Oral Presentation 43rd Lorne Genome Conference 2022

Integrating GWAS and 3D chromatin interactome data to identify multi-cancer risk genes in hormone-related cancers (#22)

Sarahi Rivera 1 2 , Jonathan Beesley 1 , Haran Sivakumaran 1 , Mahdi M Marjaneh 1 3 , Kristine M Hillman 1 , Sneha Nair 1 , Susanne Kaufmann 1 , Stacey Edwards 1 , Juliet French 1
  1. QIMR Berghofer, Herston, QUEENSLAND, Australia
  2. School of Biomedical Science and Institute of Health and Biomedical Innovation, Faculty of Health, Queensland University of Technology (QUT),, Brisbane, Queensland, Australia
  3. UK Dementia Research Institute, London, United Kingdom

Breast, endometrial, ovarian and prostate are hormone-related cancers that account for over 15% of all cancer deaths. To date, GWAS has identified 150 breast, 16 endometrial, 37 ovarian and 167 prostate cancer-associated genetic variants. In this study, we observed that those variants are often in close proximity suggesting common mechanisms underlying these cancers. We identified 45 multi-cancer risk regions (MCRRs) defined as genomic regions that contain variants associated with risk of two or more hormone-related cancers. Disease-associated variants frequently fall in DNA regulatory elements, such as enhancers, that can lie up to 1Mb from their target gene promoter. Therefore, we hypothesized that within MCRRs, genetic variants associated with different cancer types modulate the expression of common target genes through altered tissue-specific regulatory elements. To enrich for chromatin interactions between gene promoters and enhancers within MCRRs in relevant cell types, we performed promoter capture HiC (PCHiC) in twelve immortalised ‘normal’ and cancer cell lines. To identify MCR genes we integrated PCHiC interactions and risk-associated SNPs within MCRRs. We identified 78 candidate MCR genes associated with at least two hormone-related cancers. This list includes established cancer driver genes such as MYC and CCND1, but also >20 genes with little or no reported role in cancer. One example is MLLT10, a candidate breast and ovarian cancer risk gene. MLLT10 encodes a transcription factor that interacts with DOT1L, which methylates H3K79, a histone mark associated with active transcription. Reporter assays show risk SNPs induced MLLT10 promoter activity, suggesting increased MLLT10 contributes to cancer risk. CRISPR-based functional studies are currently underway to elucidate the role of MLLT10 in breast and ovarian cancer development. We anticipate that some of the identified multi-cancer risk genes may provide new drug targets for future prevention or treatment of hormone-related cancers.