This year our group was the first to examine characteristics, specifically depression history and sex, that may influence cognitive phenotypes seen in unaffected siblings of patients with schizophrenia. Non-psychotic siblings share about half the genes with their affected sister or brother and show cognitive deficits in episodic, working memory, psychomotor speed, and executive functioning similar to those seen in patients albeit at a lesser degree. These deficits serve as intermediate phenotypes in studies of the genetics of schizophrenia. However, inconsistent neuropsychological performance results have been reported for sibling studies. These inconsistencies may be attributed to a range of task design features but also to be considered were sibling characteristics like psychiatric history, where treatment and remittance have been shown to resolve cognitive deficits. Also, there still remains the question about how genetic liability for schizophrenia may influence cognition of siblings. It is known that in the general population, there is an increased rate of depression for females compared to males and, that male patients with schizophrenia have worse symptomatology, course of illness, and cognitive impairments compared to female patients. Moreover, it is undetermined whether there is a similar sex effect on sibling cognition. Using unaffected siblings recruited through the Clinical Brain Disorders Branch Sibling Study( NIH protocol NCT00001486) and controls from the NIH Volunteer office, we identified 366 non-psychotic siblings of patients with schizophrenia and 680 unrelated healthy controls to analyze how sex interacts with a history of depression on a background of genetic risk for schizophrenia. Our subgroups consisted of the following: 603 healthy controls with no psychiatric history (276 males:327 females), 77 healthy controls with a history of depression (18 males:59 females), 271 siblings with no psychiatric history (119 males:152 females), and 95 siblings with a history of depression (25 males:70 females). All were given neurological exam, blood tests, and medical history review to determine any condition that might affect brain function. To minimize heterogeneity across sibling and control subgroups some exclusionary criteria included, but were not limited to 1) eliminating subjects with estimated IQs below 70, 2) controls could not have a first degree relative with a history of psychosis, and 3) excluding alternate psychiatric histories, like bipolar disorder or obsessive compulsive disorder. To examine how risk for schizophrenia, depression history, and sex influence cognitive ability, we generated a composite measure of cognition from 18 equally-weighted cognitive measures. The result being a six factor analysis-based domain score of cognitive ability, which was also used to examine effects across individual cognitive domains.
Our aim was to address the following hypotheses: 1) non-psychotic siblings of patients with schizophrenia would have cognitive impairments in verbal memory, working memory, executive functioning and processing speed;2) depression remitted controls would show that cognitive deficits associated with a history of depression will not contribute to cognitive impairment;and 3) a sex affect on cognition of siblings would show a similar pattern to patients with schizophrenia and impairments would be greater in males than females. We speculated that controls would have no sex affect on cognition, only sex differences in verbal and spatial domains. After taking into account demographic differences between unaffected siblings and unrelated controls we examined the effect of having risk for schizophrenia on cognitive performance and then determined whether depression history and sex impacted cognition. Our group had three key findings: 1) genetic risk for schizophrenia was associated with significant impairments in general cognitive ability. This impairment was especially associated with the siblings typically poor performance on measures of processing speed, card sorting ability, and spatial working memory as assessed with the N-back task. Siblings showed no memory impairment but had small verbal memory impairment which is in contrast with previously published reports. This difference between visual and verbal memory could be due in part to differences in test batteries, psychometric properties of our assessments, covariates used, or other study details. 2) The prevalence of a depression history was significantly higher in siblings than in controls, and such a history affected sibling cognition on measures of processing speed, card sorting ability, and minimally on N-back performance. We confirmed that a history of depression did not affect cognition in controls. And lastly, we observed an interaction between schizophrenia risk, depression history, and sex, which showed important differences between subgroups. Depression history and sex interacted to influence cognition solely in siblings. In subgroups, the greatest impairments were observed in male siblings with depression history;cognition in this group scored well below that of male siblings without a history of depression. This effect was demonstrated most prominently in card sorting ability, where the overall pattern of cognition associated with this interaction was observed in 5 out of 6 cognitive domains, indicating cognitive compromised associated with a history of depression in male siblings has a broad rather effect on cognition. In female siblings cognition did not significantly differ as a function of depression history in any cognitive domain. An important implication can be derived from this study which is that cognition in siblings is not homogenous, but differs as a function of subgroup characteristics. Our results suggest the importance of considering the characteristics of subgroups when conducting genetic association studies of cognitive phenotypes in schizophrenia in order to reduce sample heterogeneity. Additional studies will be needed to confirm our findings and to determine whether subgroups differ in other dimensions (e.g. neuroimaging) and how the phenotypes relate back to the patients with schizophrenia.
Showing the most recent 10 out of 24 publications
