The endophenotype approach to studying the genetics of psychosis relies on a shared genetic origin between the endophenotype in question and the psychosis phenotype. This is usually inferred by the presence of impairments in the unaffected relatives of patients. However, population-based studies including information on relatedness, the putative endophenotype, and the psychosis phenotype offer the opportunity to test this association at the population level. In the first such study, Fowler and colleagues set out to examine the relationship between genetic influences in premorbid IQ and psychosis in a Swedish longitudinal, population-based study. Genetic relatedness was taken from the Swedish Multigeneration Register and Twin Registry, allowing the identification of pairs of MZ and DZ twins and sibling pairs with an age gap no greater than five years. A proxy for IQ was taken from the Military Conscription Register (thus restricting the analysis to males), and the psychosis phenotype from the Hospital Discharge Register.
In agreement with previous studies (Toulopoulou et al., 2007), both IQ and psychosis had high heritabilities. However, the phenotypic correlation between IQ and psychosis was relatively modest, at -0.11. Although 91 percent of this correlation was explained by genetic influences, the small size of the correlation itself meant that only 7 percent of the genetic variance for psychosis was shared with IQ at the population level; the remaining 93 percent of heritability of psychosis is independent of IQ.
This study shows the importance of examining these associations at the population level using unselected samples. Previous studies (Toulopoulou et al., 2007), using case-control designs, have been able to measure the phenotypic correlation between IQ and psychosis, which was found to be high, as in this study. However, what this study adds is that the low phenotypic correlation at the population level translates into disappointingly low overlap in genetic variance explaining IQ and psychosis at the population level. However, these results do not rule out the presence of genetic variants (such as CNVs) having a large effect on both psychosis and IQ in specific families.
1. Toulopoulou T, Picchioni M, Rijsdijk F, Hua-Hall M, Ettinger U, Sham P, Murray R. Substantial genetic overlap between neurocognition and schizophrenia: genetic modeling in twin samples. Arch Gen Psychiatry. 2007;64:1348-1355. AbstractView all comments by James MacCabe
Objections are principled but not data driven
Fowler and colleagues (Fowler et al., 2012) should be warmly congratulated on their study examining the relationship between IQ and psychotic disorders in a population-based cohort. Fowler’s excellent study addresses in a superior design some of the limitations that we recognized in our previous studies (Toulopoulou et al., 2007; Toulopoulou et al., 2010). We are therefore very grateful to them for their contribution.
However, there are a number of reasons that might suggest that the interpretations of Fowler et al. are overstated. For example, dismissing a phenotypic correlation of -0.22 between premorbid IQ and psychosis, when most of this appears to relate to shared genetic covariance, is, in my opinion, an omission. Specifically, discarding up to 14 percent of net shared genetic influences between IQ and psychosis—when the effect sizes for our best candidate risk genes are consistently tiny—makes me wonder why, exactly, IQ as an intermediate phenotype to identify genes for schizophrenia would be an unlikely successful strategy. It might turn out to be an unsuccessful strategy, but a genetic overlap of 7-14 percent should not be passed over so quickly. At least not until we have better alternatives.
The above is an edited version of a comment posted in the Archives of General Psychiatry on 13 September 2012. See the full comment.
Fowler T, Zammit S, Owen MJ, Rasmussen F. A population-based study of shared genetic variation between premorbid IQ and psychosis among male twin pairs and sibling pairs from Sweden. Arch Gen Psychiatry . 2012 May ; 69(5):460-6. Abstract
Toulopoulou T, Picchioni M, Rijsdijk F, Hua-Hall M, Ettinger U, Sham P, Murray R. Substantial genetic overlap between neurocognition and schizophrenia: genetic modeling in twin samples. Arch Gen Psychiatry . 2007 Dec ; 64(12):1348-55. Abstract
Toulopoulou T, Goldberg TE, Mesa IR, Picchioni M, Rijsdijk F, Stahl D, Cherny SS, Sham P, Faraone SV, Tsuang M, Weinberger DR, Seidman LJ, Murray RM. Impaired intellect and memory: a missing link between genetic risk and schizophrenia? Arch Gen Psychiatry . 2010 Sep ; 67(9):905-13. Abstract
View all comments by Timothea Toulopoulou