| Part IThe diagnostic value of intravoxel incoherent motion model of diffusion weighted imaging in differentiating benign from malignant breast lesions[Objective] To investigate the diagnostic value of parameters derived from diffusion-weighted imaging intravoxel incoherent motion (IVIM) model for benign and malignant breast lesions, and compared these parameters with that of monoexponential model and time-signal intensity curve (TIC) derived from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI).[Methods] 102 cases (35 cases with a total of 36 benign lesions; 67 cases with a total of 68 malignant lesions) of breast disease with breast MRI examination in our hospital were collected. All of them were examined with single b value diffusion-weighted imaging (DWI), multiple-b value DWI and DCE-MRI sequences scans. The multiple-b value DWI was performed with 12 b values from 0-1000 s/mm2. GE AW4.6 workstation was used to automatically calculate the value of ADC, D, D* and f by multiple-b value DWI images and to generate TIC using the DCE-MRI images. All parameters between benign and malignant breast lesions were compared by Mann-Whitney U test. We used the method of Binary Logistic regression to obtain the combined diagnostic parameter. The diagnostic performance of different parameters was evaluated by receiver operating characteristic curve (ROC curve) analysis, and then find the optimal value, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV).[Results] The median value of ADC, D for malignant lesions were 0.99×10-3 mm2/s and 0.85×10-3 mm2/s respectively, and they were significantly lower than those for benign lesions (1.38X 10"3mm2/s and 1.24×10-3mm2/s, respectively). The median value of f of malignant lesions was significantly higher than that of benign ones (24.60% vs.16.90%). There were all statistically significant differences between these parameters (P<0.001). There was a big overlap between the D* value in differentiating benign and malignant lesions and there was no statistically significant difference between them (P=0.064). The ratio of type I was the highest in benign lesions, while type III in malignant lesions. With ROC curve analysis for each parameter value, the area under the curve (AUC) of D value was 0.910 which was the highest one, and corresponding diagnostic sensitivity, specificity, PPV and NPV were 80.9%,88.9%,93.2% and 71.1% while taking the optimal threshold of 0.966×10-3mm2/s. AUC value can be increased to 0.965 by combining D with f value, and the corresponding diagnostic sensitivity, specificity, PPV and NPV were increased to 86.8%,94.4%,96.7% and 79.1% respectively. When the curve of type Ⅱ+Ⅲ was set as the criteria for malignant lesions, the diagnostic sensitivity can be as high as 95.6%. while only the curve of type Ⅲ was set as the criteria, the diagnostic specificity can be as high as 86.1%.[Conclusion] The parameters derived from the IVIM model can be used to distinguish benign from malignant breast lesions. The D and f value had excellent diagnostic performance. Associated with D and f value can obtain better diagnostic performance than the ADC value and TIC.Part IIDiffusion weighted imaging intravoxel incoherent motion model:does it correlate with prognostic factors and subtypes of breast cancers[Objective] To investigate whether parameters of intravoxel incoherent motion (IVIM) model correlate with prognostic factors and subtypes of breast cancers, thus to explore its noninvasive evaluation value in prognostic factors and subtypes.[Methods] From March 2014 to May 2015,110 cases, histologically confirmed breast invasive ductal cancer, were collected. All of them were examined with multiple-b value DWI before surgery or core needle biopsy. GE AW4.6 workstation was used to automatically calculate the value of D, D* and f deriving from IVIM model. Correlation between IVIM parameters and prognostic factors (including tumor size, vascular invasive status, axillary lymph node status, tumor grade, the expression status of estrogen receptor (ER), progesterone receptor (PR), HER2 and Ki67) and molecular subtype which was judged by immunohistochemistry results was analyzed.[Results] The average age of the 110 patients, which were enrolled, was 47.4±9.3 years. 