Inherited genetic susceptibility of acute lymphoblastic leukemia in Down syndrome

Austin L. Brown, Adam J. de Smith, Vincent U. Gant, Wenjian Yang, Michael E. Scheurer, Kyle M. Walsh, Jonathan M. Chernus, Noah A. Kallsen, Shanna A. Peyton, Gareth E. Davies, Erik A. Ehli, Naomi Winick, Nyla A. Heerema, Andrew J. Carroll, Michael J. Borowitz, Brent L. Wood, Wiiliam L. Carroll, Elizabeth A. Raetz, Eleanor Feingold, Meenakshi Devidas, Lisa F. Barcellos, Helen M. Hansen, Libby Morimoto, Alice Y. Kang, Ivan Smirnov, Jasmine Healy, Caroline Laverdière, Daniel Sinnett, Jeffrey W. Taub, Jillian M. Birch, Pamela Thompson, Logan G. Spector, Maria S. Pombo-de-Oliveira, Andrew T. DeWan, Charles G. Mullighan, Stephen P. Hunger, Ching-Hon Pui, Mignon L. Loh MD, Michael E. Zwick, Catherine Metayer, Xiaomei Ma, Beth A. Mueller, Stephanie L. Sherman, Joseph L. Wiemels, Mary V. Relling, Jun J. Yang, Philip J. Lupo and Karen R. Rabin

Key Points

  • IKZF1, CDKN2A, ARID5B, and GATA3 influence ALL risk in Down syndrome. CDKN2A risk allele penetrance is higher, independent of DS-ALL subtype

  • IKZF1 locus impacts enhancer activity and protein binding in B-cell super-enhancer. Knockdown leads to more proliferation in Down syndrome


Children with Down syndrome (DS) have a 20-fold increased risk of acute lymphoblastic leukemia (ALL) and demonstrate distinct somatic features, including CRLF2 rearrangement in approximately 50% of cases; however, the role of inherited genetic variation in ALL susceptibility among children with DS is unknown. Here, we report the first genome-wide association study of DS-ALL, comprising a meta-analysis of four independent studies, with 542 DS-ALL cases and 1,192 DS controls. We identified four susceptibility loci at genome-wide statistical significance: single nucleotide polymorphisms rs58923657 near IKZF1 (odds ratio [OR]=2.02, Pmeta=5.32x10-15), rs3731249 in CDKN2A (OR=3.63, Pmeta=3.91x10-10), rs7090445 in ARID5B (OR=1.60, Pmeta=8.44x10-9), and rs3781093 in GATA3 (OR=1.73, Pmeta=2.89x10-8). We performed DS-ALL versus non-DS ALL case-case analyses, comparing possible associations at these and three other established ALL susceptibility loci (BMI1, PIP4K2A, CEBPE) and found significant association with DS status for CDKN2A (OR=1.58, Pmeta=4.1x10-4). This association was maintained in separate regression models, both adjusting for and stratifying on CRLF2 overexpression, high hyperdiploidy, ETV6-RUNX1, and B-other subgroups, indicating an increased penetrance of CDKN2A risk alleles in children with DS. Finally, we investigated functional significance of the IKZF1 susceptibility locus. It maps to a B-cell super-enhancer, and the risk allele is associated with decreased enhancer activity and differential protein binding. IKZF1 knockdown resulted in significantly higher proliferation rates in Down syndrome- than non-Down syndrome lymphoblastoid cell lines. Our findings demonstrate a higher penetrance of a known ALL risk locus in children with DS and serve as a basis for further biological insights into the etiology of this disease.

  • Submitted December 12, 2018.
  • Revision received July 25, 2019.
  • Accepted July 14, 2019.