Germ-Line Genetic Variations in TP53 and Risk for Pediatric Acute Lymphoblastic Leukemia.

The tumor suppressor p53 plays an essential role in the prevention of tumor development by its ability to respond to stress signals, resulting in the inhibition of cell growth, via cell-cycle arrest or induction of apoptosis. p53 also contributes to the repair of genotoxic DNA damages and is involved in cellular senescence. Previous studies have shown that both germ-line and somatic mutations in the TP53 gene, which encodes p53, can occur in a wide spectrum of cancers, including acute lymphoblastic leukemia, ALL. Recently, we characterized the common germ-line genetic variation across TP53 by extensive re-sequence analysis (http://snp500cancer.nci.nih.gov).

Patients and Methods: We conducted a genetic association study in childhood ALL to investigate the possible contribution of common single nucleotide polymorphisms, SNPs, and the common haplotypes on which they reside. For TP53, we analyzed haplotype-tagging SNPs, which are estimated to capture greater than 95% of common haplotypes across the gene (IVS2+38C, Ex4+119 (R72P), IVS4-91, Ex6-34 (R213R), IVS6+62, IVS6-36 and 1846bp 3′ of STP). Cases of ALL (n = 513) were drawn from anonymized samples from the German co-operative therapy study (COALL-06-97) and 751 healthy anonymized blood donor controls collected from regional blood banks in Germany. Genotype analysis was performed by validated Taqman assays posted on the SNP500cancer website. Haplotypes were deduced by maximization-estimation analysis of unphased genotypes using PHASEv2.0.1 (http://www.stat.washington.edu/stephens/software.html).

Results: The single locus analysis revealed an increased risk for ALL for four of the seven SNPs individually: C allele at IVS2+38C (304/982 in patients vs. 399/1476 in controls, OR 1.21 95% CI 1.01 – 1.45), G allele at IVS4-91 (150/988 vs. 168/1456, OR 1.37 95% CI 1.08 – 1.75), A allele at Ex6-34 (R213R) (164/996 vs. 178/1450, OR 1.41 95% CI 1.11 – 1.78) and T allele at 1846bp 3′ of STP (185/990 vs. 203/1420, OR 1.38 95% CI 1.10 – 1.72). Individuals homozygous AA at IVS6+62 (16/350 vs. 9/556, OR 2.82, 95% CI 1.16 – 6.99) or TT at 1846bp 3′ of STP (24/334 vs. 15/522 OR 2.50, 955 CI 1.24 – 5.09) showed further risk for disease. The non-synonymous coding SNP at Ex4+119 (R72P) did not show any association with ALL. The distribution of the inferred haplotypes differed significantly between cases and controls (global haplotype test p = 0.02). Children carrying the haplotype CCGAAGT have an increased risk for ALL (138/611 vs. 155/972, OR 1.42, 95% CI 1.09 −1.83) in reference to the most common haplotype GGAAGGC.

Conclusions: Our data suggest that susceptibility to pediatric ALL could be associated with at least one copy of an “at risk” haplotype. The study results imply that germ-line genetic variation in TP53, a key tumor suppressor gene, could contribute to susceptibility to childhood ALL. Additional studies are required to confirm our findings and to characterize the functional elements in the at-risk haplotypes.