Studying biomarkers for Alzheimer’s disease (AD) in cognitively healthy, identical twins can be a useful approach to understand how genetic and environmental risks contribute to AD. Previous studies showed that regionally specific cortical thinning is an early diagnostic marker for AD and can be detected in asymptomatic amyloid positive individuals. Heritability for cortical thickness varies greatly across the cortex with the highest value (76% of variance explained by additive genetic factors) for frontal regions. This is the first study investigating correlations for cortical thickness in non-demented elderly twins and aims to determine the contribution of genetic and environmental influences on this AD-biomarker.


      We studied the first 42 monozygotic (MZ) twin pairs (n = 84) from the ongoing NTR-Twin substudy of the EMIF-AD PreclinAD project, which aims to enroll in total 100 twin pairs. Inclusion criteria were age > 60 years and an age, gender and education corrected CERAD delayed recall z-score > -1.5 SD. High resolution 3D T1-weighted structural MRI scans were acquired on a 3.0T Philips Intera MR scanner. Cortical thickness was estimated automatically using the publicly available Freesurfer software package. MZ twin correlations for cortical thickness (adjusted for age and gender) were calculated in SPSS.


      Subjects were on average 68 (SD 7.9) years old (range:60-86; 65.5% female; MMSE mean(SD)=28.9(1.4)). MZ twin correlations for cortical thickness ranged from .316 (left paracentral) to .825 (right postcentral) (figure 1). MZ correlations for AD signature regions, including entorhinal, inferior temporal, parahippocampal, inferior parietal, precuneus, superior parietal, supramarginal, parsopercularis, parsorbitalis, parstriangularis, and superior frontal cortices and temporal pole, were on average .57. Left and right hemisphere showed similar correlations. Correlations assessed in randomly paired subjects were lower ( < .2) and not significant.


      Cortical thickness correlates between MZ twins. These data suggest that in addition to a strong genetic background for cortical thickness in cognitively healthy elderly also non-genetic factors substantially influence this AD-biomarker. Future studies on the causes of identical twin discordance may provide novel insights into the pathophysiology of AD which may provide clues for AD prevention.
      Figure thumbnail fx1