Aim: In the AIEOP-BFM ALL (Associazione Italiana Ematologia Oncologia Pediatrica-Berlin Frankfurt Münster acute lymphoblastic leukemia) 2000 protocol, 70% of relapsed patients had favorable prognostic features and fell within less intensive polychemotherapeutic regimens, suggesting the need for better assessing lower risk stratification. Materials & methods: A novel two-phase study design selected 614 children to be genotyped for TNF-α SNP rs1800629 (-308G>A). A weighted Cox model was applied to evaluate the SNP effect on hazard of relapse, adjusting for immunophenotype, risk group, age and gender and including interaction terms. Results: Significant interaction was found with immunophenotypes (p = 0.0007, with minor allele genotypes being adverse genetic markers in B-cell acute lymphoblastic leukemia and protective ones in T-cell acute lymphoblastic leukemia), and also with risk protocols (p = 0.0041, with minor allele genotypes as prognostic factor of relapse for standard risk patients [only one T-cell acute lymphoblastic leukemia in the subgroup analyzed]). Conclusion: The presence of at least one A allele in TNF-α SNP rs1800629 should suggest a closer monitoring in B-cell acute lymphoblastic leukemia standard risk patients.
