E-GEOD-43379 - Molecular Characterization of the Vasculature of Patients with Infiltrating Ductal Carcinoma generates a Gene Signature predictive of Breast Cancer Survival

Released on 1 April 2013, last updated on 2 June 2014
Homo sapiens
Samples (80)
Array (1)
Protocols (6)
Introduction: Targeting of the tumor vasculature has evolved into an integral part of existing standard anti-cancer therapies. However, recent pre-clinical and clinical studies have shown that the efficacy of these treatments is temporary at most and frequently followed by the renewed tumor growth, indicating that these targets may not be robust. Methods: We transcriptionally profiled laser capture microdissection isolated microvessels from tumor and matching adjacent normal tissue samples obtained from breast cancer patients diagnosed with infiltrating ductal carcinoma. Gene expression profiles were generated and analyzed using Genespring and the web-based WebGestalt bioinformatics Toolkit. Several differentially expressed genes were validated by immunohistochemistry. To further mine our data set we used a novel integrative systems biology network approach to identify a gene signature that could robustly stratify breast cancer patient populations. Results: Vascular enrichment of the LCM samples was confirmed by Q-PCR of CD31. Hierarchical two-dimensional clustering of the transcriptome data identified 219 significantly upregulated transcripts in at least 4/8 patient samples (cut-off >1.3-fold, p<0.05). Several genes AGRN, FLNA and ILR4 were selected and their protein overexpression was confirmed in the tumor tissue. Additional mining of the data set using the clinical information associated with these tissues by applying an integrative bioinformatics network method generated a 15-gene ‘Vascular-Derived Prognostic Predictor’ that is able to stratify patient populations in to high- and low- risk groups. Conclusions: The overall heterogeneity observed in the vascular gene expression patterns in the infiltrating ductal carcinoma samples poses a significant problem for these individual genes to be used as a robust vascular breast cancer specific marker or therapeutic target. Using an integrative bioinformatics network approach allowed us to identify a 15-gene ‘Vascular-Derived Prognostic Predictor’ that can robustly identify patients with an increased risk of local recurrence. Stratifying patient populations using this gene signature may lead to better patient-tailored treatment regimes. Two-color microarrays comparing laser-captured microvessels from the tumors and matching adjacent normal tissues of 8 breast cancer patients
Experiment type
transcription profiling by array 
Andre Nantel, Anne E Lenferink, Catherine Collins, Christiane Cantin, Maureen D O’Connor-McCourt, Phuong U Le
Investigation descriptionE-GEOD-43379.idf.txt
Sample and data relationshipE-GEOD-43379.sdrf.txt
Raw data (1)E-GEOD-43379.raw.1.zip
Processed data (1)E-GEOD-43379.processed.1.zip
Array designA-GEOD-16481.adf.txt