Associations of endocrine stress-related gene polymorphisms with risk of autism spectrum disorders: Evidence from an integrated meta-analysis.

Autism spectrum disorders (ASD) are related to serotonin transporter (5-HTT) and catechol-O-methyl transferase (COMT) as two most monoaminergic polymorphic variations. However, multiple studies assessing rs4680 and 5-HTTLPR variants in ASD have reported inconsistent results. Therefore, we conducted an integrated meta-analysis to combine case-control and transmission/disequilibrium test (TDT) studies to determine whether COMT and 5-HTT are associated with ASD. We searched multiple electronic databases (PubMed, EmBase and Web of Science) to identify studies assessing the rs4680 and 5-HTTLPR variants in ASD from Jan 1997 to Dec 2016. Then allelic data from case-control and TDT studies were analyzed by the Catmap package in the R software. A total of 5 studies were eligible for the meta-analysis of rs4680, including 3 case-control, 1 TDT and 1 TDT & case-control studies. Meanwhile, 22 studies of 5-HTTLPR were available, including 16 TDT, 4 case-control and 2 TDT & case-control studies. The current meta-analysis included 814 ASD cases, 741 controls and 311 families related to rs4680; 749 ASD cases, 1,118 controls and 1,861 families relevant to 5-HTTLPR were also evaluated. For rs4680, the pooled OR was 1.18 (95% CI=0.87-1.59, P=0.29, P-heterogeneity<0.00001). There was no significant association of rs4680 with risk of ASD between the two subgroups. For 5-HTTLPR, the pooled OR was 1.05 (95% CI=0.92-1.20, P=0.4652, P-heterogeneity<0.00001). Meanwhile, we found no significant risk in individual case-control or TDT studies. The above findings indicated that neither COMT rs4680 nor 5-HTT 5-HTTLPR polymorphism significantly affects ASD risk. Autism Res2017, 10: 1722-1736. (c) 2017 International Society for Autism Research, Wiley Periodicals, Inc. Lay SummaryOur results showed no evidence of significant association of either COMT rs4680 or 5-HTT 5-HTTLPR variants with ASD, showing that these two genes may not be major susceptible genetic factors in ASD occurrence, and may have a reciprocal action with each other in combination with environmental factors. These findings further provide evidence that a single gene variant may not dictate autism occurrence, but possibly contributes to a specific phenotype or subtype of ASD.