oncology icon
Enrichment of CpG island shore region hypermethylation in epigenetic breast field cancerisation

Breast cancer causes the greatest number of cancer-related deaths among women and is the most diagnosed non-keratinocyte cancer. In 2018, approximately 15% of all cancer deaths among women were caused by breast cancer. In the US alone over 40,000 women died from breast cancer in 2019. Assessment of methylation in ductal carcinoma in situ (DCIS) has shown differences at early stages of the disease, however many of these studies have been conducted using normal adjacent tissue as the comparator. Very early changes in the methylome may mean that normal adjacent tissue is not representative of normal breast epithelium. A recent study identified changes in methylation in normal adjacent tumour breast tissue, which suggested that field cancerisation effects may confound our understanding of early carcinogenesis.

In this study, the authors investigated potential field effects using patient matched tissue from laser capture microdissection of both ipsilateral-normal and contralateral-normal tissue. Tissue was microdissected from archival FFPE tissue blocks from treatment naïve women who had elected for a mastectomy.

DNA from laser capture microdissected tissue was bisulphite treated and hybridised to the Illumina Infinium MethylationEPIC arrays. Data was analysed using the R minifi package and after QC resulted in 683,209 CpG loci from 25 matched samples from nine women, eight samples of tumour tissue, eight ipsilateral normal and nine contralateral-normal tissue.

Unsupervised hierarchical clustering of the top 10,000 most variable probes demonstrated separation of tumour and normal samples, using both ipsilateral or contralateral.

Methylation of Alu was decreased significantly in tumour tissue compared to contralateral-normal tissue but not ipsilateral-normal tissue. Alu methylation was not significantly different between the two normal sample types. There were no significant differences detected for methylation of long interspersed nucleotide element-1 (LINE-1).

38,346 differentially methylated CpG loci were detected in tumour tissue compared to ipsilateral-normal tissue and 63,271 CpG loci in tumour tissue compared to contralateral-normal, 33,657 were common to both analyses. In terms of gene numbers associated with the methylated loci, 9,800 unique genes were mapped using ipsilateral-normal as the control and 12,887 using contralateral-normal with 9,283 common to both.

9,562 CpG loci were detected as differentially methylated between the two control cell types. 4,962 were hypomethylated in ipsilateral-normal compared to contralateral-normal.

Locus overlap analysis (LOLA) of these hypomethylated loci against a background of MCF-7 and T47D datasets detected enrichment of transcription factor binding sites (TFBS) for c‑Fos, ESR1, FoxA1, GATA3, RAD21, RARA, RARG, PGR and SRC‑3. These transcription factors have been associated with early events of tumour formation in breast cancer.

4,620 hypermethylated loci were detected in ipsilateral-normal relative to contralateral-normal, however there were no significant enrichments of TFBSs detected.

Further analyses of these loci identified significant depletion of CpG island regions and enrichment for CpG island shore regions and south shores as well as a modest enrichment of open sea regions.

The authors suggest that early changes in DNA methylation may occur at shore regions adjacent to CpG islands and that methylation then spreads to the island regions.

Investigation of the 20 most significantly hyper shore CpG loci in the ipsilateral normal tissue compared to tumour by analysis of the distribution of methylation at these loci and their adjacent CpG islands identified hypermethylation of tumour suppressor genes FAAP20, SST and VGLL4. The analysis also demonstrated hypermthylation spreading into the CpG island for SST and VGLL4 in the tumour samples.

This study demonstrated that normal tissue from the tumour harbouring breast is epigenetically different from normal tissue of the opposite breast despite histology appearing normal in both cases. This study suggests that a field cancerisation effect may be occurring in the normal tissue surrounding the tumour and that early epigenetic changes are observed in the CpG islands and their adjacent shore regions that may impact the expression of oncogenes and tumour suppressor genes. Normal adjacent tissue is often used as a matched control for epigenetic assessments of tumour tissue but this study suggests they may not be the ideal reference material as they may already harbour some of the early markers of carcinogenesis.


Muse et al, 2020. Enrichment of CpG island shore region hypermethylation in epigenetic breast field cancerization. Epigenetics


Want to ask a question about

Laser Capture Microdissection or Oncology?

oncology icon
Epistem’s Epigenetics and LCM Services

Epistem is a GCLP-accredited laboratory specialising in providing biomarker, target discovery and personalised medicine information. We offer microarray, qPCR and NGS services for gene expression, whole genome and epigenetic analysis in all species. In addition to being able to conduct methylation analysis using NGS, Epistem also offers Illumina Infinium MethylationEPIC arrays.

We also specialise in laser capture microdissection (LCM) for the isolation of specific highly enriched populations of cells of interest for subsequent transcriptional profiling and epigenetic analysis. We have GCLP-accredited histology and IHC labs and we have used our expertise in this area to develop “RNA-friendly” stains compatible with LCM to isolate specific cell types, making them amenable for gene expression studies. All of our histology and gene expression, LCM and hair IHC applications are GCLP-compliant and we have supported many clinical studies. We also have in-house bioinformatics support for all of our genomics studies and have extensive experience identifying biomarkers or biomarker signatures in a variety of tissues.

About Epistem

Epistem's contract research service is committed to providing reliable, innovative and transferable pre-clinical models and services to support decision making throughout the drug discovery and development pipeline.

Tel: +44 (0)161 850 7600  Email: info@epistem.co.uk