FAS (TNFRSF6/CD95/APO-1) is a cell surface receptor involved in apoptotic signal transmission. Dysregulation of this pathway is believed to result in down regulation of apoptosis and subsequent persistence of a malignant clone. A single nucleotide polymorphism resulting in guanine-to-adenine transition in the FAS promoter region occurs at position −1377, affecting an SP1 transcription factor binding site; an adenine residue at position −1377 of FAS promoter significantly reduces SP1 binding compared to guanine residue causing a decrease in FAS expression. Recent data in adult leukemia indicate that a variant allele at this site increases susceptibility to AML (Morgan et al; Cancer Research 63, 4327–4330). We hypothesized that FAS genotype would also increase risk of childhood AML and, by altering susceptibility to apoptosis might impact outcome of AML therapy. Methods: 442 children treated for AML on CCG studies 2941 and 2961 (intensive induction with daunorubicin, idarubicin, Ara-C, 6-thioguanine, etoposide, dexamethasone, +/− fludarabine, Ara-C, idarubicin; consolidation with high-dose Ara-C and L-asparaginase or HLA-matched sibling BMT) were genotyped for the −1377 FAS promoter polymorphism using mismatch amplification coupled with real time PCR (MAMA assay). Genotype frequencies were compared with published normal control frequencies. In addition patient outcomes were analyzed according to genotype.
Results: Comparison of gene frequencies in AML patients and reported normal controls showed similar frequencies (78.7% GG, 18.6% GA, 2.7% AA in patients vs. 78% GG, 20% GA and 2% AA in controls; p=0.6). Stratification of cases by age at diagnosis, white blood count at diagnosis, AML subtype, or cytogenetics revealed no difference in genotype frequencies. Outcome data showed no significant difference in OS (p=0.22), EFS (p=0.31), TRM (p=0.42) or relapse rate (p=0.57) between patients with 1377GG genotype vs. 1377GA/1377AA genotypes. However, there was trend towards improved survival (OS at 5 years 75% AA, 55% GA, 49% GG; p=0.16) and EFS (66% AA, 43% GA, 39% GG; p=0.19) in AA cases, but small numbers (n=12 AA) may have contributed to the results not being statistically significant. Conclusion: These data, representing the only data in pediatric AML, suggest that FAS genotype does not affect the etiology of childhood AML and that children with AML differ from adults in terms of biology of their disease. However, further studies are needed to evaluate a potentially important role of FAS genotype in outcome of AML therapy.