Methylenetetrahydrofolate reductase (MTHFR) has a major impact on the regulation of the folic acid pathway due to the conversion of 5,10-methylenetetrahydrofolate (methylene-THF) to 5-methyl-THF. Two common non-synonymous coding region polymorphisms (677C>T and 1298A>C) in the MTHFR gene were shown to confer reduced enzyme activity in in vitro assays leading to a decreased pool of methyl-THF and the MTHFR variant 677C>T was associated with an increased risk of hyperhomocysteinemia, particularly in folate-deficient states [1–4]. With regard to the MTHFR 677C>T polymorphism, results from in vitro assays showed a decrease in enzyme activity to 65% for the heterozygous and to 30% for the homozygous state of the 677T variant [5]. For the 1298A>C polymorphism, enzyme activity in vitro is decreased in homozygous variants and, to a lesser extent, in heterozygotes compared with those homozygous for the wild-type allele [6, 7]. It is supposed that variants (677TT/1298AA or 677CC/1298CC) of MTHFR increase the level of methylene-THF leading to a subsequent reduction of uracil in DNA. The reduction of uracil decreases the amount of misincorporations of uracil instead of thymidine into DNA. This may protect DNA from double-strand breaks and, consequently, from chromosomal alterations [1, 8]. Furthermore,it is assumed that DNA methylation via MTHFR is an important epigenetic feature that modulates DNA methylation status through interaction with folate status [1, 9]. Consequently, MTHFR variants are discussed to influence disease processes and several studies in the literature reported on reduced MTHFR enzyme activity and the susceptibility to different disorders, including vascular disease, neural tube defects and lymphoid malignancies [1, 2, 5, 7, 8, 10–13]. Lymphoid malignancies arise as a consequence of point mutations, chromosomal rearrangements (e.g., chromosomal translocations), and epigenetic alterations in hematopoietic cells making the variant MTHFR an interesting candidate gene for studies on leukemogenesis [14]. However, especially with regard to susceptibility to acute lymphoblastic leukemia (ALL), previous studies led to discrepant results. Although both variants 677T and 1298C have been reported to decrease susceptibility to ALL in children and adults [8, 10, 11, 13], other investigators did not support such associations [15, 16]. Moreover, the protective effect of MTHFR variants was present only in children before folic acid supplement in pregnancy has been recommended [17]. These results suggest a role for gene-environment interactions in the association of MTHFR with ALL.

In the present study, we analyzed the association of MTHFR variants with pediatric ALL in a German study sample including 443 ALL patients and 379 healthy controls.

With regard to ALL, our results are in contrast to previous findings and do not support the assumption that higher methylene-THF levels due to MTHFR variants lead to a decreased risk of ALL in our study population [8, 10, 11, 13]. There are different explanations for these diverging results. The present study included pediatric patients from Germany. The above mentioned studies included adult patients from United Kingdom [8] and Italy [13] but also pediatric patients from United Kingdom [10] or Brazil [11]. These latter two studies on the MTHFR 677C>T and 1298A>C polymorphisms in childhood ALL included patients diagnosed between 1992–1998 and 1991–2000, respectively, while our study included pediatric ALL patients diagnosed after July 1999. Thus, probably a large proportion of patients analyzed in the studies by Wiemels et al. [10] and Franco et al. [11] were born before folic acid supplementation in pregnancy was recommended. In a Canadian study, Krajinovic et al. reported that the protective effect of MTHFR variants was accentuated and present only in children born before 1996 [17]. This may explain the diverging results with regard to pediatric ALL and is further supported by a recent study of Balta and colleagues from Turkey, where in 142 pediatric ALL patients diagnosed between February 2000 and February 2002, no significant association between the MTHFR 677C>T polymorphism and ALL was detected [16]. However, the impact of folate metabolism on the risk of ALL may also vary from population to population because of additional gene-environment and gene-gene interactions. For example, in our study population we cannot control for differences in folate status as we have not collected information on dietary intake. Thus, a potentially existing association of MTHFR variants and ALL may be masked by the inability to control for the individual folate status in our study population. Furthermore, we cannot exclude potential gene-gene interactions. As several polymorphic key enzymes are involved in folate metabolism, a more comprehensive approach would clearly yield more precise estimates of the association of folate metabolism with risk of ALL. As an additional explanation for diverging results, specifically in subgroups with chromosomal translocations, the potential importance of parental variants and folate status with respect to an in utero pathogenesis of ALL (e.g., MLL/AF4 and TEL/AML1 fusion gene-positive ALL) should be considered [22]. Lastly, our results may simply be due to chance and small patient numbers. However, with regard to sample size, our study exceeds most other studies reported in the literature, so far. In conclusion, our findings suggest that the MTFHR gene variants analyzed here do not have a major influence on the susceptibility to pediatric ALL in the German population. However, potential effects in rare specific ALL subgroups cannot be excluded.

We retrospectively evaluated the association of the MTHFR 677C>T and 1298A>C polymorphisms with childhood ALL by analyzing a study sample of 443 ALL patients consecutively enrolled onto the German multicenter trial ALL-BFM 2000 and 379 healthy controls. No significant associations between specific MTHFR variants or combinations of MTHFR variants and risk of ALL were observed neither in the total patient group nor in analyses stratified by gender, age at diagnosis, DNA index, immunophenotype, or TEL/AML1 rearrangement. Our findings suggest that the MTHFR 677C>T and 1298A>C gene variants do not have a major influence on the susceptibility to pediatric ALL in the German population.