In:
Cancer Research, American Association for Cancer Research (AACR), Vol. 72, No. 8_Supplement ( 2012-04-15), p. 5540-5540
Abstract:
Abstract: Introduction. Molecular pathology tools are increasingly important in routine diagnostic. Histological determination remains mandatory, while gene expression profiles (GEP) are successfully classifying breast cancer specimens also when minimal amount of tissue, such as core biopsies (CB), is available. The limit of this application resides however in the fact that the part of material used for RNA extraction could not represent the entire biopsy. To overcome this problem in the present work we calculated an algorithm for the identification of malignant cells in a giving CB. Patients and methods. The quantitative expression levels of 60 selected genes representative for relevant pathways in breast cancer (BC) oncogenesis were assessed by real time PCR in a total of 106 CB samples obtained from patients with neoplasia or invasive BC and in 12 biopsies with known good differentiation grade (Grade 1). The GEP of 48 CBs and of the 12 Grade 1 specimens were used as training set to build the scoring system based on the Welsh variant of t-test and best-fit logistic curves. The remaining GEPs were used as validation set. Results. A 15-genes classifier was built and cross-validated with 1000 iterations of a leave −10% out in which each of the 10 subset is used. The 15 genes were representative for proliferation (thymidine phosphorylase and E2F1), cell matrix degradation (CTSD, MMP1, MMP11, PLAU, PLAUR, SERPINE1), as well as angiogenesis (VEGFA, VEGFD, TGFbeta1). EGFR and GSTP1 were also part of this classifier able to discriminate non-malignant from malignant breast tissue with an accuracy of almost 90%. Discussion. The calculated classifier, which could also be reduced to only six genes, is robust and useful in the discrimination of malignant from non-malignant breast cancer tissue. Such an algorithm identifying normal proliferation rate and inability of invasion should be integrated in the evaluation of GEP data from microarrays before using them for other predictive purposes. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5540. doi:1538-7445.AM2012-5540
Type of Medium:
Online Resource
ISSN:
0008-5472
,
1538-7445
DOI:
10.1158/1538-7445.AM2012-5540
Language:
English
Publisher:
American Association for Cancer Research (AACR)
Publication Date:
2012
detail.hit.zdb_id:
2036785-5
detail.hit.zdb_id:
1432-1
detail.hit.zdb_id:
410466-3
Permalink