Grant Number:	1R01CA309511-01A1 Interpret this number
Primary Investigator:	Darst, Burcu
Organization:	Fred Hutchinson Cancer Center
Project Title:	Harnessing Large Biobanks to Decipher Cancer Pleiotropy
Fiscal Year:	2026

PROJECT SUMMARY Genome-wide association studies (GWAS) have identified numerous pleiotropic cancer risk regions, offering compelling opportunities to identify shared genetic mechanisms underlying cancer risk and etiology. However, most cancer GWAS only include a single cancer type, limiting our insights into shared genetic risk factors. Our previous work demonstrates the power of jointly studying multiple cancer types to discover novel germline susceptibility regions. This is particularly crucial for GWAS of rare cancers, where a notable paucity in research and lack of sufficient sample sizes have led to a critical knowledge gap. Leveraging existing genomic data across multiple cancer types provides a cost-efficient platform for conducting in-depth cancer pleiotropy studies at scale. Rich resources such as the UK Biobank and All of Us, which recently integrated whole-genome sequencing (WGS) data, as well as growing tissue- and cell-specific omics data resources, present unprecedented opportunities to advance our knowledge of the shared and distinct mechanisms contributing to cancer risk. We propose to comprehensively interrogate cancer pleiotropy and shared heritability by leveraging multiple existing data sources, including WGS data for >193K cancer cases, individual-level GWAS data for >244K cases, and GWAS summary statistics based on >1.1M cases in our discovery analyses. To validate our findings, we will conduct replication analyses leveraging newly launched initiatives which are currently accruing data for an additional 785K cancer cases with WGS or GWAS data. In Aim 1, we will conduct multi-cancer GWAS and fine-mapping to identify and characterize novel pleiotropic cancer susceptibility regions, with a comprehensive investigation of the highly pleiotropic human leukocyte antigen (HLA) cancer region. In Aim 2, we will apply a multi-pronged bioinformatics approach to characterize pleiotropic cancer regions through gene set enrichment and network analyses and conduct in silico analyses integrating single-cell omics data to prioritize putative causal variants and cell types. For a selected set of regions, we will conduct in vitro follow-up experiments, recognizing the limited budget allowed for functional validation in PAR-25-095. In Aim 3, we will derive latent genetic components underlying genome-wide genetic correlations between cancers to better characterize shared cancer risk. We will project rare cancer GWAS onto the latent component space to boost our power to detect genetic variants associated with rare cancers. We will also develop pathway-specific cancer polygenic risk scores to identify biological components of cancer risk that are shared and distinct across cancers. Our project is highly cost-efficient and leverages existing data at an unprecedented scale to close critical gaps in our understanding of cancer pleiotropy. Our results will fuel additional computational and laboratory- based follow-up studies and catalyze future projects within the broader scientific community. Ultimately, the proposed work will allow us to distinguish shared from site-specific mechanisms in cancer development, and to advance universal vs personalized preventive and therapeutic strategies.





None