Comments on “Effects of MTNR1B Genetic Variants on Individual Susceptibility to Gestational Diabetes Mellitus: A Meta-Analysis”

Effects of MTNR1B Genetic Variants on Individual Susceptibility to Gestational Diabetes Mellitus: A Meta-Analysis

Jia et al[1] recently published a meta-analysis on the association of MTNR1B variants and gestational diabetes mellitus (GDM). According to their main findings, “rs1387153 (dominant model: p = 0.0002, odds ratio [OR] = 0.78, 95% confidence interval [CI]: 0.68–0.89; recessive model: p < 0.0001, OR = 1.46, 95% CI: 1.24–1.73), rs4753426 (recessive model: p = 0.01, OR = 1.75, 95% CI: 1.14–2.68), and rs10830963 (dominant model: p < 0.0001, OR = 0.72, 95% CI: 0.65–0.78; recessive model: p < 0.0001, OR = 1.56, 95% CI: 1.40–1.74) variants were all significantly associated with the susceptibility to GDM.” However, there are some mistakes regarding the genetic models, meta-analysis, and interpretation which affected their results and should be addressed. Here we aim to fix these issues.

First of all, unfortunately, authors missed the citation of 15 from 17 included articles in their meta-analysis. As it is clear, all included articles should be cited. Therefore, we highly recommend authors to correct references in erratum.

In the meta-analysis by Jia et al,[1] dominant and recessive models should be corrected. Authors used C allele as reference allele in allelic model. It should be noted that recessive model analyzed by Jia et al[1] based on comparing G/T allele versus C is misleading. In dominant model, they compared CC genotypes versus two others. Thus, all the results referred to dominant model should be renamed to recessive model. Therefore, results of recessive model for rs1387153 (CC vs. CT + TT), rs4753426 (CC vs. CT + TT), and rs10830963 (CC vs. CG + GG) should be 0.78 (0.68–0.89), 0.89 (0.61–1.28), and 0.72 (0.65–0.78), respectively. Dominant model should be described as CC + CT versus TT (for rs1387153), CC + CT versus TT (for rs4753426), and CC + CG versus GG (for rs10830963). Results related to corrected dominant model are presented as follow in [Figs. 1] [2] [3].

Fig. 1 Dominant model related to rs1387153 and susceptibility to GDM. CI, confidence interval; GDM, gestational diabetes mellitus; OR, odds ratio.

Fig. 2 Dominant model related to rs4753426 and susceptibility to GDM. CI, confidence interval; GDM, gestational diabetes mellitus OR, odds ratio.

Fig. 3 Dominant model related to rs10830963 and susceptibility to GDM. CI, confidence interval; GDM, gestational diabetes mellitus OR, odds ratio.

In addition, for rs1387153 and rs10830963 polymorphisms, total numbers given in their results (forest plots in Supplementary Material) are incorrect. The authors included total number of studies which were excluded from meta-analysis of dominant, recessive, and additive models (Huopio et al[2] and Huerta-Chagoya et al[3]). Also for the study by Huerta-Chagoya et al,[3] the true numbers should be 408 GDM and 342 controls, while it has been inversely calculated in meta-analysis.

Finally, it seems that C allele of three MTNR1B genetic variants (rs1387153, rs4753426, and rs10830963) have protective effect on individual susceptibility to GDM. Nevertheless, more studies, especially for rs4753426, are needed to confirm these results.

Publication History

Article published online:
28 September 2019

• References

• 1 Jia G, Gao Y, Li C, Zhang Y. Effects of MTNR1B genetic variants on individual susceptibility to gestational diabetes mellitus: a meta-analysis. Am J Perinatol 2019 DOI: 10.1055/s-0039-1685446
• 2 Huopio H, Cederberg H, Vangipurapu J. et al. Association of risk variants for type 2 diabetes and hyperglycemia with gestational diabetes. Eur J Endocrinol 2013; 169 (03) 291-297
• 3 Huerta-Chagoya A, Vázquez-Cárdenas P, Moreno-Macías H. et al. Genetic determinants for gestational diabetes mellitus and related metabolic traits in Mexican women. PloS one 2015; 10 (05) e0126408