Font Size: a A A

Correlation Between The Breast Imaging Reporting And Data System Of Ultrasonography And The Molecular Subtypes Of Breast Cancer

Posted on:2016-09-05Degree:MasterType:Thesis
Country:ChinaCandidate:K XuFull Text:PDF
GTID:2284330461951657Subject:Medical imaging and nuclear medicine
Abstract/Summary:PDF Full Text Request
Background and Purpose:Analysis of gene expression arrays has revealed that breast cancer is a highly heterogeneous disease on the molecular level and under the regulation of hormone receptors. The 13 th St Gallen International Breast Cancer Conference(2013) Expert Panel adopted a new approach to the classification of patients based on the recognition of intrinsic biological subtypes subtypes within the breast cancer staining with immunohistochemisty. It includes four fundamentally different subtypes of cancer in terms of aetiology, prognosis and treatment response. A recent study have confirmed different molecular subtypes of breast showed distinct MRI features and pathologic characteristics. But a few study on the correlation of ultrasonographic features and molecular subtypes. In our study, the Breast Imaging Reporting and Data System(BI-RADS) instituted by American Association of Radiology(ACR) was applied to investigate the ultrasonographic features and pathological features of different molecular of breast cancer. Enhance the understanding of ultrasonographic features of breast cancer and provide more valuable information to determine the clinical treatment and the clinical prognosis. Materials and Methods:The data of 175 lesions which underwent primary breast cancer resection were reviewed.All of the patients had US preoperatively.The molecular subtypes of breast cancer defined by immunohistochemistry( IHC) and fluorescence in situ hybridization(FISH) were classified as Luminal A-like subtypes(i.e ER and PR positive, HER2 negative,Ki-67 low expression <14%), Luminal B-like subtypes(i.e ER and PR positive, HER2 negative or positive,Ki-67 high expression >14%),HER-2 overexpression subtypes(i.e. HER2 positive; ER and PR may be positive or negative)and Basal-like subtypes.(i.e. ER, PR and HER2 negative). Due to Luminal A-like type and Luminal B-like type with similar patterns of gene expression and biological characteristics, this study was unified defined as Luminal-like subtype. According to Classification Nottingham, Scored the proportion of glandular tissue-like structures, mitotic count and cell atypia, breast cancer was divided into three levels according to the resulting scores: level I(well differentiated, low-grade), Ⅱ grade(moderately differentiated, intermediate grade), grade Ⅲ(poorly differentiated, highly malignant).The characteristics of the tumors according to the BI-RADS-US(tumor morphology, orientation, margin, internal echo, posterior echo feathers, calcification and structural changes in the surrounding tissue) and histologic types and grades of tumors were analyzed with Chi-square test.Consistency of description lesions with BI-RADS between observer were compared with the intra-class correlation coefficien( ICC); α=0.05. Comparison between groups was considered statistically significant with P <0.05, the group pairwise comparisons using the corrected P value, was considered statistically significant with P <0.017.②The data of 79 lesions which underwent primary breast cancer resection were reviewed.All of the patients had US and SWE to easured the average Young’s modulus values of lesions(Emean) preoperatively by two high qualification sonographer follow the principle of double-blind.The molecular subtypes of breast cancer defined by immunohistochemistry and FISH were classified as Luminal-like subtypes, HER-2 overexpression subtypes and Basal-like subtypes. Consistency of measure of Emean between observer were compared with the intra-class correlation coefficien( ICC) Speannan rank analysis used to analysis the correlation of breast lesions Emean and molecular subtype, tumor size, histological grade and clinical stage. Results:Luminal-like subtypes were 89(50.9%),HER-2 overexpression subtypes were 34(50.9%),Basal-like subtypes were 52(29.7%)in 175 lesions. The shape of the breast lesions, between orientation, internal echo, calcification, vascular supply and lymphatic metastasis judged by two observers excellent consistency; the edge of the breast lesions, posterior echo, the structural changes in the surrounding tissue surround structural changes judged by two observers good consistency. The ultrasonographic features like tumor morphology(c2=17.5, P < 0.01),margin(c2=46.5, P < 0.01),internal echo(c2=24.9, P < 0.01),posterior echo feathers(c2=36.2,P<0.01) and calcification(c2=15.736,P<0.01),tumor size(c2=33.5,P<0.01)in each subtype were significantly different. However, the difference for orientation(c2=3.18,P=0.059),Associated features(c2=2.17,P=0.078),in each subtype did not reach significance. Luminal-like subtypes compared with the Her-2 overexpression subtypes, Luminal-like subtypes likely diameter <20mm(51.7% vs. 17.6%, P = 0.000). HER-2 overexpression likely posterior echo attenuation(41.2% vs.11.2%, P = 0.000), with lymph node metastasis(73.5% vs.23.6%, P =0.000).The difference of other ultrasonographic features was not statistically significant(P> 0.017). Luminal-like subtypes compared with the Basal-like subtypes, Luminal-like subtypes likely spicuated(50.6% vs.9.6%, P = 0.000), lack of blood supply(59.6% vs.13.5%, P = 0.000). Basal-like mostly edge lobulated(69.2% vs. 27.0%, P = 0.000) and more posterior acoustic enhancement(42.3% vs.18.0%, P = 0.001); no internal calcification(82.7% vs. 53.9%, P = 0.003), a ring blood flow of the edge(55.8% vs. 7.9%, P = 0.000). The difference of other ultrasonographic features was not statistically significant(P> 0.017). HER-2 overexpression subtypes compared with the Basal-like, HER-2 overexpression subtypes edge angular(23.5% vs.9.6%, P = 0.010); likely microcalcification in mass(52.9% vs.17.3%, P = 0.0108%, P = 0.003). Basal-like subtypes likely posterior acoustic enhancement(51.5% vs.11.8%, P = 0.001). The difference of other ultrasonographic features was not statistically significant(P> 0.017).②Luminal subtypes were 42(53.2%),HER-2 overexpression subtypes were 17(21.5%),Basal-like subtypes were 20(25.3%)in 79 lesions. Emean of breast lesions range of 24.2 ~ 126.5 k Pa, the average of is(69.7 ± 35.9) k Pa. Correlation analysis showed that: betwween Emean lesions and lesion size(r = 0.638), histological grade(r = 0.479), the clinical stage(r = 0.684), lymph node metastasis(r = 0.703), molecular subtypes(r = 0.417) were significantly positive correlation. Emean of Luminal-like subtypes breast cancer is lower than Emean of Her-2 overexpressing subtypes breast cancer and Basal-like breast cancers. Conclusion:1.Luminal-like subtypes breast cancer likely showed spiculated and tumor diameter <20mm.2.Her-2 overexpressing subtypes breast cancer likely showed shadowing of posterior feature, microcalcification in mass, lymph node metastasis.3.Basal-like subtypes breast cancer usually with circumscribed or indistinct margin, less spiculated, but microlobulated, anechoic necrotic zone in mass visible, less calcification in mass, a ring blood flow of the edge.4.SWE measurement Emean were significantly positively correlated with molecular subtypes of breast cancer, tumor size, histological grade, clinical stage, lymph node metastasis.
Keywords/Search Tags:breast cancer, molecular subtypes, BI-RADS-US, Young’s modulus, shear wave elastography
PDF Full Text Request
Related items