Preliminary Application Of Magnetic Resonance Diffusion Kurtosis Imaging (DKI) In The Diagnosis Of Breast Cancer

Part ?The value of the diffusion kurtosis imaging in differential diagnosis of benign and malignant breast lesionsObjective:To compare the difference of diffusion kurtosis imaging(DKI)parameters between benign and malignant breast lesions and to evaluate DKI in the differentiation and characterisation of breast lesions.Methods:70 patients with 72 breast lesions who underwent MRI conventional scan,diffusion weighted imaging(DWI)and DKI scan before surgery were collected from July 2015 to September 2016.All lesions were confirmed by biopsy or operation,including 36 benign lesions and 36 malignant lesions.Malignant lesions were:ductal carcinomain situ(DCIS)(3);invasive ductal carcinoma(IDC)(30);infiltrating mixed mucinouscarcinoma(2);basal-likebreastcarcinoma(BLBCs)(1).36 benign lesions included 23 cases of fibroadenoma,5 cases of plasma cell mastitis(PCM),4 cases of breast intraductal papilloma,abscess in 1 case,cyst disease in 3 cases.The DKI scan is based on the non-Gaussian distribution model,all parameters were measured at the workstation,ncluding the mean kurtosis(MK)and mean diffusivity(MD)of malignant lesions and benign lesions.The independent sample t test and mann-whitney U test were used to compare the parameters between malignant lesions and benign lesions.Receiver operating characteristic(ROC)was used to compare the diagnostic capability of ADC,MK and MD values.Results:There were significant statistical differences in ADC,MK and MD values between malignant and benign lesions.The values of MD and ADC in benign lesions were higher than those in malignant lesions(P<0.05).The value of MK in malignant lesions were higher than those in benign lesions(P<0.05).The areas under curve(AUC)of ADC,MK and MD were 0.940,0.945 and 0.729,respectively.Using the 1.12?0.77?1.35 as the cut off of the evalution of benign and malignant lesions respectively,the sensitivity was 100.00%,86.10%,80.60%and the specificity was 80.60%,94.40%,63.90%respectively.Conclusion:As a complement to the ADC value,MK and MD values have potential clinical value in differentiating benign and malignant breast lesions.Part ?The correlation between the quantitative parameters of MRI and biological factorsObjective:To explore the correlation between the quantitative parameters(ADC,MK and MD values)of magnetic resonance imaging(MRI)and prognostic factors(ER,PR,CerbB-2 and Ki-67)of breast cancer,to find MRI features which can predicted the molecular phenotypic expression.Methods:34 patients with 36 breast cancers confirmed by pathology were collected,including 3 cases of ductal carcinomain situ,30 cases of invasive ductal carcinoma(IDC),2 cases of infiltratingmixed mucinouscarcinoma,1 cases of basal-likebreastcarcinom(BLBCs).All patients underwent MRI conventional scan,diffusion weighted imaging(DWI)and DKI scan.The apparent diffusion coefficient(ADC),mean kurtosis(MK)and mean diffusivity(MD)values were measured by DKE and' mricron software.The breast cancer biological factors(ER,PR,CerbB-2 and Ki-67)were obtained by immunohistochemistry.According to the expression status of ER?PR?CerbB-2?Ki-67,all these cases were divided into positive group and negative group,the independent sample t test and mann-whitney U test were used to compare the parameters between two groups.The correlations between quantitative parameters(ADC,MK,MD values)and expression of biological factors used Spearman rank correlation analysis.Results:ADC and MK values were higher in breast cancer with CerbB-2 expression in positive than CerbB-2 expression in negative(P<0.05),and MK value was also higher in breast cancer with Ki-67 expression in positive than PR expression in negative(P<0.05).The expression of ADC and MK was positively correlated with the expression of CerbB-2 in breast cancer(r=0.339,P<0.05;r=0.522,P<0.05).MD value of breast cancer and expression of Ki-67 was negati-vely correlated(r=-0.511,P<0.05).MK were positively correlated with Ki-67 in breast cancer(r=0.566,P<0.05).Conclusion:The quantitative functional parameters of MRI(ADC,MK and MD values)were related to the expression of biological factors CerbB-2 and Ki-67 in breast cancer,which may be helpful for providing a basis for individual treatment options and prognosis prediction.