Elsevier

Annals of Epidemiology

Volume 26, Issue 11, November 2016, Pages 816-818.e6
Annals of Epidemiology

Rapid report
Quantitative assessment of genetic testing for type 2 diabetes mellitus based on findings of genome-wide association studies

https://doi.org/10.1016/j.annepidem.2016.09.004Get rights and content

Introduction

Some genetic tests provide useful information not only on risk prediction but on diagnosis, prognosis, and the risk of having a child or another family member with a specific disease. Although it is uncertain whether there is a clear advantage in genetic testing for type 2 diabetes mellitus (T2DM), T2DM has a strong genetic basis. The concordance of T2DM is about 70% for monozygotic twins compared to about 20%–30% for dizygotic twins [1]. Recently, genome-wide association studies (GWAS) identified an increasing number of loci associated with susceptibility to T2DM. The GWAS usually cannot identify the specific etiological cause, unlike monogenic studies that predict some forms of T2DM penetrating into family members such as maturity-onset diabetes of the young [2]. However, GWAS can help identify genes in which variations affect the susceptibility to T2DM. Actually, the number of genetic tests and the number of companies offering individual tests have proliferated based on findings of GWAS [3].

However, recent genetic tests that are directly provided to consumers without involving health-care providers such as genetic counselors or physicians have driven governments to make efforts to regulate the business of providing genetic testing. One of the main concerns about such testing is the small relative risk contributed by each genetic variant, which is a main feature of a risk allele (RA) of single-nucleotide polymorphisms (SNPs) identified by GWAS [4]. Even combining several SNPs to discriminate between individuals who will and will not develop diabetes is not useful in clinical practice [5]. Another main concern is misinterpretation by consumers which leads to pursuit of unnecessary interventions or failure to take appropriate preventive measures [3]. The aim of this comprehensive meta-analysis is to provide important information to clarify the lack of clinical utility of genetic testing that combines multiple T2DM-associated SNPs identified by GWAS, considering the recent proliferation of genetic tests for diabetes susceptibility marketed directly to consumers.

Section snippets

Methods

Electronic literature search using EMBASE and MEDLINE (up to 2014) was conducted for studies with a cross-sectional design or longitudinal design and allowed reproduction of a 2 × 2 contingency table (i.e., number of true-positive, false-negative, true-negative, and false-positive cases) that could be reproduced for the cut-off value of the genetic risk score (i.e., total number of RAs carried). Keywords and thesaurus terms related to genetic profiles and T2DM were used (Appendix 1). This

Results

Of 6517 articles retrieved by the electronic literature search, 63 studies including 64,849 cases among 166,596 participants were eligible (Appendices 2 and 3). Details on the SNPs used for T2DM screening are shown in Appendix 4. The number of SNPs used for T2DM screening ranged from 2 to 49. The most commonly used locus was TCF7L2 (27 studies) followed by CDKAL1 (25 studies), CDKN (25 studies), and HHEX (25 studies; Appendix 3).

The Figure 1 shows HSROC curves and their associated confidence

Discussion

According to the method for interpreting the test results using LR+ and LR− [8], alteration of post-test provability compared with pre-test provability was considered minimal if LR+ <2 and LR− <0.5: users of the genetic test should hardly change their attitude toward their chances of developing T2DM. Inconsistency in the pooled sensitivity between cross-sectional and longitudinal studies suggested an unknown selection bias: one speculation is that more participants who were genetically

Acknowledgments

All authors thank Ms. Haga and Ms. Tada in the Niigata University for their excellent secretarial work. This work was supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (JSPS, #26870208). The sponsors had no influence over the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.

First page preview

First page preview
Click to open first page preview

References (8)

  • J. Kaprio et al.

    Concordance for type 1 (insulin-dependent) and type 2 (non-insulin-dependent) diabetes mellitus in a population-based cohort of twins in Finland

    Diabetologia

    (1992)
  • A.L. Gloyn

    Glucokinase (GCK) mutations in hyper- and hypoglycemia: maturity-onset diabetes of the young, permanent neonatal diabetes, and hyperinsulinemia of infancy

    Hum Mutat

    (2003)
  • C.S. Bloss et al.

    Direct-to-consumer personalized genomic testing

    Hum Mol Genet

    (2011)
  • M.I. McCarthy

    Genomics, type 2 diabetes, and obesity

    N Engl J Med

    (2010)
There are more references available in the full text version of this article.

Cited by (1)

The authors declare no conflict of interest.

View full text