Transferrin receptor and hereditary hemochromatosis gene variants interact to modify childhood leukemia risk. A. E. Kennedy1, E. DeRycke1, G. Y. Lai1, K. Y. Kamdar2, P. J. Lupo2, M. F. Okcu2, M. E. Scheurer2, M. K. Baum3, D. Seminara1, M. T. Dorak4 1) Epidemiology and Genomics Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, NIH, Rockville, MD; 2) Section of Hematology-Oncology, Department of Pediatrics, Baylor College of Medicine, Houston, TX; 3) Department of Dietetics and Nutrition, Robert Stempel College of Public Health & Social Work, Florida International University, Miami, FL; 4) School of Health Sciences, Liverpool Hope University, Liverpool, United Kingdom.

   An interaction between the transferrin receptor (TFRC) rs3817672 (S142G) and hereditary hemochromatosis gene (HFE) rs1800672 (C282Y) is associated with increased risk for multiple myeloma, breast, colorectal and hepatocellular cancers as well as childhood acute lymphoblastic leukemia (ALL) risk in European studies. HFE C282Y and H63D correlate with body iron levels. Since HFE and TFRC proteins biologically interact in iron transfer across membranes, the observed statistical interaction suggests that the involvement of HFE variants in cancer risk modification is mediated via their synergistic effect on body iron levels. We sought to replicate the HFExTFRC interaction in childhood ALL in a US-based study, and used bioinformatic tools to assess the contribution of TFRC S142G to this association. Genotyping was conducted in a multi-ethnic sample from Houston, TX (161 incident cases with childhood ALL and 231 controls, all <18 yr). Being positive for either HFE variant yielded an elevated odds ratio for childhood ALL risk in males (1.4, 95% CI=0.8 to 2.4), which increased to 3.0 (95% CI=1.3 to 6.8) in the presence of the S142G allele A homozygosity (Pinteraction=0.04). A comprehensive screening of empirical and bioinformatic data showed that allele A shows allele-specific expression, the non-conservative amino acid substitution is within the helix portion of the protein and predicted to alter the secondary structure (but not in the transferrin or HFE binding sites), and the SNP alters a splicing site. S142G itself is not associated with any disease individually, except that it is in linkage disequilibrium (r2 > 0.8) with rs9859260, which correlates with a red cell index. None of the other 21 variants in LD with S142G has a higher functionality score or a trait association. One plausible mechanism may be that S142G generates a splice variant that interacts with HFE differently. Thus, it appears that S142G is the causal SNP interacting with HFE in cancer risk associations. The interaction of HFE with TFRC in modification of risk for cancers has now been replicated in several studies and in multiple cancers. Although HFE and TFRC variants do not show individual associations with cancer risk in GWAS, their interaction should be explored in existing GWAS data in cancer to ultimately establish it as a possible risk factor. The biological basis of this association has implications on public health and policy on unregulated use of iron supplements.

You may contact the first author (during and after the meeting) at