A130: Molecular profiling of normal breast epithelial cells to quantify breast cancer risk

Natalie Shenker1,Gillian Weaver2,Luca Magnani1,Van Nguyen1,Robert Brown1,James Flanagan1

1Imperial College London, London, UK,2QCCH Milk Bank, Imperial Healthcare NHS Trust, London, UK

Presenting date: Monday 2 November
Presenting time: 13.10-14.00

Background

One key barrier to epigenetic epidemiological studies has been the tissue-specific nature of epigenetic patterns. To gain a mechanistic understanding of breast cancer risk, it is crucial to assess the association between known cancer risk factors with epigenetic marks in normal breast epithelial cells. Therefore, we have initiated the Breast milk Epigenetics Cohort Study (BECS), aiming to collect and store expressed breast milk (EBM) and exfoliated breast epithelial cells along with questionnaire data on breast cancer risk factors.

Method

We have recruited 156 subjects to the pilot phase of this cohort study. Flow cytometry was used to quantify the percentage of epithelial (EMA+) and oestrogen receptor positive (ER+) cells. We performed molecular profiling using the Illumina 450k methylation array (n=36), array-based genotyping (n=80), gene expression profiling (n=8) and chromatin immunoprecipitation followed by high-throughput sequencing for ER binding (n=3).

Results

We have shown that the donated breast milk samples contain ~97% epithelial cells and >85% ER-positive cells, as confirmed by flow cytometry and qRT-PCR for ESR1 [1]. ChIP-seq data for ER identified unique ER binding sites across the genome that were almost mutually exclusive of known core ER binding sites in breast tumours. Using the methylation profiling data, we have validated previously reported associations between methylation-genotype (meth-QTLs) in breast cancer risk alleles and have developed a novel cell type specific age-signature to calculate the apparent methylomic aging rate (AMAR) in each of the individuals. Furthermore, bisulphite pyrosequencing assays showed that smoking induces methylation changes in breast epithelial cells detectable in former smokers (average, 5 years post-quitting) compared to never smokers.

Conclusion

This pilot study has demonstrated that it is possible to collect highly pure epithelial cells from breast milk samples donated to a hospital milk bank with sufficient quality to perform a range of molecular analyses required to assess novel breast cancer risk biomarkers. This supports the development of a collaborative cohort study to further investigate novel breast cancer risk biomarkers and may offer new opportunities for research.