Debby W. Tsuang’s research while affiliated with University of California, San Francisco and other places

Background and objectives: The clinical diagnosis of dementia with Lewy bodies (DLB) depends on identifying significant cognitive decline accompanied by core features of parkinsonism, visual hallucinations, cognitive fluctuations, and REM sleep behavior disorder (RBD). Hyposmia is one of the several supportive features. α-Synuclein seeding amplification assays (αSyn-SAAs) may enhance diagnostic accuracy by detecting pathologic αSyn seeds in CSF. In this study, we examine how different clinical features associate with CSF αSyn-SAA positivity in a large group of clinically diagnosed participants with DLB. Methods: Cross-sectional and longitudinal CSF samples from the multicentered observational cohort study of the DLB Consortium and similar studies within the Parkinson's Disease Biomarker Program, contributed by academic medical centers in the United States, underwent αSyn-SAA testing. Participants included those clinically diagnosed with DLB and 2 control cohorts. Associations between core DLB features and olfaction with αSyn-SAA positivity were evaluated using logistic regression. Results: CSF samples from 191 participants diagnosed with DLB (mean age 69.9 ± 6.8, 15% female), 50 age-matched and sex-matched clinical control participants, and 49 younger analytical control participants were analyzed. Seventy-two percent (137/191) of participants with DLB had positive αSyn-SAAs vs 4% of the control groups. Among participants with DLB, those who were αSyn-SAA-positive had lower Montreal Cognitive Assessment scores (18.8 ± 5.7 vs 21.2 ± 5.2, p = 0.01), had worse parkinsonism on the Movement Disorders Society Unified Parkinson's Disease Rating Scale part III (33.8 ± 15.1 vs 25.6 ± 16.4, p = 0.001), were more likely to report RBD (114/133 [86%] vs 33/53 [62%], p < 0.0001), and had worse hyposmia on the University of Pennsylvania Smell Identification Test (UPSIT) (94/105 [90%] below 15th percentile vs 14/44 [32%], p < 0.0001). UPSIT percentile had the highest area under the curve (0.87, 95% CI 0.81-0.94) in predicting αSyn-SAA positivity and participants scoring at or below the 15th percentile of age and sex normative values had 18.3 times higher odds (95% CI 7.52-44.6) of having a positive αSyn-SAA test. Among 82 participants with longitudinal CSF samples, 81 (99%) had the same αSyn-SAA result for initial and follow-up specimens. Discussion: A substantial proportion of clinically diagnosed participants with DLB had negative αSyn-SAA results. Hyposmia was the strongest clinical predictor of αSyn-SAA positivity. Hyposmia and αSyn-SAA may have utility in improving the diagnostic assessment of individuals with potential DLB. Classification of evidence: This study provided Class III evidence that CSF αSyn-SAA distinguishes patients with clinically diagnosed DLB from normal controls.

The heterogeneity of the whole-exome sequencing (WES) data generation methods present a challenge to a joint analysis. Here we present a bioinformatics strategy for joint-calling 20,504 WES samples collected across nine studies and sequenced using ten capture kits in fourteen sequencing centers in the Alzheimer’s Disease Sequencing Project. The joint-genotype called variant-called format (VCF) file contains only positions within the union of capture kits. The VCF was then processed specifically to account for the batch effects arising from the use of different capture kits from different studies. We identified 8.2 million autosomal variants. 96.82% of the variants are high-quality, and are located in 28,579 Ensembl transcripts. 41% of the variants are intronic and 1.8% of the variants are with CADD > 30, indicating they are of high predicted pathogenicity. Here we show our new strategy can generate high-quality data from processing these diversely generated WES samples. The improved ability to combine data sequenced in different batches benefits the whole genomics research community.

