Variant to Function Mapping of B-ALL Risk Loci

ABSTRACT Acute lymphoblastic leukemia (ALL) is the most common childhood malignancy and arises predominantly from developing B-cell precursors. B-cell ALL (B-ALL) remains a leading cause of cancer-related mortality in children, and B-ALL survivors face long-term therapy-related morbidities. Moreover, B-ALL disproportionately impacts

children of Hispanic/Latino ethnicity, who have the highest risk of disease and inferior patient outcomes compared to non-Hispanic white children. Prevention of B-ALL remains an essential goal, to mitigate childhood mortality and suffering and to reduce disparities, and this necessitates an understanding of the disease’s root

causes. Paramount to this will be elucidating the mechanisms of genetic predisposition to B-ALL. Previous genome-wide association studies (GWAS) have identified at least 15 risk loci associated with childhood B-ALL, but in most cases the causal variants have not been identified. Our proposal aims to integrate results from a

large multi-ancestry GWAS of childhood ALL with single-cell epigenomic data followed by targeted base editing in hematopoietic stem and progenitor cells (HSPCs), to pinpoint causal variants and discern their impact on normal human B cell development. Through this innovative and collaborative approach, we hypothesize that we

will uncover mechanistic insights into the root causes of B-ALL development that could inform future preventive strategies. In our first aim, we will conduct a multi-ancestry GWAS followed by statistical fine-mapping analysis of known B-ALL risk loci using existing genotype data. These results will be integrated with single-cell accessible

chromatin data spanning human hematopoiesis and B cell development using the SCAVENGE method we have developed previously, to identify specific B cell developmental stages that underlie B-ALL predisposition and to identify causal variants at each B-ALL risk locus. In our second aim, we will generate new single-cell multi-omic

data, including from joint scRNA- and scATAC-seq in single cells, from human HSPCs undergoing B cell differentiation in vitro. These data will enable regulatory mapping in cells across B cell development, including rare intermediate cells, and will be used to map relevant cell states and target genes for the fine-mapped B-ALL

risk variants. Finally, in our third aim, we will investigate the functional impact of B-ALL risk variants using genome editing. We will prioritize six causal variants from our other mapping approaches, including variants in ARID5B and GATA3 that have previously been identified as causal and that contribute to the increased B-ALL risk in

Hispanic/Latino children. We will recreate these six variants individually and in specific combinations via base editing in human HSPCs from multiple donors that will then be subject to B cell differentiation and investigate their effects on B cell development and on illegitimate V(D)J recombination, a known driver of B-cell

leukemogenesis. Our innovative variant-to-function mapping approach will reveal the precise mechanisms through which B-ALL association loci may increase a child’s risk of developing leukemia and may highlight novel therapeutic targets and inform future approaches for childhood leukemia prevention.