Epigenetic and genetic burden measures are associated with tumor characteristics in invasive breast carcinoma. Issue 5 (3rd May 2016)

• Record Type: Journal Article
• Authors: O'Sullivan, Dylan E.; Johnson, Kevin C.; Skinner, Lucy; Koestler, Devin C.; Christensen, Brock C.
• Is Part Of: Epigenetics. Volume 11:Issue 5(2016)
• Page Start: 344
• Page End: 353
• Publication Date: 2016-05-03
• Subjects: Breast cancer -- DNA methylation -- deregulation -- dysregulation -- epigenome -- TCGA; Epigenesis -- Periodicals; Epigenetica; 572.86505
• DOI: 10.1080/15592294.2016.1168673
• Languages: English
• ISSNs: 1559-2294

Abstract

ABSTRACT: The development and progression of invasive breast cancer is characterized by alterations to the genome and epigenome. However, the relationship between breast tumor characteristics, disease subtypes, and patient outcomes with the cumulative burden of these molecular alterations are not well characterized. We determined the average departure of tumor DNA methylation from adjacent normal breast DNA methylation using Illumina 450K methylation data from 700 invasive breast tumors and 90 adjacent normal breast tissues in The Cancer Genome Atlas. From this we generated a novel summary measure of altered DNA methylation, the DNA methylation dysregulation index (MDI), and examined the relation of MDI with tumor characteristics and summary measures that quantify cumulative burden of genetic mutation and copy number alterations. Our analysis revealed that MDI was significantly associated with tumor stage ( P = 0.017). Across invasive breast tumor subtypes we observed significant differences in genome-wide DNA MDIs ( P = 4.9E–09) and in a fraction of the genome with copy number alterations (FGA) ( P = 4.6E–03). Results from a linear regression adjusted for subject age, tumor stage, and estimated tumor purity indicated a positive significant association of MDI with both MCB and FGA ( P = 0.036 and P < 2.2E–16). A recursively partitioned mixture model of all 3 somatic alteration burden measures resulted in classes of tumors whose epigenetic and genetic burden profile were associated with the PAM50 subtype and mutations in TP53, PIK3CA, and CDH1 . Together, our work presents a novel framework for characterizing the epigenetic burden and adds to the understanding of the aggregate impact of epigenetic and genetic alterations in breast cancer.