Update on “Associations of Estrogen Receptor Alpha Gene Polymorphisms with Type 2 Diabetes Mellitus and Metabolic Syndrome: A Systematic Review and Meta-Analysis”
Zulvikar Syambani Ulhaq
Abstract
Letter to: Associations of Estrogen Receptor Alpha Gene Polymorphisms with Type 2 Diabetes Mellitus and Metabolic Syndrome: A Systematic Review and Meta-Analysis Horm Metab Res 2018; 50(06): 469-477 DOI: 10.1055/a-0620-8553 Dear Editor, Associations of Estrogen Receptor Alpha Gene Polymorphisms with Type 2 Diabetes Mellitus and Metabolic Syndrome: A Systematic Review and Meta-Analysis Yang et al. [ 1 ] published a meta-analysis on the association of ESR1 (estrogen receptor alpha) polymorphism and the risk of type 2 diabetes mellitus (T2DM). Based on to their findings, recessive model ESR1 rs2234693 polymorphism (OR=0.673, 95 % CI=0.550–0.823, p random-effect model<0.001, p fixed-effect model<0.003) was significantly associated with the susceptibility to T2DM. However, there are some mistype on the genetic models, it should be written recessive model rather than the regressive model in the manuscript. Moreover, in this letter, I also update the results of the meta-analysis for the association between ESR1 polymorphisms with T2DM. In this present study, two studies published by Akhmad et al. [ 2 ] and Ereqat et al. [ 3 ] are included ([ Table 1 ]). It showed that no associations were observed between ESR1 rs2234693 or rs9340799 polymorphisms with T2DM ([ Tables 2 ] and [ 3 ]). However, when the two studies from ESR1 rs2234693 in which the genotype distribution for the control groups deviated from HWE were excluded, a significant association was observed in the homozygous model (CC vs. TT, OR=0.64; 95% CI 0.47–0.88, p=0.006, I 2 =39.15%) ([ Table 2 ], [ Fig. 1 ]). No change was observed from ESR1 rs9340799, although three studies deviated from HWE were excluded ([ Table 3 ]). No publication bias was observed among studies (p>0.05, data now shown). Together, these results suggest that homozygous of CC genotype ESR1 rs2234693 shows a protective effect against T2DM. Indeed, the estrogen receptor and its signaling play an important role in regulating insulin sensitivity [ 4 ]. Nevertheless, more studies, especially for ESR1 rs2234693, are needed to verify these results. Fig. 1 Association of homozygous model (CC vs. TT) in ESR1 rs2234693 polymorphism with T2DM. Table 1 Characteristics of individual studies for association between ESR1 polymorphisms and T2DM. rs2234693 ( Pvu II) No. Author [Ref] Year Country Sample size (T2DM/Control) Genotype distribution T2DM Control P HWE * Allele distribution T2DM Control TT TC CC TT TC CC T C T C 1 Akhmad et al. [ 2 ] 2013 Indonesia 40/25 9 18 13 4 14 4 0.2008 44 36 22 28 2 Chen et al. [ 5 ] 2015 China 20/40 4 10 6 4 25 11 0.0671 18 22 33 47 3 Dong et al. [ 6 ] 2002 China 104/60 16 62 25 7 26 27 0.8465 95 113 40 80 4 Ereqat et al. [ 3 ] 2019 Palestine 102/112 27 57 18 43 47 22 0.1685 111 93 133 91 5 Ganasyam et al. [ 7 ] 2012 India 100/100 39 43 18 52 32 16 0.0081 121 79 136 64 6 Huang et al. [ 8 ] 2006 China 299/341 36 172 91 34 149 158 0.8962 224 354 217 465 7 Meshkani et al. [ 9 ] 2012 Iran 152/299 52 73 27 64 173 62 0.0065 177 127 301 297 8 Mohammadi et