Background Neuropathological hallmarks of AD, amyloid plaques and neurofibrillary tangles, have been described in dementia with Lewy bodies (DLB). Less is known about changes in pre‐mortem cerebrospinal fluid (CSF) amyloid and tau biomarker status in DLB. CSF biomarkers for amyloid (A), tau (T), and neurodegeneration (N) can be utilized to define pre‐mortem pathological status. The overall aim of this investigation was to utilize the ATN research framework to characterize pre‐mortem AD‐related pathology longitudinally in a rigorously characterized cohort of DLB from the Dementia with Lewy Bodies Consortium (DLBC) study. Our hypothesis was that ATN biomarkers would identify DLB with coexistent AD and ATN status would remain the same after 12 months. Method Cerebrospinal fluid (CSF) samples from an AD/ADRD cohort were obtained from the Cleveland Alzheimer’s Disease Research Center (CADRC), Cleveland Clinic Lou Ruvo Center for Brain Health Aging and Neurodegenerative Disease Biobank (CBH‐biobank) and DLBC. CSF Aβ 40 , Aβ 42 , Aβ 42/40 (A), p‐Tau181 (T), and t‐Tau (N), were evaluated. The cohort was subtyped by CSF ATN category and sub‐group of DLB cases were neuropathologically evaluated. A sub‐cohort of DLB cases had longitudinal data with an evaluation at baseline and 12 months later. Result DLB patients who were positive for both CSF Aβ 42/40 (A+) and p‐Tau181 (T+) remained positive from baseline to 12 months. In contrast, some DLB patients who were positive for CSF Aβ 42/40 (A+), but negative for p‐Tau181 (T‐), changed from CSF Aβ 42/40 positive to negative at 12 months (p‐value = 0.0153) in the A+T‐N‐ and A+T‐N+ group. Findings from a small DLB subgroup with post‐mortem neuropathologic analyses indicated that most of the cases that were p‐Tau181 positive (T+) also had high Braak stage. Conclusion These findings suggest that DLB patients who are positive for CSF Aβ 42/40 (A+), but negative for p‐Tau181 (T‐), may have CSF Aβ 42/40 levels that can fluctuate over time. In the small DLB subgroup with post‐mortem neuropathologic analysis, there appeared to be a relationship between AD pathologic change and ATN findings that warrants further study in a larger autopsied cohort to fully understand the relationship between ATN status and underlying AD neuropathologic change in DLB.

Background Dementia with Lewy Bodies (DLB) is defined by abnormal deposits of alpha‐synuclein (Lewy bodies), in the brain. DLB may be accompanied by variable co‐pathology, especially Alzheimer’s Disease (AD). DLB is diagnosed clinically, with definitive diagnosis only possible at autopsy. Misdiagnosis may occur due to overlapping symptoms in different dementias. The alpha‐synuclein (αSyn) seed amplification assay (SAA) is a qualitative test that detects misfolded aggregates of αSyn in synucleinopathies, including DLB, with high sensitivity and specificity. In autopsy‐confirmed AD with limbic or neocortical synuclein pathology, SAA was positive even when DLB phenotypic features were mild (Arnold et al., 2022). We are now evaluating SAA in cerebrospinal fluid (CSF) samples and clinical features in patients with DLB followed through a multi‐Center DLB Consortium (NINDS U01NS100610), which is part of the NINDS Parkinson’s Disease Biomarkers Program (PDBP). Method CSF from 3 groups: 1) analytical controls: youngest available age, no neurological symptoms, normal neurological exam; 2) clinical controls: no symptoms/signs, age and gender match to DLB; 3) DLB. SAA will be run blind to diagnosis using standardized assay methods by Amprion. We will compare Core clinical features of DLB patients and SAA. We hypothesize that SAA‐ patients will have less clear clinical features for DLB, and will be likely to meet clinical and/or CSF biomarker criteria for other diagnoses, e.g., AD. Core clinical features will be defined as strongly or weakly present, e.g., REM Sleep Behavior Disorder: PSG‐proven (strong); +questionnaire (weak); Parkinsonism: ≥ 2 cardinal features (strong); 1 feature (weak). Hallucinations and fluctuations are described in standardized rating scales/interviews. CSF AD biomarkers (Aβ42, t‐Tau and p‐Tau181) were analyzed for almost all DLB subjects. Sensitivity and specificity of each Core clinical feature will be calculated in relation to SAA+ and SAA‐ results. Result Subjects were recruited at 8 clinical sites across the USA from 2017‐2021. Demographics of subjects with available CSF are shown in Table 1. Demographics, clinical features and CSF AD biomarker results will be presented in SAA+ and SAA‐ groups . Conclusion Results will inform how well different core clinical features predict pathological αSyn (CSF SAA+) in people diagnosed with DLB.

