Background: The oldest-old (≥85 years) have the highest rates of dementia. However, because of pronounced brain atrophy, standard neuroimaging programs face unique challenges in this cohort. Here, using data from both The 90+ Study and Alzheimer’s Disease Neuroimaging Initiative (ADNI), we report the prevalence of errors when processing brain MRIs from oldest-old participants using standard approaches and present some strategies for ameliorating these issues. Method: First, we analyzed image quality issues and segmentation errors (FreeSurfer) for the last available scans by age-groups (oldest-old, ≥85 years, vs younger-old, 55 to 85 years) in ADNI. Second, we evaluated the utility of an age-specific template (the 90+ T1w Template) for realigning MRIs from the oldest-old compared to a standard template (Montreal Neurological Institute T1w template, MNI) using linear and nonlinear registration (FSL) and evaluating alignment quality using visual and deformation (Jacobian) measures for oldest-old participants (≥90 years) in both The 90+ Study and ADNI. Lastly, we evaluated whether stringent preprocessing (bias field removal, denoising, brain extraction, intensity normalization using Computational Anatomy Toolbox (CAT12)) helped reduce FreeSurfer segmentation errors in frontal, parietal, and temporal regions in MRIs from the oldest-old. Comparisons were performed using Chi-square tests. Result: In ADNI data (Table-1), MRIs from oldest-old participants (∼10% of cohort) were around twice as likely to have image quality issues (20% vs 13%, P<0.001) and brain segmentation errors (45% vs 29%, P<0.001) compared to younger-old participants. Figure-1 shows a histogram of image quality issues and segmentation errors, which become more common with greater age. Compared to the MNI template, MRIs from participants aged 90 and older from both The 90+ Study (N=163, Table-2) and ADNI (N=75) had fewer registration errors (∼5% vs ∼23%, P<0.001 for all) and less severe deformation when using the 90+ T1w Template (Figure-2). Our proposed preprocessing pipeline (see Figure-3 for example) roughly halved segmentation errors in the oldest-old (P<0.001 for all regions, Figure-4). Conclusion: Our results show that adapting pipelines specifically for processing brain MRIs of oldest-old participants minimizes errors, enabling more effective use of neuroimaging to study neural correlates of neurodegeneration and dementia in this important portion of our aging population.

Hippocampal sclerosis of aging (HS-A) is a common age-related neuropathological lesion characterized by neuronal loss and astrogliosis in subiculum and CA1 subfield of hippocampus. HS-A is associated with cognitive decline that mimics Alzheimer's disease. Pathological diagnosis of HS-A is traditionally binary based on presence/absence of the lesion. We compared this traditional measure against our novel quantitative measure for studying the relationship between HS-A and other neuropathologies and cognitive impairment. We included 409 participants from The 90+ study with neuropathological examination and longitudinal neuropsychological assessments. In those with HS-A, we examined digitized H&E and LFB stained hippocampal slides. The length of HS-A in each subfield of hippocampus and subiculum, each further divided into three subregions, was measured using Aperio eSlide Manager. For each subregion, the proportion affected by HS-A was calculated. Using regression models, both traditional/binary and quantitative measures were used to study the relationship between HS-A and other neuropathological changes and cognitive outcomes. HS-A was present in 48 (12%) of participants and was always focal, primarily affecting CA1 (73%), followed by subiculum (9%); overlapping pathology (subiculum and CA1) affected 18% of individuals. HS-A was more common in the left (82%) than the right (25%) hemisphere and was bilateral in 7% of participants. HS-A traditional/binary assessment was associated with limbic-predominant age-related TDP-43 encephalopathy (LATE-NC; OR = 3.45, p < 0.001) and aging-related tau astrogliopathy (ARTAG; OR = 2.72, p = 0.008). In contrast, our quantitative approach showed associations between the proportion of HS-A (CA1/subiculum/combined) and LATE-NC (p = 0.001) and arteriolosclerosis (p = 0.005). While traditional binary assessment of HS-A was associated with impaired memory (OR = 2.60, p = 0.007), calculations (OR = 2.16, p = 0.027), and orientation (OR = 3.56, p < 0.001), our quantitative approach revealed additional associations with impairments in language (OR = 1.33, p = 0.018) and visuospatial domains (OR = 1.37, p = 0.006). Our novel quantitative method revealed associations between HS-A and vascular pathologies and impairment in cognitive domains that were not detected using traditional/binary measures.

TDP-43 proteinopathy is a salient neuropathologic feature in a subset of frontotemporal lobar degeneration (FTLD-TDP), in amyotrophic lateral sclerosis (ALS-TDP), and in limbic-predominant age-related TDP-43 encephalopathy neuropathologic change (LATE-NC), and is associated with hippocampal sclerosis of aging (HS-A). We examined TDP-43-related pathology data in the National Alzheimer's Coordinating Center (NACC) in two parts: (I) availability of assessments, and (II) associations with clinical diagnoses and other neuropathologies in those with all TDP-43 measures available. Part I: Of 4326 participants with neuropathology data collected using forms that included TDP-43 assessments, data availability was highest for HS-A (97%) and ALS (94%), followed by FTLD-TDP (83%). Regional TDP-43 pathologic assessment was available for 77% of participants, with hippocampus the most common region. Availability for the TDP-43-related measures increased over time, and was higher in centers with high proportions of participants with clinical FTLD. Part II: In 2142 participants with all TDP-43-related assessments available, 27% of participants had LATE-NC, whereas ALS-TDP or FTLD-TDP (ALS/FTLD-TDP) was present in 9% of participants, and 2% of participants had TDP-43 related to other pathologies ("Other TDP-43"). HS-A was present in 14% of participants, of whom 55% had LATE-NC, 20% ASL/FTLD-TDP, 3% Other TDP-43, and 23% no TDP-43. LATE-NC, ALS/FTLD-TDP, and Other TDP-43, were each associated with higher odds of dementia, HS-A, and hippocampal atrophy, compared to those without TDP-43 pathology. LATE-NC was associated with higher odds for Alzheimer's disease (AD) clinical diagnosis, AD neuropathologic change (ADNC), Lewy bodies, arteriolosclerosis, and cortical atrophy. ALS/FTLD-TDP was associated with higher odds of clinical diagnoses of primary progressive aphasia and behavioral-variant frontotemporal dementia, and cortical/frontotemporal lobar atrophy. When using NACC data for TDP-43-related analyses, researchers should carefully consider the incomplete availability of the different regional TDP-43 assessments, the high frequency of participants with ALS/FTLD-TDP, and the presence of other forms of TDP-43 pathology.