THE EFFECT OF INTELLIGENCE AND EDUCATIONAL ATTAINMENT ON THE BRAIN IN THOSE WITH FAMILIAL HIGH RISK FOR SCHIZOPHRENIA OR BIPOLAR DISORDER: AN ENIGMA-RELATIVES STUDY

Abstract Background Schizophrenia (SZ) and bipolar disorder (BD) are both associated with generally lower IQ test results and show overlapping structural brain abnormalities, albeit with lower effect sizes in BD. In contrast, our recent ENIGMA-Relatives meta-analysis showed that patients’ first-degree relatives (FDRs) have divergent patterns of global brain measures (De Zwarte et al. Biol. Psychiatry, 2019). FDRs-BD had larger intracranial volumes (ICV) than matched controls, a pattern not found in FDRs-SZ; when we adjusted for ICV, no differences were detected between FDRs-BD and controls. In contrast, FDRs-SZ had significantly smaller brain volumes, mean cortical thickness and larger ventricle volume than controls. Here, we extend this work by adding measures of local cortical thickness and surface area and by investigating the effect of IQ and educational attainment (EA) on global and local brain measures in FDRs. Methods 6,134 participants from 36 cohorts worldwide were included: N(FDRs-SZ)=1,103, N(FDRs-BD)=867, N(controls)=2,529 (and N(SZ-patients)=942, N(BD-patients)=693). Most cohorts provided information on IQ and/or EA (years completed education in those aged >25 yr.). (Sub-)cortical reconstruction and volumetric segmentations were performed with FreeSurfer. Linear mixed model analyses were performed on brain measures, IQ, and EA within each cohort comparing FDRs to controls (taking family relatedness into account). Cohen’s d effect sizes (95%CI) were calculated. Results FDRs-SZ had a thinner cortex across most cortical regions, compared to controls, with a thinner pars orbitalis surviving correction for multiple testing (left d=– 0.17, right =– 0.16, q<0.05 corrected). FDRs-BD had larger regional surface area in many cortical areas than controls, with a significantly larger cortical surface area in the left transverse temporal, left parahippocampal, right superior temporal, right supramarginal and right transverse temporal regions surviving correction for multiple testing (d’s >+ 0.15, q<0.05, corrected). Mean IQ test scores were lower in both FDRs-SZ (d=–0.42, p<0.001) and FDRs-BD (d=–0.23, p=0.045); while relatives did not differ on EA from controls. The IQ-EA correlation was r=0.39 [0.31–0.47]. When adjusting for IQ or EA, the group differences in brain measures changed, albeit modestly. In FDRs-SZ, controlling for IQ explained part of the effect of familial risk for schizophrenia in total brain, gray and white matter volumes (i.e., reduced effect sizes), while in FDRs-BD IQ correction resulted in a larger average ICV compared to controls. Discussion This study showed differential patterns of cortical thickness and surface area abnormalities in FDRs-SZ and FDRs-BD. While present in both relative groups, cognitive deficits (but only IQ not EA) were more pronounced in FDRs-SZ. We found no evidence that larger ICV in FDRs-BD was related to IQ, suggesting that the differential brain developmental trajectories underlying predisposition for schizophrenia or bipolar disorder may be unrelated to IQ. These large-scale studies inform the debate on whether schizophrenia and bipolar disorder represent truly independent diagnostic categories or whether they fall on a continuum of overlapping symptom profiles.