Reference
Complete the bibliographic reference for the article according to AJE format.

Ulvik A, Evensen ET, Lien EA et al. Smoking, folate, and methylenetetrahydrofolate reductase status as interactive determinants of adenomatous and hyperplastic polyps of colorectum. Am J Med Genet 2001;101:246-54.

Category of HuGE information
Specify the types of information (from the list below) available in the article:

1. Prevalence of gene variant
2. Gene-disease association
3. Gene-environment interaction
4. Gene-gene interaction
5. Genetic test evaluation/monitoring

1. Prevalence of gene variant
2. Gene-disease association
3. Gene-environment association

Study hypotheses or purpose
The authors study hypotheses or main purpose for conducting the study.

Hypothesis: Epidemiologic studies suggest a positive association between smoking and risk of colorectal polyps.  Evidence also suggests that folate may be protective against colorectal polyps and that the C677T polymorphism of the MTHFR gene may influence risk, although evidence regarding the latter is inconclusive.  The authors hypothesize that folate status and genotype may interact with smoking status to influence risk of colorectal polyps.  This study examines the association of smoking, folate status, MTHFR genotype, and interactions among these factors, with risk of colorectal polyps.

Gene(s)
Identification of the following:

1. Gene name
2. Chromosome location
3. Gene product/function
4. Alleles
5. OMIM #

1. Gene:MTHFR
2. Chromosome location: 1p36.3
3. Gene product/function: Folate-metabolizing enzyme that catalyzes irreversible conversion of 5,10-methylenetetrahyrdofolate to 5-methyltetrahydrofolate; involved in shunting folates to either methionine or DNA synthesis
4. Alleles: 677C/T
5. OMIM #: 236250

Environmental factor(s)
Identification of the major environmental factors studied (infectious, chemical, physical, nutritional, and behavioral)

Health outcome(s)
Identification of the major health outcome(s) studied

1. High-risk adenomatous polyps of the colon and rectum (high risk defined as ³ 10 mm or severe dysplasia or villous pathology)
2. Hyperplastic polyps of the colon and rectum

1. Case-control
2. Cohort 
3. Cross-sectional
4. Descriptive or case series
5. Clinical trial
6. Population screening

Case definition
For study designs 2, 3, and 6, the following are defined, where available:

1. Case selection criteria
2. Exclusion criteria
3. Gender
4. Race/ethnicity
5. Age
6. Time period
7. Geographic location
8. Number of participants

Control definition  
For study design 1, the following are defined, if available.

1. Control selection criteria
2. Matching variables
3. Exclusion criteria 
4. Gender
5. Race/ethnicity
6. Age
7. Time period
8. Geographic location
9. Number of participants

Cohort definition  
For study designs 2, 3, and 6, the following are defined, if available.

1. Cohort selection criteria 
2. Exclusion criteria
3. Gender
4. Race/ethnicity
5. Age
6. Time period
7. Ggeographic location
8. Number of participants

1. Cross-sectional selection criteria : Participants in the Telemark Polyp Study-I (TPS-I) who accepted an offer of colonoscopy and polypectomy; TPS-I was a prevention trial initiated in1983 in which half the participants were randomized to endoscopic screening
2. Exclusion criteria: Not specified
3. Gender: 229 male; 214 female
4. Race/ethnicity: Not specified
5. Age: 63-72 years
6. Time period: Questionnaire, colonoscopy, and laboratory analysis in 1996
7. Geographic location: Norway
8. Number of participants: 442,139

1. Environmental factor
2. Exposure assessment
3. Exposure definition
4. Number of participants with exposure data (%
   of total eligible)

1. Environmental factor 1: Smoking status
2. Exposure assessment: Self-report on written survey
3. Exposure definition: Current smoker; # cigarettes/day; former smoker
4. Participants with exposure data: 442 (100%)
   

