Hu, Mingzhao

Abstract Background Multi‐site aging and dementia studies are challenged by data heterogeneity caused by scanner (manufacturer and protocol) related variability. Quantitative analysis of white matter hyperintensity (WMH) from pooled samples, a traditional marker of cerebrovascular disease, is sensitive to machine variability. Our goal was to compare Deming (symmetric) linear regression and ComBat, a common method used for harmonizing neuroimaging measurements, for harmonizing WMH measurements on participants scanned on two different scanners (GE and Siemens) and protocols (2D FLAIR on GE and 3D FLAIR on Siemens). Methods Cognitively unimpaired (CU) ( n =53, age (mean (SD))=67 (15) years) and cognitively impaired (CI; MCI/dementia) ( n =34, age (mean (SD))=82 (9) years) participants from the Mayo Clinic Study of Aging and Mayo ADRC were scanned on GE and Siemens scanners within a week of each other (typically done on the same day). WMH as a percentage of total intracranial volume was measured using multispectral segmentation of FLAIR and T1‐weighted images (Gunter et. al. AAIC 2023). We used Deming regression as it includes errors in both input dimensions. ComBat was applied using age, sex, and diagnosis as covariates, with scanner vendor and protocol as the batch variable. We assessed the inter‐scanner similarity using Maximum Mean Discrepancy (MMD) analysis, where a value of zero indicates identical distributions, p >0.05 indicates no significant difference under H 0 . (Gretton et al., JMLR 2012). Results Our results showed that the distribution of WMH measurements, which were different in the CU group, were more similar after Deming than ComBat harmonization (Figure 1). This was confirmed by the lower MMD in the CU group from Deming regression (‐0.01 ( p =0.76) from 0.03 ( p =0.05)) compared to ComBat (0.04 ( p =0.02))(Figure 1). However, ComBat and Deming were unable to create distributions as compatible as original measurements for the CI group. Conclusion Using a unique dataset of the same participants scanned close in time on different scanner and protocols, we found that (1) MMD measurements were useful to test the need/improvement due to harmonization; (2) ComBat was not substantially better at creating compatible distributions between scanner vendors than Deming regression, and both were sometimes inadequate. Future work will require alternative models to linear scale and shift transformations.

Abstract Background Cerebrovascular disease (CeVD), a common co‐pathology with Alzheimer’s Disease, has not been sufficiently operationalized, limiting its usage for accurate estimation of vascular contributions to cognitive impairment (VCI). The proposed CeVD biomarker classification uses more recently developed diffusion MRI markers (fractional anisotropy of Genu (FA), free water (FW), peak width of skeletonized mean diffusivity (PSMD), and ARTS score (ARTS)) as surrogates of early CeVD changes and white matter hyperintensities (WMH) as surrogate of late CeVD changes. Our primary goal was to characterize the temporal progression of these biomarkers to enhance CeVD biomarker staging. Our secondary goal was to establish cut‐points for normal/abnormal CeVD biomarkers. Methods Using CeVD biomarker data from 3,953 Mayo Clinic Study of Aging participants (split 80‐20 for training‐testing with training used for cut‐point estimation), we classified participants into VCI+/VCI‐ (Fazekas≥2 and Clinical Dementia Rating (CDR)≥0.5/not). We evaluated cut‐points using four criteria: specificity‐based, sensitivity‐based, accuracy young referent (VCI+ vs. 30‐49 yrs VCI‐), and accuracy age‐/sex‐matched referent (VCI+ vs. age‐/sex‐matched VCI‐). After selecting a cut‐point based on sensitivity and positive predictive value (PPV), we used nonlinear mixed‐effects (NLME) models to examine temporal progression in 2,047 participants (aged 43.39‐92.51 years, 47% female) with longitudinal data incorporating participant‐specific time shifts, correlations between random effects of each biomarker, and covariates (age, sex, vascular risk, education, APOE4 status). Results The accuracy young referent criterion produced the best sensitivity‐PPV tradeoff across all diffusion MRI markers (Figure 1). Diffusion MRI biomarkers consistently progressed 10 years earlier than WMH (Figure 2). Whole‐brain diffusion MRI measures (FW, PSMD, ARTS) were highly correlated with WMH, showing synchronized progression curves with a time offset. The regional genu measure (interhemispheric disconnection) progressed earlier with a steadier slope across decades and had comparatively lower correlation with WMH. Conclusions We developed cut‐points for diffusion MRI markers using classification of VCI based on Fazekas scale and CDR. Cut‐points from the accuracy young referent (VCI+ vs. 30‐49 yrs VCI‐) outperformed the accuracy age‐/sex‐matched referent (VCI+ vs. age‐/sex‐matched VCI‐), suggesting that older VCI‐ individuals may have unmeasured CeVD changes. Diffusion MRI markers progressed earlier relative to WMH, highlighting their potential for early detection, improved clinical assessment, and targeted interventions.

Abstract Background Despite advances in Alzheimer’s disease (AD) research, limited information exists regarding the absolute risk of mild cognitive impairment (MCI) in cognitively unimpaired (CU) individuals with abnormal AD biomarkers, particularly when accounting for competing risks of death. This knowledge gap is critical as the field addresses disease‐modifying treatments targeting CU individuals with preclinical AD. We focus on the predictive value of AD amyloid stages defined by amyloid PET Centiloid values. Methods We included 5,858 participants from the Mayo Clinic Study of Aging (MCSA) to evaluate AD amyloid stage as a predictor of clinical progression to MCI or dementia. The data includes long‐term follow‐up information on death and dementia beyond active study participation, which mitigates potential bias due to dropout. We predicted 10‐year and lifetime risks of MCI and dementia, accounting for the competing risks of death, given amyloid PET stages, sex, APOE4 status, and baseline age. Results are based on a hidden Markov model. Results AD amyloid staging based on amyloid PET Centiloid values was the strongest predictor of lifetime risk for MCI or for dementia (Figure 1). Higher Centiloid levels amplified age effects on the risk of MCI, whereas for dementia, amyloid stage effects surpassed age effects. For 10‐year risk, age was the dominant factor, whereas for lifetime risk, amyloid stages had a greater influence. While smaller in magnitude than the effects of amyloid stages and age, APOE4 carrier status and (on average) male sex were also associated with increased risk of MCI and dementia. For a female APOE4 carrier at age 65, lifetime risk was 41%/61%/76% for low/moderate/high amyloid PET levels. The 10‐year absolute risk for a female APOE4 carrier with high amyloid PET was 4%/29%/58% when assessed at starting ages 65, 75, and 85. Conclusions AD amyloid staging is a critical predictor of lifetime risk for cognitive impairment, underscoring its importance in guiding treatment decisions if interventions are approved for preclinical AD in the future. Effective patient evaluation should focus on biomarker‐based amyloid staging rather than simple binary biomarker classification, with careful consideration of age, sex, and APOE4 status to ensure a comprehensive benefit‐risk assessment for future therapeutic strategies.