||5R03CA245998-02 Interpret this number
||De Smith, Adam
||University Of Southern California
||Exploring the Shared Genetic Architecture of White Blood Cell Variation and Childhood Acute Lymphoblastic Leukemia
Acute lymphoblastic leukemia (ALL) is a cancer of white blood cells, and the most common malignancy
in children under 15 years of age. Despite significant advances in treatment, children who survive ALL face
severe long-term therapy-related morbidities. Moreover, ALL is still one of the leading causes of cancer-related
mortality in children. Prevention of childhood ALL, therefore, remains an essential goal, and is dependent on
understanding the etiology of this disease, which forms the focus of this study.
Genome-wide association studies (GWAS) have identified genetic risk factors for childhood ALL at
several genes, many of which control white blood cell (WBC) development, including IKZF1, ARID5B, and
CEBPE. The majority of variants associated with ALL risk are also associated with variation in WBC traits,
suggesting a shared genetic basis between normal blood cell development and blood cancer. Thus, a
comprehensive study of the role of heritable WBC trait variation in risk of childhood ALL is warranted. To this
end, the first aim of this study is to catalog common genetic variants associated with variation in WBC traits,
namely counts of lymphocytes, monocytes, neutrophils, eosinophils, basophils, and total WBCs, via GWAS
using genetic and blood cell phenotype data from ~330,000 individuals of European ancestry, free of cancer
and immune-related disease, from the UK Biobank study.
The second aim will be to assess the contribution of WBC trait-associated variants to risk of childhood
ALL by: (1) testing association of these variants with childhood ALL using GWAS results from a European-
ancestry case-control study from the California Cancer Records Linkage Project, including 1184 ALL cases
and 3551 controls, with replication of novel risk loci in an independent GWAS of 959 ALL cases from the
Children's Oncology Group and 2624 controls from the Wellcome Trust Case Control Consortium; and (2)
using the WBC trait-associated variants as genetic instruments in Mendelian randomization analyses to test
the causal relevance of each WBC trait in risk of childhood ALL.
Our final aim is to identify putative causal genes associated with childhood ALL, and to assess the
shared biological pathways underlying WBC trait variation, through transcriptome-wide association studies
(TWAS) using genetic prediction models of whole blood gene expression. We will perform TWAS in: (1) the
ALL case-control study, to discover genes the expression of which is associated with ALL risk; and (2) the UK
Biobank data to identify genes whose expression is also associated with each WBC trait. This will be the first
examination of the causal relationship between WBC indices and ALL risk. Our results will shed light on the
biology of ALL, provide insight into the extent to which ALL risk genes operate via WBC-associated
mechanisms, and may highlight potential avenues for risk-stratification and prevention.