ABSTRACT
This application is being submitted in response to the Notice of Special Interest (NOSI) identified as
NOT-CA-22-056. Acute lymphoblastic leukemia (ALL) is the most common pediatric cancer and, despite
advances in treatment, it is still one of the leading causes of childhood death in the United States (US), and
survivors face significant lifelong treatment-related morbidities. Children of Latino ethnicity have the highest
and fastest-increasing risk of ALL in the US, and have lower survival than non-Latino whites; however, this
ethnic disparity in incidence and outcome is not fully understood. Elucidating the increased risk of ALL in Latino
children may reveal novel insights in the etiology of ALL in both Latino and non-Latino populations, and may
highlight potential avenues for disease prevention. We hypothesize that germline genetic variation plays an
essential role in the increased ALL risk in Latinos, that this risk is imparted via Native American ancestry, and
that ALL risk alleles were selected in Native Americans during European colonization of the Americas due to
their beneficial effects on immune response to new infections. We have assembled the largest ever case-
control set of childhood ALL in Latinos, including over 5,400 cases and 27,000 controls from three independent
studies in California, plus studies in Texas, Children’s Oncology Group/St. Jude Children’s Research Hospital,
and Guatemala. In our first aim, we will perform three complementary approaches to discover novel common
risk loci associated with childhood ALL: i) a genome-wide association study (GWAS) meta-analysis, ii)
admixture mapping to capitalize on the recently admixed nature of Latino genomes, and iii) a transcriptome-
wide association study to identify novel loci and to pinpoint causal genes at known and novel risk regions. In
our second aim, we will characterize the genetic variants in terms of their association with local Native
American ancestry and whether they exhibit evidence of directional natural selection on the Native American
branch of the human population tree. We will also characterize the aggregate effects of common variants on
childhood ALL risk and how this varies by ethnicity, via comprehensive modeling of polygenic risk scores
(PRS) for ALL in Latinos and in non-Latino whites. Finally, we will incorporate our genetic findings into
epidemiologic analyses, by accounting for common (PRS) genetic variants in ALL risk models for immune-
related risk factors in Latinos and non-Latino whites, including cesarean delivery and in utero cytomegalovirus
infection, both of which have shown stronger effects on ALL risk in Latinos and are potentially modifiable risk
factors. The results of this study will shed light on the etiology of childhood ALL in general and of the increased
risk of ALL in Latino children, which will help to alleviate this ethnic disparity and may inform future approaches
for childhood leukemia prevention.