Abstract Background We examined whether brain amyloid PET, hippocampal volume, or plasma biomarkers are better predictors of conversion to mild cognitive impairment (MCI). Method In the Baltimore Longitudinal Study of Aging (BLSA), plasma Aβ 42 , Aβ 40 , glial fibrillary acidic protein (GFAP), and neurofilament light chain (NfL) concentrations were measured using Quanterix Simoa Neurology 4‐plex E. Plasma p‐tau181 and p‐tau231 concentrations were measured using in‐house Simoa assays at University of Gothenburg. Brain amyloid was quantified using mean cortical 11 C‐Pittsburgh compound B PET distribution volume ratio (mcDVR). Hippocampal volume was calculated from T 1 ‐weighted MRI and adjusted for intracranial volume. Of 178 cognitively unimpaired (CU) participants with both plasma and neuroimaging measurements at baseline, 33 developed MCI during follow up (median time to MCI onset: 3.7 [interquartile range (IQR) 2.1–6.2] years), 17 of whom had elevated brain PET amyloid at baseline. We used Cox proportional hazards models to examine the association between risk of conversion to MCI (time to conversion right‐censored at death or last visit for CU participants) and baseline plasma or neuroimaging biomarker, adjusting for age, sex, race, and education. To enable comparison of hazard ratios (HRs), we negated plasma Aβ 42 /Aβ 40 and hippocampal volume, and standardized each biomarker using statistics calculated in a cross‐sectional dataset of CU individuals aged 60–80 (preferring the visit closest to age 70 per participant) drawn from the larger BLSA. For standardization, we examined subtracting the median and dividing by the IQR and binarization (top tertile vs. not). Result In analyses using IQR normalized scores, mcDVR and plasma Aβ 42 /Aβ 40 and GFAP had statistically significant HRs (Table 2). In binarized biomarker analyses, mcDVR and plasma GFAP, p‐tau181, and p‐tau231 were statistically significant. Using all plasma biomarkers together yielded the highest concordance: 0.877±0.026 and 0.861±0.026 for the continuous and binarized biomarker analyses, respectively. Conclusion Plasma biomarkers provide information beyond demographics regarding the prediction of MCI conversion within 4 years. Risk prediction conveyed by several individual plasma biomarkers exceeded that of brain amyloid PET. Plasma GFAP was consistently identified as contributing to the prediction of MCI conversion, with Aβ 42 /Aβ 40 and p‐tau exhibiting statistically significant associations in a subset of analyses.
Abstract Background There is growing recognition that white matter microstructural integrity is affected in Alzheimer’s disease. The goal of this study was to characterize sex, racial/ethnic, and apolipoprotein (APOE)‐ε4 allele differences in white matter integrity. Methods This study included participants from ADNI, BLSA, ROS/MAP/MARS, and VMAP, all longitudinal cohorts of aging. This combined dataset included 6,837 imaging sessions from 2,619 participants age 50+ with diffusion MRI (dMRI) and demographic and clinical data (60% female, 31.4% APOE‐ε4 carriers, 78.9% White). dMRI was preprocessed using the PreQual pipeline. Free‐water (FW) correction was used to generate FW and FW‐corrected intracellular metrics including fractional anisotropy (FA FWcorr ), mean diffusivity (MD FWcorr ), axial diffusivity (AxD FWcorr ), and radial diffusivity (RD FWcorr ). Conventional and FW‐corrected metrics were harmonized using the Longitudinal ComBat package. Linear mixed‐effects models related sex, race/ethnicity, and APOE‐ε4 allele status to longitudinal diffusion metrics in 48 white matter tracts, adjusting for age at baseline, sex, education, race/ethnicity, APOE‐ε4 carrier status, cognitive status at baseline, and converter status. All models were corrected for multiple comparisons using the FDR approach. Result Sex differences in white matter were most notable in projection tracts (Figure 1A) and were primarily in FW‐corrected metrics. Females had lower FA FWcorr and higher RD FWcorr , indicative of worse microstructure, but lower AxD FWcorr . This sex difference was most pronounced for FA FWcorr in the ventral premotor projection tract (p=1.53x10 ‐62 ). There were global differences in white matter integrity by race/ethnicity (Figure 1B). Non‐Hispanic White participants tended to have higher conventional FA, FA FWcorr and AxD FWcorr and lower RD FWcorr . There was no association between APOE‐ε4 status and white matter integrity and no significant sex x race/ethnicity, sex x APOE‐ε4, or race/ethnicity x APOE‐ε4 interactions for conventional or FW‐corrected metrics when corrected for multiple comparisons. Conclusion There were striking sex and racial/ethnic (but not APOE‐ε4) differences in white matter tract integrity in a large cohort of aging adults. Female participants tended to have measures reflective of worse white matter integrity, and non‐Hispanic White participants tended to have measures reflective of greater integrity. Additional research exploring the etiology of these differences will be important to better understand disparities in Alzheimer’s disease.
