Abstract 6618: Pleiotropic GATA3 locus is associated with multiple childhood cancers: harnessing the existing data from the St. Jude Lifetime Cohort and the Childhood Cancer Survivor Study for genetic etiology research

Abstract

Abstract Background: A genetic variant (rs3824662) at GATA3 was previously associated with risk of Philadelphia chromosome-like acute lymphoblastic leukemia (Ph-like ALL) and relapse. Our recently published childhood Hodgkin lymphoma (HL) study identified a genome-wide significant association for another genetic variant (rs3781093) in high linkage disequilibrium (D’=1.0, R2=0.91) with rs3824662. We hypothesize that GATA3 is a pleiotropic locus associated with multiple childhood cancers. Methods: The study was performed by combining 3 cohorts of childhood cancer survivors: the St Jude Lifetime Cohort Study (SJLIFE), the original and expansion cohorts of the Childhood Cancer Survivor Study (CCSS). The CCSS original cohort had SNP array genotyping and imputation, and the SJLIFE and CCSS expansion cohort underwent whole-genome sequencing. Associations between the variants mapped to GATA3 and risks of different cancer types were analyzed (a specific cancer type was selected as the case group and all other unselected cancer types served as controls, then repeated for 13 cancer types). Summary statistics were subsequently analyzed with the ASSET (Association analysis for SubSETs) method which is powerful for pooling association signals across multiple cancer types when true effects may exist only in a subset and be in opposite directions across cancer types. Results: Genotype data were available for 11623 childhood cancer survivors (5325 from the CCSS original cohort, 2428 from the CCSS expansion cohort, and 3870 from the SJLIFE, all European ancestry) were initially considered. After excluding individuals with dual enrollment or as outliers in the principal components analysis (n=667), a total of 10956 individuals remained for genetic association analysis. We identified a bi-directional association for a subset of childhood cancers with rs3781093 (2-sided P value of 6.22 × 10−19): the same effect allele (C) was associated with increased risk of ALL or soft tissue sarcoma [STS] (OR=1.21; 95% CI=1.13-1.29; P=7.4 × 10−9), and was associated with decreased risk of HL (OR=0.66; 95% CI=0.59-074; P=1.80 × 10−12). The frequencies of the effect allele (C) for rs3781093 in ALL cases (17.9%), STS cases (17.8%) and HL cases (12.5%) relative to non-cancer community controls in the SJLIFE (15.6%) and non-cancer gnomAD samples (14.9%) corroborate this finding. In our combined dataset, there were 1380 Hodgkin lymphoma cases, 2961 ALL cases and 692 STS cases. Conclusion: We discovered a novel variant in the GATA3 locus exhibiting pleiotropy with both tumor-promoting and protective effects, suggesting a shared mechanistic pathway involving GATA3 may underlie the carcinogenesis for ALL, STS and HL. Conducting systematic searches for childhood cancer pleiotropy represents an important research direction. Citation Format: Cheng Chen, Xijun Zhang, Qian Dong, Heather L. Mulder, John Easton, Xiaotu Ma, Jinghui Zhang, Jun Yang, Kim E. Nichols, Gregory T. Armstong, Kirsten K. Ness, Melissa M. Hudson, Hui Wang, Nilanjan Chatterjee, Cindy Im, Zhaoming Wang. Pleiotropic GATA3 locus is associated with multiple childhood cancers: harnessing the existing data from the St. Jude Lifetime Cohort and the Childhood Cancer Survivor Study for genetic etiology research [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6618.