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

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

2. Gene-disease association
3. Gene-environment interaction
4. Gene-gene interaction

Hypothesis:  This study tests the hypothesis that the presence of certain polymorphisms in both the glutathione S-transferase (GST) gene and the catechol-O-methyl-transferase (COMT) gene synergistically increase susceptibility to breast cancer and that hormone replacement therapy, in the presence of these mutations, interacts to augment the risk.

Gene(s)
Identification of the following:

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

1. Gene: Glutathione S-Transferase, Theta-1 (GSTT1)
2. Chromosome location:  22q11.2
3. Gene product/function: In humans, the GSTT1 enzyme is one of the four major polymorphic forms of the GSTs. GSTT1 is involved in the biotransformation of low-molecular-weight halogenated compounds and reactive epoxides; thus, the enzyme may act as a detoxification sink.
4. Alleles:  Null (A proportion of the population, varying from 12%-20% in Europeans to 65% in Asians carries a null polymorphism (deletion). Such people do not express the gene and therefore do not have any GSST1 enzyme activity.)
5. OMIM #: 600436

1. Gene: Gluathione S-transferase, Mu-1; (GSTM1)
2. Chromosome location:  1p13.3
3. Gene product/function: The GSTM1 enzyme is the mu form, one of the four major polymorphic forms of the GSTs existing in humans. The GSTs are a family of enzymes responsible for the metabolism of a broad range of xenobiotics and carcinogens. This enzyme catalyzes the reaction of glutathione with a wide variety of organic compounds to form thioethers, a reaction that is sometimes a first step in a detoxification process. 
4. Alleles:  Null (Deletions in GSTM1 occur at a frequency of about 15% in human populations. Individuals who are deletion homozygous, i.e., GSTM1 null, exhibit and absence of enzyme activity. A null allele at the GSTM1 locus is found in 40%-45% of Caucasians. GSTM1 deficiency may be a risk factor for cancer by providing increased sensitivity to chemical carcinogens. The mechanism of carcinogenicity may be related to an increased formation of DNA adducts in the presence of the null deletion.)
5. OMIM #:138350

1. Gene: Gluathione S-transferase, Mu-3; (GSTM3)
2. Chromosome location:  1p13.3
3. Gene product/function: The GSTM3 enzyme is a variant of the mu form of GST. This variant has a different isoelectric point than GSTM1 but has overlapping substrate specificity. The enzyme may be involved in regulation. 
4. Alleles:  *B (A deletion of three bp in intron 6 of this this gene results in the generation of a recognition sequence for the YY1 transcription factor. This deletion may have both negative and positive regulatory effects.) 
5. OMIM #: 138390

1. Gene:Gluathione S-transferase, Pi-1 (GSTP1)
2. Chromosome location:  11q13
3. Gene product/function: The GSTP1 enzyme is the pi form, one of the four major polymorphic forms of the GSTs existing in humans. The GSTs are a family of enzymes responsible for the metabolism of a broad range of xenobiotics and carcinogens.  
4. Alleles:  *B, *C (For GSTP1, two variant alleles may be associated with disease risk:GSTP1*B, VAL105, in which isoleucine is changed to valine at position 105, and GSTP1*C, VAL 105 and VAL 114, in which isoleucine is changed to valine at position 105 and alanine is changed to valine at position 114. These polymorphic forms exhibit altered specific activity and affinity for substrates, and reduced enzyme activity.)
5. OMIM #:134660

1. Gene: Catechol-O-methyltransferase; (COMT)
2. Chromosome location:  22q11.2
3. Gene product/function: COMT catalyzes the transfer of a methyl group from S-adenosylmethionine to catecholamines and catalyzes the methylation of catechol estrogens (a carcinogenic metabolic biproduct of estrogen hydroxylation) to less polar monomethyl esthers. If not completely conjugated, catachol estrogens can be further oxidized to reactive quinone/semiquinone intermediates, which can form free radicals or DNA adducts. Thus COMT may decrease susceptibility to estrogen-linked carcinogenisis by catalyzing the formation of noncarcinogenic conjugates. 
4. Alleles:  LL, HL, HH (COMT enzyme activity is genetically polymorphic in human red blood cells (RBCs) and liver, with a trimodal distribution of low, intermediate, and high levels of activity. This genetic polymorphism results in a 3- to 4-fold difference in COMT activity. About 25% of Caucasians are homozygous for the low activity allelle, COMT(LL). The homozygous polymorphism could be related to breast cancer caused by reduced capacity to inactivate catechol estrogens.
5. OMIM #: 116790

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

Hormone replacement therapy (HRT)

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

Breast cancer

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

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

1. Disease case definition: Breast cancer. Study included all breast cancer cases referred to the Kuopio University Hospital during 1990–1995. Breast cancers were characterized by stage, lymph node status, estrogen receptor status and grade of cancer
2. Exclusion criteria: No cases were excluded by design. 51 cases were missed that occurred in private patients who did not enter the hospital by standard procedures. Another 11 cases were missed during a nurses’ strike that occurred in 1995.  
3. Gender: Women
4. Race/ethnicity: Caucasians only (Finnish)
5. Age: Case-patients were 44.3 – 91.6 years
6. Time period: 1990 - 1995
7. Geographic location: Finland
8. Number of participants: 483

