Medical University were retrospectively analyzed.All patients received IVIM and DCE-MRI scans,and the quantitative parameters of IVIM(true diffusion coefficient(D),pseudo diffusion coeffificient(D*),perfusion fraction(f))and the quantitative pa-rameters of DCE-MRI(volume tansfer constant(Ktrans),rate constant(Kep),extracellular extravascular volume fraction(Ve))were measured.Independent risk factors for predicting molecular subtypes and prognosis were screened using logistic regression and the predictive models were built.The receiver operator characteristic(ROC)curves were drawn to evaluate the diagnostic effi-ciency of models,and DeLong test was used to compare area under the ROC curve(AUC).Results:Logistic regression results showed that D,Ktrans and Ve were independent risk factors for Luminal and triple-negative types,respectively(P=0.003,<0.001,0.002;P=0.027,0.002,<0.001).D+Ktrans+Ve showed the best prediction for Lunimal(AUC=0.883)and triple-negative subtypes(AUC=0.850).Meanwhile,D and Ktrans were independent risk factors for HER2 overexpression type,and the D+Ktrans model showed an optimal diagnostic efficiency,with an AUC of 0.913.Additionally,Ktrans was an independent factor for predicting Nottingham prognostic index(NPI),with an AUC of 0.790;and D was the only independent risk factor for predicting the ex-pression of antigen Ki-67 with an AUC of 0.747.Conclusions:The combined IVIM and DCE-MRI diagnostic model showed favorable diagnostic efficiency in predicting molecular subtypes and prognosis of stage Ⅰ～Ⅲ breast cancer,which might be helpful with the clinical therapy decision-making in a certain degree.