Evaluation of the TCF7L2 gene as a predictor of type 2 diabetes


A. Cecile J.W. Janssens, PhD
Center for Medical Decision Making
Department of Public Health
Erasmus University Medical Center Rotterdam, the Netherlands

Educational objectives

After reading this case study, you should be able to:

  • interpret gene-disease associations in terms of clinical validity of a genetic test
  • calculate basic indicators of clinical validity and utility
  • discuss the potential clinical or public health implications of predictive testing based on a single disease susceptibility gene.


Type 2 diabetes mellitus represents a significant health problem in the United States. According to the most recent assessment by the Centers for Disease Control and Prevention (CDC), the prevalence of type 2 diabetes is nearly 6% nationwide; prevalence exceeds 10% in some ethnic subgroups, including African-Americans, Hispanics and Native Americans. Although the risk of type 2 diabetes increases with age, incidence has been rising in all age groups and the increase in children is of particular concern. In 2000, diabetes was the sixth leading cause of death.

In type 2 diabetes, the body’s impaired production and use of insulin results in chronically elevated blood glucose levels and long-term tissue damage. Evidence for genetic susceptibility to type 2 diabetes comes from studies of population admixture, familial aggregation, mono- and dizygotic twins, and genetic association and linkage. Diabetes is also a major feature of certain monogenic disorders including maturity-onset diabetes of the young (MODY). Genome-wide linkage studies have identified susceptibility loci for type 2 diabetes on many chromosomes and associations with numerous candidate genes (including mitochondrial genes) have been reported. Most of these associations are small (OR 1-2) and inconsistent among studies.

Case Study

In February 2006, Grant et al. reported that a variant of the TCF7L2 gene is associated with type 2 diabetes (T2D).

Grant SF, et al. Variant of transcription factor 7-like 2 (TCF7L2) gene confers risk of type 2 diabetes. Nat Genet 2006;38:320-323.

This discovery gained enormous media exposure all over the world. The principal investigator, Dr Kari Stefansson, told the New York Times that a practical consequence of the discovery could be a diagnostic test to identify people who carry the variant gene. People who knew of their extra risk, he said, would be motivated to prevent the lifestyle habits that lead to diabetes. Noting that a Scottish scientist had led the research team, the Glasgow Herald reported, “Discovery of holy grail will help scientists treat diabetes.”

Grant et al. reported an association of the TCF7L2 gene with T2D. They examined this association in three unrelated case-control series.

  1. What is the interpretation of the relative risk column in Table 1?
    1. The relative risk is an odds ratio.
    2. The relative risk is probably overestimated.
    3. Both a and b are correct
    4. Both a and b are incorrect

Table 1 compares the allele frequencies in affected (case) and control groups in the three study populations.

  1. Which allele frequencies differed significantly different between case and control groups in all three populations?
    1. Allele 0
    2. Allele 4
    3. Allele 0 and 12
    4. All alleles ≥ 4
  2. What do the genotypes OX and XX mean? What do you think of this classification?
    1. OX means one allele is allele 0, the other is not allele 0
    2. XX means that both are alleles are not risk alleles
    3. Both a and b are correct
    4. Both a and b are incorrect

The overall population attributable risk (PAR) of the X allele (OX and XX genotypes) was 21%.

  1. What is the interpretation of PAR?
    1. 21% of the population carries a risk allele
    2. 21% of the patients carry a risk allele
    3. Both a and b are correct
    4. Both a and b are incorrect


The lifetime risk of T2D is estimated at 33% (Narayan et al. 2003, JAMA). Based on this estimate and the results of Grant et al., we constructed the following table to reflect the association of TCF7L2 genotype with T2D in a hypothetical population of 100,000 persons.

graph of T2D and no T2D
  1. What is the population disease risk of T2D? What are the relative risks of the OX and XX genotypes?
    1. The population disease risk is 33%.
    2. The relative risks of OX and XX are 1.45 and 2.41
    3. Both a and b are correct
    4. Both a and b are incorrect
  2. What is the T2D risk associated with each genotype (penetrance)?
    1. The T2D risk in OO carriers is 33.3%
    2. The T2D risk in XX carriers is 63.3%
    3. Both a and b are correct
    4. Both a and b are incorrec


Suppose you took this test to learn whether you should do something to prevent diabetes.

  1. Would any of these test results motivate you to go on a diet or increase your physical activity?
    1. These results would not motivate me at all.
    2. Learning that I am an OX or XX carrier would increase my motivation.
    3. Learning that I am an XX carrier would increase my motivation.
    4. All of the results would motivate me.


Not all carriers will develop T2D.

  1. Calculate the percentages of OX and XX carriers who will not develop T2D. Which is the correct answer?
    1. 62% of the OX carriers will not develop T2D
    2. 37% of the XX carriers will not develop T2D
    3. Both a and b are correct
    4. Both a and b are incorrect


Not all carriers will develop T2D due to their carrier status.

