Abstract

Abstract Background: Ductal carcinoma in situ (DCIS) currently accounts for 20% of new breast cancer diagnoses. DCIS is typically treated with surgery and radiation therapy, despite having a much better prognosis than invasive disease. Previous studies have documented features and biomarkers in DCIS that are associated with a higher risk of invasion, and have shown that the tumor microenvironment plays a role in invasive progression of DCIS. In this study we leveraged publicly available gene expression datasets in a meta-analysis to identify biological processes underlying differences in invasive potential. We focused on studies using laser capture microdissection to elucidate these differences within both the epithelial and stromal compartments. Methods: All studies included in this analysis were publicly available datasets downloaded from the Gene Expression Omnibus. Studies were included if they reported gene expression of laser-capture microdissected human breast tissue processed on Affymetrix microarrays and included samples that had the histology of IDC or DCIS. We included 5 studies in our analysis, totaling 83 DCIS epithelial samples, 67 IDC epithelial samples, 26 normal epithelial samples, 22 DCIS stromal samples, 20 IDC stromal samples, and 14 normal stromal samples. The raw datasets were normalized using the Robust Multiarray Average algorithm, and batch effects associated with study and platform were corrected using ComBat. To assess differences in gene expression between DCIS and IDC, we conducted differential gene expression analysis using a t-test and Benjamini-Hochberg correction. Next, we used the DAVID Bioinformatics functional annotation tool to identify enriched pathways in the differentially expressed (DE) gene sets. Finally, hierarchical clustering was used to identify groups of samples within the epithelial and stromal groups based on the DE genes. Results: We found 422 and 118 genes that were differentially expressed (FDR < 0.05) between DCIS and IDC, in epithelium and stroma respectively. Interestingly, 102 of the 118 DE genes in the stroma were downregulated in IDC compared DCIS. Within the epithelial DE genes, pathways related to extracellular matrix reorganization, collagen catabolic processes, and cell-matrix adhesion were upregulated in IDC compared to DCIS. Within the stromal DE genes, cell-adhesion, leukocyte migration, and the Ras/MAPK pathways were downregulated in IDC compared to DCIS. When hierarchically clustered based on DE genes, we identify 6 clusters. Two sets of three clusters associated perfectly with histology (stromal vs epithelium). Within each set of histological clusters, two clusters were comprised entirely or nearly entirely (>97%) of specimens from patients with either IDC or DCIS, while the last cluster (mixed) was comprised of samples from both disease subtypes. Interestingly, only a minority (26%) of stromal samples were classified as mixed, in contrast to a majority of epithelial samples (53%). Conclusions: Our meta-analysis identified DE gene sets that were enriched in processes relating to the tumor microenvironment in both the epithelial and stromal components. Many of the DE genes between DCIS and IDC epithelia were related to collagen production, likely due to the inclusion of fibroblasts in the dissected IDC epithelia. Further analysis will seek to extricate underlying biological changes from differences occurring due to sample dissection and processing. The widespread downregulation of genes in the stroma of IDC supports prior theories that invasion may be related to the loss of a suppressive environment in the stroma. Our results suggest that evaluation of the stromal component of DCIS may be critical in understanding its invasive potential and could therefore guide treatment recommendations. Citation Format: Andre Dempsey, Christina Yau, Olivier Harismendy, Jonah Donnenfield, Amrita Basu, Michael Campbell, Alexander D Borowsky, Gillian Hirst, Laura J Esserman. Gene expression changes in the epithelium and stroma of invasive ductal carcinoma and ductal carcinoma in situ: A meta-analysis [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P6-06-07.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call