|ECR 2019 / C-1620|
|Usefulness of fusion MRI in ADC analysis after neoadjuvant chemotherapy in patients with breast cancer|
Methods and materials
Patients population: in our retrospective study we enrolled 50 patients with histological diagnosis of breast cancer by core needle biopsy before NACT . They underwent two breast MRI exam in the period from June 2015 to October 2018, one before the fist NACT administration and the other after 2-3 chemotherapy cycles, about 6 months after the first MRI exam/ corresponding about at 6 months later than the first MRI exam .
According to the histological classification of breast cancer, our 50 case were comprised of 42 invasive ductal carcinomas, 5 invasive lobular carcinomas, 2 invasive micropapillary carcinoma and 1 carcinomatous lymphangitis.
10 patients were later excluded from the study because they had one breast - MRI in other hospital or they were claustrophobic or they stopped the MRI exam.
Pre-chemotherapy clinical stages of the 40 remaining patients included: 23 patients T2a, 8 patients T2b, 3 patients T3a, 3 patients T3b.
Data acquisition: 30 breast MRI exams were performed using a 1.5 Tesla (GE SIGNA 1.5T MR EXCITE) and 10 breast MRI exams were performed using 3Tesla (GE SIGNA 3T DISCOVERY MR 750). Patients were placed in prone position. At 1.5 Tesla the MRI protocol consisted of: axial T2 with fat suppression (Ax T2 FSE TE=102, TR= 380, 3 NEX, Frequence =256, Phase= 224, FOV=37, Slice Thickness=4, Spacing=1); COR DWI with b-values of 0, 800 TE minimum, TR=10000, 6 NEX, Frequence =92, Phase=92, FOV=40, Slice Thickness=4,5, Phase=1; BILATERAL 3D T1 FSHR ( Dynamic Contrast Enhanced, DCE) with TE in phase, Flip Angel=10°, 1 NEX, Frequence=320, Phase =1 and BILATERAL 3D T1 FSHR Special.
The 3Tesla the MRI protocol consisted of: axial T2 FSE IDEAL Slice thickness 3, Spacing 0.3, Phase FOV 1, TE 120, Matrix 320x288, NEX 1; Ax DWI 50-800-1200, slice thickness 3, Spacing 0.3, TR 7000, b value 3 (50 NEX 2, 800 NEX 3, 1200 NEX 6), ACCELERATION PHASE 2, Matrix 96x198; Ax VIBRANT MPH+C (Dotarem 0,5 M or Gadovist 0,1 M) slice thickness 2, Matrix 360x360, ACCELERATION 2 and ax VIBRANT ISO slice thickness 2, Matrix 360x360, ACCELERATION 2.
In 1.5 Tesla MRI pre-contrast examinations ADC maps were automatically calculated from b=0, b=800 DWI series and at 3 Tesla MRI pre-contrast examinations ADC maps were calculated from b=0 and b=1200 s/mm2 DWI series by dedicated software.
Data analysis: MR imaging results were reviewed by experienced radiologists of San Salvatore Hospital of L’Aquila who were blinded to pathologic findings and treatment responses to NACT.
In each examination , diameters and volumes of each breast tumor were measured using axial morphologic sequences (T2) and DCE, then ADC values were calculated, in the post-processing phase, by overlapping the ADC map and DCE through FUSION technique (AW Volume Share 7-Release AW4.7 Ext 10 Software – Property of GE Company USA ).
ADC values, used for the study, were obtained as a mean of three ROIs (circle of 5-10 mm2) randomly traced in tumor lesion. Fusion technique was very useful to exclude cystic areas, necrosis, liquefactions, or hemorrhage tissue in patients with reduction of tumor lesion >80%. (Fig 1- Fig.2).
Statistical analysis: To evaluate the percentage of response we obtained tumor volumes mean, before and after chemotherapy, and, at the same time, we obtained a mean of ADC values before (ADC-pre) and after (ADC-post) chemotherapy. Than we compared volumes before and after chemotherapy and we obtained the reduction percentage.
Mean values of ADC-pre and ADC-post were compared through non parametric statistical test which were Wilcoxon test .
A P value less than 0.05 was considered statistically significant.
Thematically related posters
ECR 2019 / C-0525
Influence of ROI placement: different ADC metrics on diagnostic accuracy of DWI in NME breast tumours
ECR 2019 / C-1117
Multiparametric MRI with dynamic contrast enhanced MRI and diffusion weighted imaging for the assessment of non-mass enhancing breast tumours