Search

Background: MRI-derived brain-age prediction is a promising biomarker of biological brain aging. Accelerated brain aging has been found in Alzheimer’s disease (AD) and other neurodegenerative diseases. However, no previous studies have investigated the relationship between specific pathophysiological pathways in AD and biological brain aging. Here, we studied whether glial activation and synaptic dysfunction are associated with biological brain aging in the earliest stages of the Alzheimer’s continuum. Method: We included 418 cognitively unimpaired individuals (CU) from the ALFA+ study with available structural MRI, and CSF biomarkers of amyloid-ß (Aß42/40) and tau pathology (p-tau181), synaptic dysfunction (neurogranin, GAP43, SYT1, SNAP25), glial activation (sTREM2, YKL40, GFAP, interleukin-6 and S100b) and a-synuclein (Table 1). Aß42/40, neurogranin and the glial activation biomarkers were measured using the Roche NeuroToolKit. We computed brain-age delta as the difference between chronological and predicted brain-age. The latter was estimated using a previously pretrained machine learning algorithm on cerebral morphological measurements on individuals from the UKBioBank cohort (N = 22.000). General linear modeling was used to test the associations between CSF biomarkers and brain-age delta, adjusting by p-tau, age, APOE status and sex. For the biomarkers whose associations were significant, we evaluated the interaction term “biomarker” × AT status while adjusting by age, APOE status and sex. AT staging was performed using pre-established cut-off values. We then used hippocampal volume as a marker of AD-related neurodegeneration and repeated the same association studies with CSF biomarkers, adjusting by p-tau, age, APOE status, sex and TIV. Result: Brain-age delta was negatively associated with CSF sTREM2 (Padjusted<0.001), meaning that younger-appearing brains showed higher levels of this biomarker (Table 1). None of the other biomarkers survived multiple compar, Peer Reviewed, Postprint (author's final draft)

Additional file 1: Supplementary Table 1. Characteristics of Single Nucleotide Polymorphisms (SNPs) associated with longer telomere length. The effect allele refers to the allele that is associated with longer telomere length. Chromosomal position of the SNPs (genome assembly GRCh37 (hg19)) according to the public archive for genetic variation within and across different species developed and hosted by the National Center for Biotechnology Information (NCBI) in collaboration with the National Human Genome Research Institute (NHGRI) (dbSNP).

Additional file 3: Supplementary Table 1. Characteristics of the study participants with information for cognition outcomes. Mean and SD are shown for continuous variables. Supplementary Table 2. Characteristics of the study participants with information for neuroimaging outcomes. Mean and SD are shown for continuous variables. Supplementary Table 3. Characteristics of the study participants with information for CSF biomarkers. Mean and SD are shown for continuous variables.

Additional file 4: Supplementary Table 1. Results of the effect of genetically predicted longer telomere length on AD endophenotypes in the entire sample. Supplementary Table 2. Results of the effect of genetically predicted longer telomere length on AD endophenotypes among APOE-ɛ4 carriers. Supplementary Table 3. Results of the effect of genetically predicted longer telomere length on AD endophenotypes among APOE-ɛ4 non-carriers. Supplementary Table 4. Results of the effect of genetically predicted longer telomere length on AD endophenotypes among individuals at high genetic predisposition to Alzheimer's disease. Supplementary Table 5. Results of the effect of genetically predicted longer telomere length on AD endophenotypes among individuals at low genetic predisposition to Alzheimer's disease.

Additional file 2: Supplementary Table 1. Linear regression estimates for cognition outcomes in the entire sample. All models are adjusted for covariates: age, sex, education, and APOE status. Supplementary Table 2. Linear regression estimates for neuroimaging outcomes (i.e., Alzheimer’s disease and aging signatures) outcome in the entire sample. All models are adjusted for covariates: age, sex, education, and APOE status. Supplementary Table 3. Linear regression estimates for CSF biomarkers outcomes in the entire sample. All models are adjusted for covariates: age, sex, education, and APOE status. Supplementary Table 4. Linear regression estimates for cognition outcomes in APOE-ɛ4 carriers. All models are adjusted for covariates: age, sex, and education. Supplementary Table 5. Linear regression estimates for neuroimaging outcomes (i.e., Alzheimer’s disease and aging signatures) in APOE-ɛ4 carriers. All models are adjusted for covariates: age, sex, and education. Supplementary Table 6. Linear regression estimates for CSF biomarkers outcomes in APOE-ɛ4 carriers. All models are adjusted for covariates: age, sex, and education. Supplementary Table 7. Linear regression estimates for cognition outcomes in APOE-ɛ4 non-carriers. All models are adjusted for covariates: age, sex, and education. Supplementary Table 8. Linear regression estimates for neuroimaging outcomes (i.e., Alzheimer’s disease and aging signatures) in APOE-ɛ4 non-carriers. All models are adjusted for covariates: age, sex, and education. Supplementary Table 9. Linear regression estimates for CSF biomarkers outcomes in APOE-ɛ4 carriers. All models are adjusted for covariates: age, sex, and education. Supplementary Table 10. Linear regression estimates for cognition outcomes among individuals at high genetic predisposition to AD. All models are adjusted for covariates: age, sex, and education. Supplementary Table 11. Linear regression estimates for neuroimaging outcomes (i.e., Alzheimer’s disease and aging s