ABSTRACT
A high impact of ARID5B SNPs on acute lymphoblastic leukemia (ALL) susceptibility has been described in Hispanic children; therefore, it is relevant to know if they influence the high incidence of childhood-ALL in Mexicans. Seven SNPs (rs10821936, rs10994982, rs7089424, rs2393732, rs2393782, rs2893881, rs4948488) of ARID5B were analyzed in 384 controls and 298 ALL children using genomic DNA and TaqMan probes. The SNPs were analyzed for deviation of Hardy-Weinberg equilibrium; Fisher's exact test was used to compare the genotypic and allelic frequencies between controls and patients. The association between SNPs and ALL susceptibility was calculated, and haplotype and ancestry analyses were conducted. All SNPs were associated with ALL, pre-B ALL, and hyperdiploid-ALL susceptibility (p < 0.05). No association with T-ALL and gene fusions was found (p > 0.05). The seven SNPs were associated with risk of pre-B ALL in younger children; however, rs2393732, rs2393782, rs2893881, and rs4948488 were not associated with susceptibility in older children and adolescents. The CAG haplotype (rs10821936, rs10994982, rs7089424) was strongly associated with ALL risk in our population (p < 0.00001). The frequency of all risk alleles in our ALL, pre-B, and hyperdiploid-ALL patients was higher than that in Hispanic children reported. This is the first report showing the association between rs2393732, rs2393782, and rs4948488 with pre-B hyperdiploid-ALL children. The G allele at rs2893881 confers major risk for pre-B hyperdiploid-ALL in Mexican (OR, 2.29) than in Hispanic children (OR, 1.71). The genetic background of our population could influence the susceptibility to ALL and explain its high incidence in Mexico.