1. Environmental factor 2 : Folate status
2. Exposure assessment: 1 st: red blood cell folate (nmol/L); 2 nd: serum folate (nmol/L); 3 rd: plasma homocysteine (micromol/L); all collected from whole blood
3. Exposure definition: RBC folate above or below median of 263 nmol/L
4. Participants with exposure data: 442 (100%)

1. Gene
2. DNA source
3. Methodology
4. Number of participants genotyped (% of total eligible) 

1. Gene: MTHFR
2. DNA source: Fresh whole blood
3. Methodology: 1 microliter of whole blood was overlaid with 50 microliters of PCR mastermix and subjected to 33 thermocycles. Allele-specific PCR products were analyzed by multiple injection capillary electrophoresis.
4. Participants genotyped: 441 (99.8%)

1. Prevalence of gene variant
2. Gene-disease association
3. Gene-environment interaction
4. Gene-gene interaction
5. Genetic test evaluation/monitoring

1. Prevalence of gene variant

Prevalence of MTHFR polymorphisms in cross-sectional study participants. (Note that study sample was not randomly selected.)

2. Gene-disease association - Cross-sectional**

*adjusted for age, gender, current smoking, RBC folate, use of NSAIDS (yes/no), flexible sigmoidoscopy in 1983, and BMI (tertiles)
** p for trend = 0.08

• Univariate odds ratios in italics were calculated based on numbers presented in the paper.

3. Gene-environment interaction (1): Risk of high-risk adenoma stratified by folate status. Numbers of subjects (i.e., cell counts) were not presented.

*p for trend = 0.002
**p for trend = 0.22

• Adjusted for age, gender, current smoking, RBC folate, use of NSAIDS (yes/no), flexible sigmoidoscopy in 1983, and BMI (tertiles)
• Numbers in each cell were not presented

Gene-environment interaction (2): Risk of high-risk adenoma stratified by smoking status; genotypes CT and TT were similar and therefore collapsed into one group (“CT or TT”)

2x4 table:

Gene-environment interaction (3): Risk of high-risk adenoma stratified by smoking and folate status. Genotypes CT and TT were similar and therefore collapsed into one group (“CT or TT”). Numbers of subjects (i.e., cell counts) were not presented.

*p for trend = 0.005
**p for trend = 0.002
***p for trend = 0.002

• p for interaction of genotype/smoking variable and folate = 0.006
• adjusted for age, gender, current smoking, RBC folate, use of NSAIDS (yes/no), flexible sigmoidoscopy in 1983, and BMI (tertiles)
• A multiplicative interaction between smoking and genotype (CT or TT) would be expected to yield an OR for the “Smoker/CT or TT” combination of 
• OR smoking * OR CT/TT = 1.77*3.64 = 6.44.
•  An additive interaction between smoking and genotype (CT or TT) would be expected to yield an OR for the “Smoker/CT or TT” combination of
• OR smoking + OR CT/TT - 1 = 1.77 + 3.64 – 1 = 4.41

Based on the results of the cross-sectional analysis, the authors conclude that folate status and C677T MTHFR polymorphism modulate the carcinogenic effect of smoking on colorectal mucosa.  The authors emphasize the significant interaction between smoking, folate status, and MTHFR genotype in association with high-risk colorectal adenomas.  The less common T allele increased the risk in smokers with low folate but was protective in nonsmokers with high folate.  Among smokers, high-risk adenoma was associated with low folate and the CT or TT genotype and with high folate and the CC genotype.

Comments
Provide additional insight, including methodologic issues and/or concerns about the study

The main finding of the strong three-way interaction is interesting, yet the discussion of the unexpected results it yields and their consistency with biological mechanisms is not fully explored.  For instance, the credibility of the findings would be enhanced if an explanation of the reversal of the folate association in nonsmokers were addressed.

In general, it would be useful to present all two-way interactions before presenting the three-way interaction.  In this case, it would be useful to see the two-way interaction between folate and genotype stratified by smoking status.

It is not apparent that stratification of the analysis by genotypes adds practical information for improving colorectal cancer interventions.