Background Hippocampal and amygdala subfields variably affect cognitive impairment in neurodegenerative diseases. Subfields of these regions can be well segmented using modern neuroimaging tools but their role in neurodegenerative disease is under active investigation. In this study, we identified hippocampal and amygdala subregions predictive of cognitive performance and motor symptoms severity measured by MoCA (Montreal Cognitive Assessment) and MDS‐UPDRS III, respectively, in patients with dementia with Lewy Bodies (DLB). Method We selected all participants with probable DLB (N = 48, mean age = 71±7 years, 15% female) enrolled in the Dementia with Lewy Bodies Consortium (DLBC) as of July 2022, with concurrent measures of 3D T1 MRI sequence, MoCA, and MDS‐UPDRS III scores. We performed cortical reconstruction and volumetric segmentation of hippocampal subfields and nuclei of the amygdala using FreeSurfer (v 7.2). We used combat harmonization to account for site and scanner differences. We trained and applied a bootstrapped, bidirectional stepwise regression model of 29 predictor variables comprised of sub‐fields and mean cortical thickness against MoCA and MDS‐UPDRS III, respectively, with an 80‐20 train‐test split ratio, and 5000 repetitions, corrected for age and sex. Result Subfield segmentation is shown in Figure 1A. The best fitting model for MoCA included mean cortical thickness, parasubiculum, hippocampal and amygdala transition area, corticoamygdaloid transition area, and CA3 body (Figure 1B, adjusted R² = 0.51). The best fitting model for MDS‐UPDRS III included the cortical nucleus of the amygdala and CA1 body (Figure 1C, adjusted R² = 0.22). This model was considered a poor fit. We considered MoCA for further analysis and closely predicted scores in our 20% partitioned test sample (Figure 2A, R² = 0.38). Conclusion We report model‐based selection of hippocampal and amygdala subfields to predict MoCA scores in DLB. Atrophy in these regions has been associated with global cognitive deficit in mild cognitive impairment and Alzheimer disease cohorts. The model fit for MDS‐UPDRS III scores was poor, providing evidence that these brain regions do not serve a role in motor control.

Background Dementia with Lewy bodies (DLB) and Alzheimer’s disease (AD) are challenging to differentiate. Mobile health devices that measure motor and sleep domains between clinic visits may improve differential diagnosis. Method Participants with probable AD and probable DLB wore ActiGraph triaxial wGT3X‐BT actigraphy devices on their nondominant wrists for 2 weeks and multichannel Sleep ProfilerX8 TM EEG devices on their foreheads for 2 consecutive nights. ³ Actigraphy data were processed and scored using GGIR and activity counts in R. The Sleep Profiler characterizes abnormal slow‐wave sleep and non‐REM hypertonia (NRH). Result At baseline, participants with DLB (n = 9) had higher UPDRS scores and a higher likelihood of psychosis than participants with AD (n = 8; see Table 1). Using univariate Wilcoxon Rank‐Sum tests, we found that at baseline participants with DLB had lower daily activity counts, higher % time in NRH, and lower % time in REM sleep than participants with AD (unadjusted p‐values < 0.05; See Figure 1). In a recent 6‐month followup, activity levels were lower on average than baseline activity levels for both DLB (n = 5) and AD (n = 5). at 6‐months, the differences in % sleep in NRH increased between DLB participants (n = 5) and AD participants (n = 3) as average % NRH increase 6% in DLB and decreased 1% in AD (see Table 2). These 6‐month differences are not statistically significant, but data collection are ongoing. An exploration of methods for processing and scoring actigraphy data revealed that processing can result in wide range of findings. As a result, we have focused on actigraphy outcomes which are more suitable to older participants with dementia. Conclusion Participants with DLB demonstrated lower baseline physical activity and higher rates of baseline sleep disturbances than participants with AD; sleep differences between DLB and AD appeared to increase over 6 months. Actigraphy and sleep‐monitoring devices are well tolerated in participants with dementia and they can reliably collect valid, useful activity and sleep data in between clinic visits.

