27-Hydroxycholesterol, cognition, and brain imaging markers in the FINGER randomized controlled trial

Table 3 Longitudinal associations between change in 27-OH and change in cognition and neuroimaging markers

	Control, β (p)	Intervention, β (p)	Interaction, p
Change in cognition
NTB total composite	0.11 (0.62)	− 0.46 (0.04)	0.07
NTB memory	0.12 (0.58)	− 0.54 (0.01)	0.02
NTB processing speed	−0.04 (0.85)	0.24 (0.31)	0.41
NTB executive function	0.12 (0.59)	−0.31 (0.17)	0.18
Change in MRI
Total gray matter volume	0.18 (0.45)	−0.39 (0.10)	0.08
Hippocampal volume	0.06 (0.79)	−0.28 (0.25)	0.30
AD signature cortical thickness	0.05 (0.84)	−0.20 (0.42)	0.50
White matter lesion volume	−0.05 (0.82)	−0.05 (0.85)	0.97
Change in FDG-PET
Cortical composite	0.24 (0.32)	0.01 (0.97)	0.42
Prefrontal cortex	0.08 (0.74)	0.04 (0.89)	0.85
Parietal cortex	0.21 (0.36)	0.11 (0.70)	0.68
Lateral temporal cortex	0.20 (0.40)	−0.03 (0.91)	0.46
Precuneus	0.23 (0.33)	−0.04 (0.89)	0.41
Anterior cingulate gyrus	0.25 (0.29)	−0.05 (0.85)	0.34
Posterior cingulate gyrus	0.26 (0.27)	0.00 (0.99)	0.34
Medial temporal cortex	0.25 (0.30)	0.13 (0.64)	0.59
Lateral occipital cortex	0.23 (0.32)	−0.28 (0.29)	0.18
Striatum	0.38 (0.10)	−0.08 (0.77)	0.15
Cerebellum	−0.01 (0.97)	0.22 (0.41)	0.50
Change in PIB-PET
Cortical composite	−0.04 (0.87)	−0.10 (0.69)	0.81

Values are standardized β coefficients (p values) from linear regression models with change in cognitive and neuroimaging measures (left column) as dependent variables and change in 27-OH as an independent variable. Change in all measures was calculated as the difference between 2-year and baseline values. Values for the control and intervention groups are shown from analyses stratified by the randomization group. p value for interaction is shown from models including change in 27-OH, randomization group, and their interaction. p value considered significant (bold) if < 0.05. AD signature cortical thickness was calculated as the average of cortical thickness in the entorhinal, inferior temporal, middle temporal, and fusiform regions. The changes in composite scores for FDG-PET and PiB-PET were determined as the average change of the prefrontal, parietal, lateral temporal, anterior cingulate, posterior cingulate, and precuneus ROIs
NTB, neuropsychological test battery; MRI, magnetic resonance imaging; AD, Alzheimer’s disease; FDG-PET, 2-deoxy-2-fluorine-18fluoro-d-glucose positron emission tomography; PiB, 11 C-Pittsburgh compound B

ISSN: 1758-9193