Folate and genetic variation in folate metabolism in the aetiology of colorectal cancer
The aim of this thesis was to investigate the roles of folate and polymorphisms in folate metabolising genes in the aetiology of colorectal cancer. The objectives were to: (1) conduct systematic reviews of (a) folate and colorectal neoplasia and (b) polymorphic genes in the folate pathway and colorectal neoplasia; (2) undertake a case-control study of folate, polymorphisms in the methylenetetrahydrofolate reductase (MTHFR) gene and colorectal cancer in the north-east of Scotland; and (2) re-analyse the study data as if it were from a case-only study, comparing the results with those of the case-control analysis. The first systematic review concluded that there was considerable evidence that folate status modulates the risk of developing colon cancer, adenomatous polyps and, possibly, rectal cancer. The risk reduction in the highest compared to the lowest intake group, for colon cancer and adenomas, was in the region of 30-40%. The second systematic review considered polymorphisms in MTHFR, methionine synthase (MTR), methionine synthase reductase (MTRR), cystathionine beta-synthase (CBS) and thymidylate synthase (TS). Homozygosity for the MTHFR 677TT and 1298CC variants was associated with moderately reduced colorectal cancer risk in the majority of studies. There was evidence of gene-environment interaction; in four of five studies, the lowest cancer risk was in 677TT subjects who had higher folate levels (or a “high methyl diet”) while in three studies 677TT homozygotes with the highest alcohol intake had the highest cancer risk. The available evidence on MTHFR C677T and adenomatous polyps was inconclusive. A population-based case-control study of colorectal cancer was undertaken in Grampian health board area. Eligible cases were Grampian residents diagnosed with histologically confirmed primary colorectal cancer during September 1998-February 2000. The primary analysis focused on all colorectal cancer. Disease risk followed an inverted “U” pattern with increasing (total and dietary) folate intake. There was little evidence of relations between disease and intakes of total or dietary vitamin B12, vitamin B6 or riboflavin or of interactions between these factors and folate. There was a suggestion that protein intake was inversely related to colorectal cancer, which also drove the observation of a modestly increased risk with low, as compared to high, methyl intake (a combination of folate, protein and alcohol intake). A modest, non-significant, reduction in risk was found in those with the 677 TT, compared to the CC, genotype. For A1298C, a slight, non-significant, reduction in risk for homozygotes for the variant allele was found. Significant interactions between the two polymorphisms and total folate intake were observed, although the patterns of interaction were different for each variant and not entirely consistent with published studies.