Genomic functional regions: enriched for cancer genetic risk signals

From the Lindstroem Group, Division of Public Health Sciences

Genetic variability is known to be a major contributor to cancer risk. It is estimated that 31% and 57% of risk can be explained by genetics for breast and prostate cancer, respectively. With these being two of the most commonly diagnosed types of cancer in the United States, increased understanding of their associated genetic risks could have a substantial public health impact. However, gaps remain in the identification of genetic risk signals and in the enrichment of signals among different genomic regions. Researchers from Dr. Sara Lindstroem’s research group in the Public Health Sciences Division recently published a paper in the journal Human Genetics that reports on an in-depth investigation of genetic associations of breast and prostate cancer and the functional genomic regions of genetic risk signals. With the focus on functional but non-coding DNA regions, this study is distinct from previous work that focused on genetic differences within coding regions of genes or single nucleotide polymorphisms of unknown function.

Dr. Lindstroem provided background on the study, “Only about 1.5% of our DNA is coding, making up genes that are then transcribed to proteins. For a long time, the remaining 98.5% of our genome was considered “junk DNA” without any known function. We know now that this is not true, and that a large proportion of our genome has some function in at least one cell type. In this paper, we studied regulatory regions in our DNA and how they affect breast and prostate cancer risk.”

To assess the enrichment patterns of previously identified genetic signals associated with these cancers, the authors utilized extensive genome-wide association study (GWAS) data and annotations of five categories of DNA regulatory elements: DNase1 hypersensitive sites (DHS), histone modification H3K27ac and H3K4me1, typical enhancers, and super enhancers. Hundreds of annotations were generated collectively among these five categories of regulatory elements from different cell lines and tissue types. Dr. Lindstroem described the previous research efforts of two major initiatives that enabled the work, “The first is the OncoArray network, a large international collaboration to study the genetics of multiple cancers. Basically, a group of scientists in specific cancers (e.g. breast, prostate) came together to design a genotyping array that assessed genetic variation linked to the different cancers. The collaboration made it possible for us to genotype a large number of individuals, which drove down the per-sample cost. We used the results from the main breast and prostate cancer papers that came out from the OncoArray initiative (see references). The other initiative is the ENCODE and RoadMap Epigenomics projects that mapped regulatory DNA regions across multiple cells.” The authors also employed a new and improved statistical method (GARFIELD 2.0) for their analyses.

Enrichment of GWAS-identified genetic signals categorized by five different classes of DNA regulatory elements for breast and prostate cancer.
Enrichment of GWAS-identified genetic signals categorized by five different classes of DNA regulatory elements for (A) breast and (B) prostate cancer. Results shown for p value threshold level of p < 10-8. Image from Hongjie Chen

The first step involved an overall assessment of the enrichment of GWAS signals across the five categories of DNA regulatory elements for breast and prostate cancer (see Figure). This analysis revealed a general trend for greater enrichment of signals for breast cancer compared to prostate cancer in each of the categories. Next, the authors assessed the enrichment of genetic signals across different tissue and cell types, including many that were non-breast or non-prostate derived, for three classes of the DNA regulatory elements. Dr. Lindstroem summarized the findings, “We found that regulatory regions are enriched with genetic variants that show strong association with breast and prostate cancer risk, and that in particular, regulatory DNA regions identified in breast tissue were enriched for breast cancer associations and regulatory DNA regions identified in prostate tissue were enriched for prostate cancer associations. These results demonstrate that genetic variation in regulatory DNA regions affect breast and prostate cancer risk and highlights the importance of including relevant tissue types when conducting these studies.” Specifically, the greatest enrichment of signals in regulatory elements for breast cancer were identified in the breast cancer cell lines MCF-7 and HCC1954, while those for prostate cancer were identified in the prostate cancer cell line LNCaP.

The authors plan to undertake a number of follow-up studies, “We will continue similar research as more cell-type specific annotations are being released. There were some questions we could not adequately address with the current data including a robust comparison of the importance of regulatory regions identified in normal vs. tumor tissue,” explained Dr. Lindstroem. There are also plans to expand the investigations to other types of cancers. “Although we had initially planned to include more cancer types, we only had access to comprehensive GWAS results and cell-type specific annotation data for breast and prostate cancer. We hope that as more annotations are released by the ongoing initiatives, we will be able to expand these studies,” Dr. Lindstroem added.


This work was supported by the National Institutes of Health.

Fred Hutch/UW Cancer Consortium member Dr. Sara Lindstroem contributed to this research.

Chen H, Kichaev G, Bien SA, MacDonald JW, Wang L, Bammler TK, Auer P, Pasaniuc B, Lindstrom S. 2019. Genetic associations of breast and prostate cancer are enriched for regulatory elements identified in disease-related tissues. Human Genetics. doi: 10.1007/s00439-019-02041-5.

Additional references:
Michailidou K, et al. Association analysis identifies 65 new breast cancer risk loci. Nature. 2017 Nov 2;551(7678):92-94. doi: 10.1038/nature24284. 

Schumacher FR, et al.  Association analyses of more than 140,000 men identify 63 new prostate cancer susceptibility loci. Nature Genetics. 2018 Jul;50(7):928-936. doi: 10.1038/s41588-018-0142-8. 

Roadmap Epigenomics Consortium, et al. Integrative analysis of 111 reference human epigenomes. Nature. 2015 Feb 19;518(7539):317-30. doi: 10.1038/nature14248. 

ENCODE Project Consortium. An integrated encyclopedia of DNA elements in the human genome. Nature. 2012 Sep 6;489(7414):57-74. doi: 10.1038/nature11247.