In this study, we explored the association of the T102C (rs6313) polymorphism of the 5-HTR2A gene with suicidal behavior. First, a case–control study was conducted. Subsequently, we performed a meta-analysis to assess the current evidence of association between rs6313 and suicidal behavior. In the first part, we found an association between rs6313 and suicidal behavioral in a Mexican sample. To our knowledge, this is the first study addressing the genetic association between the T102C allele and suicidal behavior in a Mexican population.
The genetic analysis revealed a slight preponderance of the T allele in the cases group. In 2000, Du et al. described an association between the 102C allele and ideation suicide . Since 2000, growing interest on this issue prompted a variety of studies in Caucasian and Asian populations dealing with 5HTR2A and suicidal behavior. Although many studies have reported this association [21, 22, 33, 42], there are other reports that have not encountered an association of either the T or C allele in rs6313 of 5HTR2A [20, 23, 31, 32, 35, 38, 43] with suicidal behavior. The slight preponderance of the T allele is in agreement with other reports in the literature. In 1999, Tsai et al.  reported that the allele frequency of 102T is higher than that of 102C in both patient and control groups. Recently, Saiz et al.  found an excess of the 102T allele in a group of 193 patients and a control group of 420 individuals when comparing non-impulsive suicide attempts with impulsive suicide attempts. This evidence suggests that the rs6313 polymorphism of the 5-HTR2A gene may be involved in suicidal behavior. However, no conclusive outcomes have yet been attained. In agreement, our results in the Mexican population suggest that allele 102T of 5HTR2A could be associated with suicidal behavior.
There are several possible explanations for the discrepancies in the results regarding this association. First, there are differences in the populations of patients. For example, Du et al.  and other authors [13, 23] have studied patients with major depression and suicidal behavior; Wrzosek et al. studied patients with attempted suicides and alcohol dependence ; other studies have analyzed suicidal behavior in patients with schizophrenia , and finally, there are studies such as ours that have analyzed only suicide attempters [8, 31, 36]. Second, there are differences in genetic heterogeneity. Several authors have shown that allele frequencies of the T102C polymorphism may be ethnic-dependent [34, 36, 44]. However, our sample was made of a relatively homogenous Mexican population, because only patients of the state of Tabasco with parents and grandfathers born in Tabasco were included in this study. In consequence, we considered necessary to carry out a meta-analysis of all the published evidence concerning rs6313 and suicidal behavior.
In the current meta-analysis we utilized the allelic model and ethnicity groups. There are two previous meta-analyses -the last including 25 studies- that have explored the association between rs6313 and suicidal behavior [41, 45]. Both meta-analyses reported no association between rs6313 and suicidal behavior. However, several additional case–control studies concerning this relationship and using larger samples have been recently reported, but they are not included in the above-mentioned meta-analyses. Therefore, to get a more comprehensive estimation we included our case–control study.
We performed a meta-analysis for the C (OR: 1.14; 95% CI 0.96-1.35) and T (OR: 0.99; 95% CI: 0.90-1.08) alleles separately; however, no association was observed. The results suggest no association between rs6313 (T102C) and suicidal behavior. One possibility for this lack of association may lie in the different criteria used in the selection of patients. In addition, samples sizes of several of the studies included in our meta-analysis are in the low range compared with genetic studies for other diseases [7, 46].
There is the possibility that the C allele may be regarded as a factor risk in the Caucasian population, because in a first analysis we observed an association when we considered a fixed effects with heterogeneity. However, we adopted the random effects, which accounts for additional sources of inter-study variation when heterogeneity exists. This model is more conservative than the fixed effects, since the latter assumes the same true genetic effects, whereas the former assumes normally distributed effects and parametrizes inter-studies variation . When we performed a second analysis, in which the studies giving rise to heterogeneity were discarded, no association was observed between rs6313 and suicidal behavior in a Caucasian population. Therefore, future studies on suicidal behavior comprising larger samples are important to determine this association. Similarly, the accuracy of the phenotype definition in the association studies is a relevant issue . However, it is necessary to take into consideration other aspects such as different inheritance patterns and gene X environment interaction .
Finally, we recognize some limitations in the present study; 1) the sample size of the case–control study is small and may not have sufficient power to detect an association between suicidal behavior and small effects polymorphism. In relation to our meta-analysis, there are also some considerations to bear in mind: first; included studies were limited to published articles; second, we did not perform sub-analyses including attempted or completed suicide; third, we did not carry out a subgroup analysis based on gender, and fourth, we did not report psychopathologies related to suicide attempters.