To compare apparent diffusion coefficient (ADC) values of malign breast lesions with different histopathological types on diffusion weighted magnetic resonance imaging (DWI) and cellular region/stroma (CR/S) ratio and histopathological results

Purpose: To compare Apparent Diffusion Coefficient (ADC) values of malign breast lesions with different histopathological types on diffusion weighted imaging (DWI) and cellular region/stroma (CR/S) ratio and histopathological results. Materials and methods: Between August 2011 -October 2013, breast magnetic resonance imaging (MRI) and DWI findings of 59 patients were evaluated retrospectively at Dokuz Eylul University Hospital radiology Department. The largest size of breast mass identified as BI-RADS 4,5 and 6 was 7.5 cm. Patients who had prior treatment or intervention regarding their breast masses were not included to the study. Breast MRI were performed by 1.5 Tesla MRI device (Gyroscan Achive Intera Philips). In order to calculate CR/S ratio, breast wide excional biyopsy or mastectomy specimen slices were stained with Haematoxylin and Eosin and evaluated by histopathological image atlas editor software. For statistical analysis, Statistical Package for Social Sciences for windows 15 package was used. Statistical significance level was accepted as p<0.05. Results: Fifty nine lesions including malign breast lesions and normal breast tissue ADC values were calculated using ''b'' factor 1000 s/mm². Mean ADC values were 1.725 ± 0.508 x10ˉ³mm²/sn for normal breast tissues and 1.098 ± 0.410 x10ˉ³mm²/sn for malign breast lesions. Mean ADC values for intraductal carcinoma (IDC) (n=25) and intralobulary carcinoma (ILC) (n=17) were calculated as 1.078 ± 0.294 x10ˉ³mm²/sn and 1.151 ± 0.340 x10ˉ³mm²/sn respectively. Receiver operating caharavteristic (ROC) analyses was performed for malign lesions, IDC lesions, and ILC lesions separately. The threshold ADC value to differentiate malignant and benign masses was 1.391 x10ˉ³mm²/sn with 79.7% sensitivity and 81.4% specificity (AUC=85.7%). Likewise, threshold ADC value for IDC was 1.260 x10ˉ³mm²/sn with 80.8% sensitivity and 81.4% specificity; threshold ADC value for ILC was 1.391 x10ˉ³mm²/sn' with 8.2% sensitivity and 79.5% specificity (AUC= 87.2% and 84.3% respectively). There was no significance in the ADC values of IDC and ILC (p=0.45). The median value for CR/S ratio was 0.46. It was divided into two groups; values above the median value was classified as high and below the median value as low CR/S ratio. There was significant relationship and positive correlation between histopathological grade and CR/S ratio according to Spearman Correlation Test (p<0.0001) (r=+0.583). ADC value in lesions with low CR/S ratio (n=21) was 1.135 ± 0.429 x10ˉ³mm²/sn while ADC value in high CR/S ratio group (n=18) was 1.155 ± 0.429 x10ˉ³mm²/sn. Although it is not statistically significant, there was pozitif correlation between ADC values and CR/S ratio (p=0.62; r=+ 0.08). Conclusion: Adding ADC and DWI to conventional breast MRI demonstrates high sensitivity and specificity in order to differentiate normal breast tissue from histopathological subtypes of malign lesions; nonetheless, ADC value calculation does not seem to be used as an alternative for histopathology detecetion which is the gold standard for differentiation of subtypes of malign breast cancer. In addition, as there is a positive correlation between CR/S ratio and ADC values, it might be a strong marker to use in order to evaluate the stromal component of the lesions.