Abstract Background We examined whether brain amyloid PET, hippocampal volume, or plasma biomarkers are better predictors of conversion to mild cognitive impairment (MCI). Method In the Baltimore Longitudinal Study of Aging (BLSA), plasma Aß42, Aß40, glial fibrillary acidic protein (GFAP), and neurofilament light chain (NfL) concentrations were measured using Quanterix Simoa Neurology 4‐plex E. Plasma p‐tau181 and p‐tau231 concentrations were measured using in‐house Simoa assays at University of Gothenburg. Brain amyloid was quantified using mean cortical 11C‐Pittsburgh compound B PET distribution volume ratio (mcDVR). Hippocampal volume was calculated from T1‐weighted MRI and adjusted for intracranial volume. Of 178 cognitively unimpaired (CU) participants with both plasma and neuroimaging measurements at baseline, 33 developed MCI during follow up (median time to MCI onset: 3.7 [interquartile range (IQR) 2.1‐6.2] years), 17 of whom had elevated brain PET amyloid at baseline. We used Cox proportional hazards models to examine the association between risk of conversion to MCI (time to conversion right‐censored at death or last visit for CU participants) and baseline plasma or neuroimaging biomarker, adjusting for age, sex, race, and education. To enable comparison of hazard ratios (HRs), we negated plasma Aß42/Aß40 and hippocampal volume, and standardized each biomarker using statistics calculated in a cross‐sectional dataset of CU individuals aged 60‐80 (preferring the visit closest to age 70 per participant) drawn from the larger BLSA. For standardization, we examined subtracting the median and dividing by the IQR and binarization (top tertile vs. not). Result In analyses using IQR normalized scores, mcDVR and plasma Aß42/Aß40 and GFAP had statistically significant HRs (Table 2). In binarized biomarker analyses, mcDVR and plasma GFAP, p‐tau181, and p‐tau231 were statistically significant. Using all plasma biomarkers together yielded the highest concordance: 0.877±0.026 and 0.861±0.026 for the continuous and binarized biomarker analyses, respectively. Conclusion Plasma biomarkers provide information beyond demographics regarding the prediction of MCI conversion within 4 years. Risk prediction conveyed by several individual plasma biomarkers exceeded that of brain amyloid PET. Plasma GFAP was consistently identified as contributing to the prediction of MCI conversion, with Aß42/Aß40 and p‐tau exhibiting statistically significant associations in a subset of analyses.
Abstract Infections have been associated with the incidence of Alzheimer disease and related dementias, but the mechanisms responsible for these associations remain unclear. Using a multicohort approach, we found that influenza, viral, respiratory, and skin and subcutaneous infections were associated with increased long-term dementia risk. These infections were also associated with region-specific brain volume loss, most commonly in the temporal lobe. We identified 260 out of 942 immunologically relevant proteins in plasma that were differentially expressed in individuals with an infection history. Of the infection-related proteins, 35 predicted volumetric changes in brain regions vulnerable to infection-specific atrophy. Several of these proteins, including PIK3CG, PACSIN2, and PRKCB, were related to cognitive decline and plasma biomarkers of dementia (Aβ 42/40 , GFAP, NfL, pTau-181). Genetic variants that influenced expression of immunologically relevant infection-related proteins, including ITGB6 and TLR5, predicted brain volume loss. Our findings support the role of infections in dementia risk and identify molecular mediators by which infections may contribute to neurodegeneration.
Abstract Background Cognitive impairment with age remains undetected until it interferes daily life activity or presents dementia symptoms. In the US, 61% of dementia population is not diagnosed, which is in part due to limited sensitivity of clinical neuroimaging modalities in assessing early gray matter (GM) changes. Here we look at microstructural changes in GM using mean apparent propagator (MAP‐MRI) in cognitively underperforming (CU) and healthy aging (HA) cohorts, grouped according to their cognitive performance based on the NIH Toolbox. Methods 725 subjects aged 36 to 90 years were included in this study using the Human Connectome Project‐Aging (HCP‐A) imaging dataset (T1w, multi‐shell diffusion ‐dMRI). Nine NIH Toolbox measures were used to classify CU and HA population groups ( Figure 1 ). Each test score was z‐normalized and subjects in the lower and upper 50 th percentile in all 9 tests were considered as CU and HA, respectively. Diffusion data was processed using TORTOISE dMRI processing package. Corrected DWIs were processed using MAP‐MRI, which yielded the following parameters: return to origin/axis/plane probability (RTOP, RTAP, RTPP), non‐Gaussianity (NG), and propagator anisotropy (PA) ( Figure 2a ). The T1w images were processed generating 125 SLANT cortical and subcortical regions of interest ( Figure 2b ). Two‐way ANOVA was used to investigate group differences while accounting for age, sex, education and site. All statistical analysis was done in Matlab2022b. Results The list of tests and respective scores of the two groups is provided in Figure 1. Zero‐displacement probabilities were more sensitive in detecting group differences compared to NG and PA. The two groups displayed notable differences in the following brain regions: the operculum, involved in visuospatial cognition; the precuneus, key to memory retrieval; and the posterior cingulate, which plays a role in processing speed, executive function, and memory. Additionally, the auditory and sensory‐related areas exhibited significant variations, aligning with expectations. These group differences are shown in Figure 3. Conclusion This study underscores the common occurrence of cognitive aging among the healthy population and demonstrates that MAP‐MRI is a valuable tool for assessing GM microstructural changes, particularly when comparing these changes to those observed in resilient aging populations.
Apolipoprotein E (APOE) ε4 allele is the strongest genetic risk factor for late onset Alzheimer's disease (AD). This case-cohort study used targeted plasma biomarkers and large-scale proteomics to examine the biological mechanisms that allow some APOEε4 carriers to maintain normal cognitive functioning in older adulthood.
