To the Editor: Gustavo Turecki, M.D., Ph.D., and colleagues (1) recently suggested that an increase in the density of serotonin receptor 2A (5-HTR_2A) in the brains of suicide victims (2, 3) may be genetically mediated. They called for a reproduction of their genetic findings in an independent study group of similar composition. We investigated the T102C polymorphism in the 5-HTR_2A gene in brain samples (Brodmann’s area 9 of the cortex) from 24 depressed suicide victims and 31 matched comparison subjects and, in agreement with Dr. Turecki et al., found no significant differences in allelic or genotypic distribution between suicide victims and comparison subjects (4). In a subset of 10 depressed suicide victims and 15 comparison subjects, the density of 5-HTR_2A was significantly higher than in the comparison subjects, but there were no differences in receptor densities among the three genotypes of 5-HTR_2A: T/T, T/C, and C/C. We think this might have been because of our small group size—in particular, a low level of binding values in the T/T group.

We re-evaluated the phenotypic-genotypic relationship in a larger group (17 depressed suicide victims and 35 comparison subjects) of the same provenience and found that 5-HTR_2A binding in the brains of suicide victims carrying the T allele was higher (mean=153 fmol/mg protein, SD=20) than that of their respective comparison subjects (mean=114 fmol/mg protein, SD=13) (unpublished data). However, C allele carriers who committed suicide also had higher levels of 5-HTR_2A binding (mean=159 fmol/mg protein, SD=18) than did the comparison subjects carrying the same allele (mean=119 fmol/mg protein, SD=12). A two-way analysis of variance revealed a significant main effect of suicide on 5-HTR_2A binding levels (F=6.93, df=1, 100, p<0.01) but no significant effect on the two alleles of the 5-HTR_2A gene (F=0.20, df=1, 100, p=0.65).

We have thus confirmed the observation by Dr. Turecki et al. of no differences in 5-HTR_2A gene allelic or genotypic distribution between suicide victims and comparison subjects, as well as the effect of suicide on 5-HTR_2A densities in the frontal cortex. Our data agree with the result of their stepwise logistic regression. It was conducted with suicide as the main outcome and indicated that 5-HTR_2A binding, but not 5-HTR_2A genetic variation, was significant in predicting suicide. However, we were unable to confirm their finding that allelic variation significantly affects 5-HTR_2A densities in suicide victims and comparison subjects; we thus cannot support their suggestion that 5-HTR_2A binding is genetically determined by an allelic variation in the T102C polymorphism of the 5-HTR_2A gene.