al. [ 10 ] 2013 Iran 174/174 50 67 57 25 84 65 0.7981 167 181 134 214 9 Motawi et al. [ 11 ] 2015 Egypt 90/40 24 49 17 15 19 6 0.9967 97 83 49 31 rs9340799 ( XbaI ) No. Author [Ref] Year Country Sample size (T2DM/Control) Genotype distribution T2DM Control P HWE * Allele distribution T2DM Control AA AG GG AA AG GG A G A G 1 Akhmad et al. [ 2 ] 2013 Indonesia 40/25 4 25 11 18 33 14 0.8767 47 33 61 69 2 Chen et al. [ 5 ] 2015 China 20/40 3 5 12 8 25 7 0.1127 11 29 41 39 3 Dong et al. [ 6 ] 2002 China 104/60 11 74 19 21 38 1 0.001 96 112 80 40 4 Ereqat et al. [ 3 ] 2019 Palestine 102/112 44 44 24 40 101 73 0.6284 132 92 247 118 5 Ganasyam et al. [ 7 ] 2012 India 100/100 5 20 75 4 10 86 0.0001 30 170 22 186 6 Huang et al. [ 8 ] 2006 China 299/341 33 149 117 33 158 150 0.3526 383 215 458 214 7 Meshkani et al. [ 9 ] 2012 Iran 152/299 57 68 27 76 167 56 0.0348 182 122 319 279 8 Mohammadi et al. [ 10 ] 2013 Iran 174/174 53 60 61 24 77 73 0.6096 166 182 125 223 9 Motawi et al. [ 11 ] 2015 Egypt 90/40 34 39 17 20 15 5 0.4209 55 25 107 73 10 Speer et al. [ 12 ] 2001 Hungary 49/138 22 25 2 20 88 30 0.0009 69 29 128 148 * p-Values for Hardy–Weinberg equilibrium test in controls. Table 2 Meta-analysis for the association between ESR1 rs2234693 ( Pvu II) polymorphism and T2DM. All relevant articles were included (n=9) Contrast Test of association Model * Test of heterogeneity OR 95% CI p p I 2 (%) C vs. T 0.88 [0.70; 1.09] 0.26 Random 0.0056 63.08 CC vs. CT+TT 0.78 [0.58; 1.04] 0.09 Random 0.0632 45.94 CC+CT vs. TT 0.87 [0.57; 1.31] 0.51 Random 0.0008 70.03 CC vs. TT 0.73 [0.50; 1.07] 0.11 Random 0.0712 44.56 CT vs. TT 0.92 [0.59; 1.44] 0.73 Random 0.0005 71.20 Articles deviated for HWE were excluded (n=7) Contrast Test of association Model * Test of heterogeneity OR 95% CI p p I 2 (%) C vs. T 0.85 [0.66; 1.09] 0.21 Random 0.0257 58.26 CC vs. CT+TT 0.74 [0.51; 1.07] 0.11 Random 0.0558 51.19 CC+CT vs. TT 0.85 [0.53; 1.37] 0.51 Random 0.013 62.85 CC vs. TT 0.64 [0.47; 0.88] 0.006 Fixed 0.1307 39.15 CT vs. TT 0.92 [0.55; 1.55] 0.76 Random 0.0082 65.32 * If the p-value for Q-statistic was<0.10 or the I 2 value ≥50%, random-effect model was used, otherwise fixed-effect model was adopted. Table 3 Meta-analysis for the association between ESR1 rs9340799 ( Xba I) polymorphism and T2DM. All relevant articles were included (n=10) Contrast Test of association Model* Test of heterogeneity OR 95% CI p p I 2 (%) G vs. A 0.91 [0.65; 1.27] 0.604 Random 0.0001 85.36 GG vs. TC+CC 0.94 [0.62; 1.43] 0.794 Random 0.0002 72.04 GG+AG vs. AA 0.84 [0.48; 1.46] 0.541 Random 0.0001 84.24 GG vs. AA 0.89 [0.45; 1.73] 0.738 Random 0.0001 81.03 GA vs. AA 0.82 [0.48; 1.40] 0.479 Random 0.0001 80.20 Articles deviated for HWE were excluded (n=7) Contrast Test of association Model* Test of heterogeneity OR 95% CI p p I 2 (%) G vs A 0.93 [0.68; 1.28] 0.688 Random 0.0001 80.21 GG vs. TC+CC 1.00 [0.67; 1.47] 0.993 Random 0.0065 66.48 GG+AG vs. AA 0.77 [0.46; 1.29] 0.335 Random 0.0002 76.98 GG vs. AA 0.85 [0.46; 1.55] 0.608 Random 0.0004 75.91 GA vs. AA 0.72 [0.44; 1.18] 0.198 Random 0.0017 71.77 * If the p-value for Q-statistic was<0.10 or the I 2 value ≥ 50%, random-effect model was used, otherwise fixed-effect model was adopted.