A series of recent studies have pointed out the association between the known functional polymorphisms of a main folate-metabolizing enzyme, methylenetetrahydrofolate reductase (MTHFR), and the risk of malignant lymphoma in Japanese adults1 or the risk of acute lymphoblastic leukemia in British adults2 and in childhood,3 suggesting that folate metabolism plays an important role in their genesis. But these studies are case-control ones performed in patients recently diagnosed with these lymphoproliferative diseases, and we have, to date, no available data on the evolution of homogeneous cohorts of patients who were initially tested for MTHFRpolymorphisms for nonmalignancy related reasons.
Our group has focused on the links between the metabolism of folates and the risk for venous thromboembolism. All patients investigated for idiopathic thromboembolism in our outpatient department of hematology were therefore tested, beside the classical hemostasis-related biological risk factors, for the MTHFR 677C>T polymorphism since it has been described. We finally found that low levels of red blood cell methylfolate are strongly associated with the risk of venous thromboembolism, particularly in subjects who do not carry the mutated TT genotype.4
We report here data concerning 852 consecutive adults aged between 55 and 76 years old and diagnosed between September 1995 and September 2000. All these patients and their general practitioners were thereafter questioned, between September and December 2001, for the subsequent occurrence of any malignant lymphoproliferative disease: the answer was based on the nature of known antecedents and on the results of a recent (less than one month prior) clinical exam, blood cell count, and serum immunofixation electrophoresis, with secondary specific investigations in case of any abnormalities. Forty-two cases were finally recorded: 29 chronic lymphocytic leukemias, 10 non-Hodgkin lymphomas, 1 Hodgkin disease, and 2 multiple myelomas. The median values [lower-upper quartiles] (range) of lengths of follow-up, given in months, were respectively 39 [23-52] (12-76) in patients negative for secondary lymphoproliferative diseases and 42 [27-55] (14-76) in positive patients (P = .79).
The frequencies of the 3 MTHFR genotypes are given in Table1. The distribution of the various genotypes is apparently different in groups of patients with or without the subsequent development of any chronic lymphoproliferative disease, with a relative excess in the homozygous mutated genotype among asymptomatic patients (presence versus absence of the MTHFR677T homozygous genotype: P = .02 using the Fisher exact test). Using logistic regression, the wild-type genotype being taken as a reference, the calculated odds ratio for developing a chronic lymphoproliferative disease is 0.91 (95% confidence interval: 0.48-1.76, P = .79) in the case of an heterozygous genotype, but, in the case of a 677T homozygous genotype, is 0.26 (0.077-0.88, P = .031).
Thus in a large group of patients with a minimum age of 55 years and investigated for idiopathic thomboembolic disease, the MTHFR677T homozygous genotype seems to protect against the subsequent development, within 1-6 years after testing, of a chronic lymphoproliferative disease. No mutated gene-dose effect could be observed in this study, a fact concordant with what has been suggested by Matsuo et al for lymphomas.1 Patients with idiopathic venous thromboembolism are at increased risk for malignant diseases, with odds ratios for a new cancer diagnosis within 6-12 months after the episode in the range of 4- to 7-fold increased risk5: this is a sensitizing factor that probably concurred with the positive result of this study. As DNA analysis had been performed after its extraction from blood before any evidence of circulating abnormal leukocytes, the MTHFR genotypes that we obtained are likely to be constitutional ones and are probably not modified by any loss of heterozygosity in circulating abnormal lymphoid cells. But only a minority of the patients developed a lymphoproliferative disease, and quite limited differences in the distribution of their MTHFRgenotypes would have nullified the statistical conclusion of our study: a confirmation in larger series of thrombotic patients with secondary lymphoid neoplasia is required.
The MTHFR 677T homozygous genotype leads to reduced MTHFR activity and results in the upstream accumulation of 5,10-methylene tetrahydrofolate, which accelerates the methylation of uridylate to thymidylate and, in turn, reduces the chance for the misincorporation of uracil into DNA, a cause of genetic instability due to double-strand breaks during uracil excision repair by uracil DNA glycosylase. If this interpretation is correct, in patients with idiopathic venous thromboembolism not carrying the MTHFR 677T homozygous genotype, increasing 5,10-methylene tetrahydrofolate levels by folic acid supplementation, which may protect from recurrence,4may also reduce the risk for chronic lymphoproliferative diseases: these speculations warrant further investigations.