  1. What proportion of T2D in OX and XX carriers will develop due to other causes?
    1. 0%
    2. 26%
    3. 33%
    4. 50%


Suppose this discovery led to a preventive intervention that specifically targets the effects of the TCF7L2 gene and is not useful for other causes of diabetes.

  1. What is the percentage of the total population that should adopt this intervention to prevent the 21% of cases indicated by PAR?
    1. 26%
    2. 33%
    3. 45%
    4. 100% (everybody)


We noted that not all carriers will develop T2D and that those who do may develop T2D from other causes.

  1. What is your opinion about the benefit of TCF7L2 testing for the prevention of T2D?
    1. It is an inefficient prevention strategy.
    2. It is too premature to evaluate this when there is no adequate preventive intervention.
    3. It is not the most important implication of this research study.
    4. All of the above
  2. Knowing that the preventive intervention specifically targets the effects of the TCF7L2 gene, what percentage of the OX and XX group will not benefit from the intervention?
    1. 88% of the OX carriers
    2. 63% of the XX carriers
    3. Both a and b are correct
    4. Both a and b are incorrect


Correct Answers:

Question 1:

Answer c is correct:
The relative risk is actually an odds ratio. It can be calculated directly from the frequencies in the affected and control populations. When a disease is rare, the OR is a good proxy of the RR, but T2D is not rare and hence the RR is likely overestimated. Note that the OR can also be interpreted as a RR when the control frequency refers to non-transmitted parental alleles and the affected frequency to transmitted parental alleles. This was the case in one of their previous papers, but not in this one.

Question 2:

Answer c is correct:
Only the frequencies of allele 0 and 12 were significantly different in cases and controls.

Question 3:

Answer a is correct:
The researchers combined the non-zero alleles. Alleles 4, 8, 12 and 16 were combined to define the risk allele, “X”. The collapsing was not based on the results of Table 1, but because the alleles share a “relatively recent common ancestor.”

Question 4:

Answer d is correct:
PAR indicates the number of cases that can be prevented when the effects of the risk factor are effectively eliminated. Thus, if X alleles are in fact a cause of diabetes and all carriers of X alleles adopted a 100% effective intervention, then 21% of diabetes cases could be prevented.

Question 5:

Answer c is correct:
The population disease risk is 33,333/1,000,000 = 0.33 or 33%.
The relative risk of OX is the ratio between the disease risk in OX carriers and the risk in non-carriers (OO). It is calculated as:
OX: (14,465/38,000) / (14,439/55,000) = 1.45
XX: (4,429/7000) / (14,439/55,000) = 2.41

Question 6:

Answer b is correct:
This is estimated as the number of carriers who develop T2D divided by the number of carriers.
OO: 14,439/55,000 = 0.263 or 26.3%
OX: 14,465/38,000 = 0.381 or 38.1%
XX: 4,429/7,000 = 0.633 or 63.3%

Question 7:

Any answer can be correct: (This is not counted in the results.)
In OX carriers the T2D risk increases from 33% to 38% and in XX carriers the increase is from 33% tot 63%. The risk increase for OX carriers is not very impressive, so learning that you are an OX carrier may not be sufficient motivation to adopt a healthier lifestyle. XX carriers have a 63% risk, which may or may not offer incentive.

Question 8:

Answer c is correct:
This is calculated as 1- the number of carriers who develop T2D divided by the number of carriers.
OO: 1- 14,439/55,000 = 0.737 or 74%
OX: 1- 14,465/38,000 = 0.619 or 62%
XX: 1- 4,429/7,000 = 0.367 or 37%

Question 9:

Answer b is correct:
Only part of the T2D risk of OX and XX carriers can be attributed to the TCF7L2 gene. OO carriers still have a T2D risk of 26%. Since only the excess risk can be attributed to TCF7L2 genotype, 26% of the OX and XX carriers develop T2D due to other causes.

Question 10:

Answer c is correct:
To prevent 21% of the cases, all carriers of the X allele should adopt preventive intervention: OX carriers (38%) plus XX carriers (7%)—or 45% of the total population.

Question 11:

Answer d is correct:
Undeniably, the discovery of an association between TCF7L2 and type 2 diabetes is noteworthy. Type 2 diabetes is a leading cause of morbidity and mortality in the developed world and is increasing in prevalence worldwide. The association is robust; it was replicated in three large, independent study populations and it offers potential new insight into the pathobiology of diabetes. However, genetic testing for TCF7L2 variants is not currently useful for the prevention of T2D.

Question 12:

Answer c is correct:
Among all OX carriers, 62% will not develop T2D and 26% will develop T2D from other causes. Hence, 88% of the OX carriers and 63% (37% + 26%) of the XX carriers will not benefit from the intervention.

Page last reviewed: June 15, 2009