Repressive histone tail modifications have been previously characterized in breast cancer, but the relationship between these marks, tumor features and breast cancer molecular subtypes has not been studied in depth. We have tissue microarray data with marker information on 971 incident breast cancers that developed from 1976-2000 in women enrolled in the Nurses’ Health Study. Tissue microarray sections were stained for repressive H3K9 trimethylation (H3K9me3) and H3K27 trimethylation (H3K27me3) marks, and tumors with 50% or more positively stained cells were scored as histone mark positive. We found that 73.3% of tumors were H3K9me3 positive, and 84.9% of tumors were H3K27me3 positive. The relationships between histone marks and tumor features (tumor size, grade, stage, nodal involvement and hormone receptor status) were analyzed. While we found no association between H3K9me3 and tumor features, age-adjusted analyses suggested that H3K27me3 was associated with lower grade tumors (OR=5.51, 95% CI 2.64-11.50), p<0.0001), estrogen receptor-positive (ER+) tumors (OR=2.40, 95% CI 1.54-3.74, p=0.0001) and progesterone receptor-positive (PR+) tumors (OR=1.96, 95% CI 1.31-2.93, p=0.0012). In multivariate analyses, a strong association was observed between H3K27me3 and lower grade comparing grade 1 to grade 3 tumors as the referent (OR=5.17, 95% CI 2.12-12.62, p=0.0003) after adjusting for age at diagnosis, age at menarche, age at menopause, parity or age at first birth, family history of breast cancer in a first-degree relative, personal history of benign breast disease, menopausal status or PMH use, BMI at age 18, alcohol consumption, lactation and smoking. We also found associations between H3K27me3 and ER+ tumors (OR= 2.28, 95% CI 1.32-3.93, p=0.003) and with PR+ tumors (OR=2.14, 95% CI 1.30-3.50) in the fully adjusted model. The relationships between histone marks and breast cancer molecular subtypes were also analyzed. Molecular subtypes were defined using immunohistochemistry staining for markers, and grade. H3K27me3 was associated with Luminal A subtype compared to all other subtypes as the referent (OR=2.16, 95% CI 1.41-3.32, p=0.0004) in age-adjusted analyses and remained significant in the fully adjusted model (OR=1.72, 95% CI 1.04-2.88, p=0.0359). Our results indicate that H3K27me3, but perhaps not H3K9me3, is independently associated with lower grade and ER+ tumors. Our results are also suggestive of a relationship between H3K27me3 and Luminal A subtype. Overall, our findings indicate that repressive histone marks may influence specific tumor features and molecular pathways in breast cancer etiology, which may help to explain the basis of disease heterogeneity and further elucidate aberrant epigenetic mechanisms in breast cancer.

Citation Format: Megan A. Healey, Andrew H. Beck, Stuart J. Schnitt, Rulla M. Tamimi, Aditi Hazra. Repressive histone marks in breast cancer in relation to molecular phenotype: results from the Nurses’ Health Study. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3637. doi:10.1158/1538-7445.AM2013-3637