基于扩散峰度成像初探HER-2表达浸润性乳腺癌瘤内瘤周弥散信息OA

Objective:To investigate the feasibility of using quantitative intratumoral and peritumoral parameters derived from diffusion kurtosis imaging(DKI)for discriminating breast cancers with different HER-2 expression levels.Methods:A ret-rospective analysis was performed on 150 female patients with pathologically confirmed unilateral solitary invasive breast cancer(from December 2019 to December 2023).Based on HER-2 expression,patients were divided into three groups:over-expres-sion,low-expression,and zero-expression.All patients underwent preoperative breast DKI.The largest cross-section of the sol-id tumor component was selected,and regions of interest(ROIs)were manually drawn within the tumor(ROI1),and at 2 mm(ROI2)and 5 mm(ROI3)beyond the tumor border.The maximum,minimum,and mean values of mean kurtosis(MK)and mean diffusivity(MD)were recorded for each ROI.The Kruskal-Wallis test and one-way ANOVA were used to compare intratumoral and peritumoral MK and MD values among groups.The diagnostic performance of individual parameters and combined models was evaluated using receiver operating characteristic(ROC)curves and the Delong test,with P<0.05 considered statistically sig-nificant.Results:For ROI1,the zero-expression group showed the highest MDmax and MDmean values,significantly greater than those in the low-expression and over-expression groups(P=0.010,0.019).For ROI2,the zero-expression group also had the highest MDmax and MDmean values(P=0.023,P=0.001),but the lowest MKmean value(P=0.034).In ROI3,the zero-expression group exhibited the highest MDmax value(P=0.029).The AUC values for distinguishing zero-expression from over-expression groups us-ing parameters from ROI1,ROI2 and ROI3 were 0.623,0.714 and 0.659,respectively.Combined models(ROI1+ROI2,ROI1+ROI3 and all three)yielded AUCs of 0.768,0.696 and 0.767,respectively,all significantly higher than those of single param-eters(P<0.05).Conclusion:DKI-derived diffusion parameters from both intratumoral and peritumoral regions can help differenti-ate invasive breast cancers with varying HER2 expression levels,providing new imaging evidence to guide endocrine therapy.