60 of them were premenopausal women, while the rest were postmenopausal women. In all the 114 lesions which belong to 110 breast cancer patients, there were 56 lesions in the left breast and 58 lesions in the right one. The median D value of the low aggressive group (including histological grade of ductal carcinoma in situ and Grade 1), PR-negative group and HER2-positive group (including scores of 3+and 2+combined with FISH gene amplification) were significantly higher than that of the highly aggressive group (including histological grade of Grade 2 and 3), PR-positive group and HER2-negative group (including scores of-,1+and 2+without FISH gene amplification) (1.11×10’3 mm2s vs.0.78×10-3 mm2s, P<0.001; 0.87×10-3 mm2/s vs. 0.80×1O-3 mm2/s, P=0.042; 0.90×10-3 mm2/s vs.0.79×10-3 mm2/s, P=0.001). The median D* value of HER2-positive lesions was significantly higher than that of HER2-negative ones (21.8×10-3 mm2/s vs.10.5×10-3 mm2/s; P=0.033). The median f value of the low aggressive lesions was significantly lower than that of the highly aggressive ones (20.3% vs.27.3%, P<0.001). The median D value of HER2 enriched subtype tumor (1.11 × 10-3 mm2/s) was significantly higher than that of Luminal and triple-negative subtype one (0.80 ×10-3 mm2/s,0.82×10-3 mm2/s; respectively), there were statistically significant differences among them (P=0.026, P=0.048; respectively).[Conclusion] D value deriving from IVIM model was negatively correlated with tumor grade and PR status, while positively correlated with HER2 status. D* value was positively correlated with HER2 status, and f value was positively correlated with tumor grade. There were no correlation among the parameters of IVIM model and other prognostic factors. There was potential value for D to distinguish the subtype of HER2 enriched from that of luminal and triple-negatinve.PartⅢRole of the intravoxel incoherent motion diffusion weighted imaging in the pre-treatment prediction and early response monitoring to neoadjuvant chemotherapy in locally advanced breast cancer[Purpose] To explore whether intravoxel incoherent model (IVIM) diffusion weighted imaging (DWI) can probe pre-treatment differences or monitor early response in patients with locally advanced breast cancer receiving neoadjuvant chemotherapy (NAC).[Materials and Methods] Fifty-seven patients with locally advanced breast cancer were imaged using multiple-b DWI with 12 b values ranging from 0 to 1000 s/mm2 at the baseline, and forty-nine patients were repeatedly scanned after the second cycle of NAC. Subjects were divided into pathologic complete response (pCR) and non-pathologic complete response (non-pCR) groups according to the surgical pathologic specimen. Parameters (D, D*, f, maximum diameter (MD) and short diameter (SD)) before and after 2 cycles of NAC and their corresponding change (Δparameter) between pCR and non-pCR groups were compared using the Student’s t test or non-parametric test. The diagnostic performance of different parameters was judged by the receiver operating characteristic curve analysis.[Results] Prior to NAC, the f value of pCR group was significantly higher than that of non-pCR (35.50% vs.23.40%, P<0.001). At the end of the second cycle of NAC, the D value was significantly higher and the f value was significantly lower in pCR than that in non-pCR (1.38±0.25x10-3mm2/s vs.0.98±0.19×10-3mm2/s, P<0.001; 15.19%±4.73% vs.21.62%±4.43%, P<0.001, respectively), whereas the D* value, MD and SD of the pCR group was slightly lower than that of the non-pCR group (P=0.204; P=0.053; P=0.080, respectively). ΔD was higher in pCR (-0.47×10-3mm2/s) than that in non-pCR (-0.07x10-3mm2/s) after 2 cycles of NAC (P<0.001). Δf value in the pCR group was significantly higher than that in the non-pCR group (17.45% vs.3.10%, P<0.001). There was no significant difference in AD* between the pCR and non-pCR group (P=0.451). The prediction performance of ΔD value was the highest (AUC=0.928,95%CI=0.817 to 0.982). When the optimal cut-off was set at-0.261 X 10"3mm2/s, the values for sensitivity, specificity, positive predictive value and negative predictive value were up to 90.0%, 89.7%,85.7% and 92.9%, respectively.[Conclusion] IVIM-derived parameters, especially the D and f value, showed potential value in the pre-treatment prediction and early response monitoring to NAC in locally advanced breast cancer. ΔD value had the best prediction performance for pathologic response after NAC. |
PDF Full Text Request