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

1. Control definition: Healthy Finnish woman in same catchment area. 
2. Matching variables: No special attention was paid to matching a specific case-patient with a control, but all women living in the same catchment area without any history of breast cancer were possible controls. They were selected randomly. 
3. Exclusion criteria: Any previous history of breast problems; non-Finnish origin; no DNA available
4. Race/ethnicity: Caucasion only (Finnish)
5. Age: 37.5–77.2 years
6. Time period: Same, 1990–1995
7. Geographic location:Finland, same catchment area
8. Number of participants: 482

Cohort definition
For study designs 2, 3, and 6, define the following if available:

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

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

1. Environmental factor: HRT use
2. Exposure assessment: Done by interview using trained nurses for both case-patients and controls
3. Exposure definition: Ever use of HRT was defined as use of HRT for at least one month. Users were further subcategorized into those who used HRT for 1 – 30 months and those who used HRT for > 30 months. Use of HRT for > 30 months was considered long-term or longer than average use
4. Number of participants with exposure data: 100%

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

1. Genes: COMT, GSTM1, GSTM3, GSTP1, GSTT1
2. DNA source:  Blood
3. Methodology: Genomic DNA was extracted from blood lymphocytes, and genotyping was done by polymerase chain reaction (PCR) and restriction analysis.

   Authors referred to previous papers describing the following analysis techniques:   GSTM1and GSTT1-specific primer pairs were used in a multiplex PCR analysis with B-globin primers as an internal control.  GSTM3 genotyping was done by PCR followed by restriction digestion with Mn/I. GSTP1 genotyping was done by PCR followed by restriction with SnaBI.   COMT genotyping was done by PCR followed by restriction with NlaIII.

   
   
4. Number of participants genotyped: A total of 483/516 (93.6%) patients were genotyped. Genotyping was done on 492/514 (95.7%) controls. COMT data were not obtained for two case-patients and two controls, GSTM1 for four controls and two case-patients, GSTM3 for two controls and two case-patients, GSTP1 for one control, and GSTT1 for four controls and two case-patients. Assuming all of these omissions were random, complete genotypes were done on 495 cases and 511 controls.

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

Results:

Gene-gene interaction/disease association:
Combined effect of COMT and GST genotypes in breast cancer risk

Gene-gene, environment interaction/disease association: 
Combined effect of COMT and GST genotype in breast cancer risk according to duration of HRT use

Risks were significantly increased for women with HRT who carried the low-activity COMT-L allele or COMT-LL genotype together with either the GSTP1 Ile/Ile or the GSTT1 null genotypes. Although on the basis of small numbers, a clear tendency of increased risk by increasing duration of HRT use and increased number of at-risk alleles was observed. This effect was dose-dependent with respect to length of HRT use. No overall modifying effect was found in breast cancer risk for combinations of COMT and GST genes alone. Nor was any association detected for HRT use alone and breast cancer risk. Thus, increased overall breast cancer risk for users of HRT may be at least partly explained by differences in genetic susceptibility!

Second, the study proposes that some of the previously discrepant results regarding the association of GST and COMT genes with breast cancer may be due to inappropriate or incomplete study designs, or studies starting with too few cases. The authors state that “studies should be well-defined and with high sample numbers, and done in ethnically well-defined populations.”

The study was designed to elucidate whether polymorphisms of COMT and GST genes synergistically modify the risk for breast cancer. Although breast cancer risk was not associated with combinations of polymorphisms alone, risk was significantly increased with combinations of the low-activity variants of COMT and some GSTT and GSTP polymorphisms when combined with long-term HRT use. The fact that the various alleles were found to be in Hardy-Weinberg equilibrium in the study population tends to verify the study design. Thus, the results are generalizable within the population in which the study was done.

Very little evidence was found of selection bias in the study because most cases, 84% (483/516) and controls, 72% (482/514) ultimately met all inclusion criteria and participated in the study. However, because 51/62 cases that were lost to the study were private patients, one might consider the possibility of some genetic differences in these patients based upon class structure. The Hardy-Weinburg equilibrium of alleles observed among included cases argues against that.

Because controls tended to be younger than case-patients, there could have been some negative volunteer bias. Older subjects may have tended to refuse to be controls because the imposition of participation. However, older case-patients may have been innately more likely to participate, becuase they needed treatment. Furthermore, the age difference in the case and control groups may have inflated results because of the possibility that the younger control group may have been less likely to be postmenopausal, less likely to have been on HRT for long periods, and less likely to have developed cancer.  

Although the cooperation rate among controls was lower than that among case-patients, more controls than case-patients reported HRT use. This could have resulted in some selection bias because normally women with healthier lifestyles are more likely to participate in studies. However, the respective participation rates were not apparently related to genotype.

The low case numbers in some at-risk categories is a fallacy of the study. This resulted in wide confidence intervals, and some of the conclusions were based upon very low sample numbers. However, the study population was large enough to detect at least some cases and controls in each category. 

The study points out the synergistic interplay between genetic and environmental factors in increasing the risk for breast cancer among the study population. Nevertheless, the results need to be confirmed in other groups and populations, and with larger numbers of cases before public health recommendations can be made. Further studies also would be necessary to determine whether the results are generalizable to ethnically different or ethnically mixed populations.