Integrative Genomic Profiling To Identify Drivers of Breast Cancer-Associated Inflammation
Michael A Seidman, Susan C Lester, Nicole B Johnson, Kimberly H Allison, Yunn-Yi Chen, Rachel E Factor, Gary MK Tse, Sandra J Shin, David A Eberhard, Puay Hoon Tan, Stuart J Schnitt, Laura C Collins, Kristin C Jensen, Konstanty Korski, Frederic M Waldman, Andrew H Beck. Cancer Genome Atlas (TCGA) Breast Cancer Expert Pathology Committee, Bethesda, MD
Background: Host immune response to tumors is variable between individuals and is a strong prognostic factor in breast cancer, with increased lymphocytic inflammation associated with improved prognosis. The tumor genomic determinants of host inflammation are not well understood.
Design: To determine genomic factors associated with the inflammatory response in breast cancer, we assembled a Breast Cancer Expert Pathology Committee to score inflammation on digital whole slide images of breast cancer specimens undergoing molecular profiling as part of The Cancer Genome Atlas project. Inflammation was scored as either present or insignificant/absent on 281 cases. We analyzed genome-wide DNA copy number variation data (both normalized log2 ratio values and GISTIC-processed copy number calls) and expression profiling data (acquired both by microarray and by RNAseq) to identify genes that were significantly associated (positively or negatively) with inflammation by both DNA copy number and RNA expression, and thus restricted to tumor cells. This analysis was implemented by performing a series of two-class SAM analyses and identifying genes significantly associated, either positively (inducers) or negatively (repressors), with inflammation in all of the analyses.
Results: This analysis identified 1271 genes significantly associated with host inflammation by both RNA expression and DNA copy number in all analyses at a false discovery rate threshold of 5%. We performed gene set enrichment analysis of this gene signature, including 1081 genes where DNA copy number gain and over-expression is associated with increased inflammation (i.e. inducers) and 190 genes where DNA copy number loss and under-expression are associated with increased inflammation (i.e. repressors). The gene-set is significantly enriched for pathways related to cell cycle regulation, transcription, and DNA repair, pathways related to breast cancer pathogenesis, and pathways relating to leukocyte migration and T-cell function. Additionally, several drug associated gene signatures were identified as enriched in the inflammation-associated genomic signature. These include a subset of down-regulated genes also reportedly down-regulated by specific HDAC1 inhibitors, suggesting a possible adjuvant therapeutic for breast cancer.
Conclusions: This integrative analysis of breast cancer genomic data and host inflammatory response provides insights into the biology of tumor associated inflammation as well as identifies candidates for new therapeutic approaches to modulate host inflammatory response to breast cancer.
Category: Special Category - Pan-genomic/Pan-proteomic approaches to Cancer
Monday, March 4, 2013 9:30 AM
Poster Session I Stowell-Orbison/Surgical Pathology/Autopsy Awards Poster Session # 278, Monday Morning