Background Dementia with Lewy bodies (DLB) and Alzheimer’s disease (AD) are challenging to differentiate. Mobile health devices that measure motor and sleep domains between clinic visits may improve differential diagnosis. Method Participants with probable AD and probable DLB wore ActiGraph triaxial wGT3X‐BT actigraphy devices on their nondominant wrists for 2 weeks and multichannel Sleep ProfilerX8TM EEG devices on their foreheads for 2 consecutive nights.3 Actigraphy data were processed and scored using GGIR and activity counts in R. The Sleep Profiler characterizes abnormal slow‐wave sleep and non‐REM hypertonia (NRH). Result At baseline, participants with DLB (n = 9) had higher UPDRS scores and a higher likelihood of psychosis than participants with AD (n = 8; see Table 1). Using univariate Wilcoxon Rank‐Sum tests, we found that at baseline participants with DLB had lower daily activity counts, higher % time in NRH, and lower % time in REM sleep than participants with AD (unadjusted p‐values < 0.05; See Figure 1). In a recent 6‐month followup, activity levels were lower on average than baseline activity levels for both DLB (n = 5) and AD (n = 5). at 6‐months, the differences in % sleep in NRH increased between DLB participants (n = 5) and AD participants (n = 3) as average % NRH increase 6% in DLB and decreased 1% in AD (see Table 2). These 6‐month differences are not statistically significant, but data collection are ongoing. An exploration of methods for processing and scoring actigraphy data revealed that processing can result in wide range of findings. As a result, we have focused on actigraphy outcomes which are more suitable to older participants with dementia. Conclusion Participants with DLB demonstrated lower baseline physical activity and higher rates of baseline sleep disturbances than participants with AD; sleep differences between DLB and AD appeared to increase over 6 months. Actigraphy and sleep‐monitoring devices are well tolerated in participants with dementia and they can reliably collect valid, useful activity and sleep data in between clinic visits.

Background The application of machine learning (ML) tools in electronic health records (EHRs) can help reduce the underdiagnosis of dementia, but models that are not designed to reflect minority population may perpetuate that underdiagnosis. Method To address the underdiagnosis of dementia in both Black Americans (BAs) and white Americans (WAs), we sought to develop and validate ML models that assign race‐specific risk scores. These scores were used to identify undiagnosed dementia in BA and WA Veterans in EHRs. More specifically, risk scores were generated separately for BAs (n = 10K) and WAs (n = 10K) in training samples of cases and controls by performing ML, equivalence mapping, topic modeling, and a support vector‐machine (SVM) in structured and unstructured EHR data. Scores were validated via blinded manual chart reviews (n = 1.2K) of controls from a separate sample (n = 20K). AUCs and negative and positive predictive values (NPVs and PPVs) were calculated to evaluate the models. Result Among Veterans with undiagnosed dementia who underwent chart review, those diagnosed by expert clinician reviewers with possible/probable dementia had higher ADRD ML risk scores compared to those diagnosed with no dementia (0.45 [0.38] vs. ‐0.02 [0.51] for BAs, and 0.38 [0.41] vs. ‐0.02 [0.47] for WAs; (see figure 1). Of the 1,200 Veterans who underwent chart review, 15.3% (n = 92) of BAs and 9.5% (n = 57) of WAs were identified with possible/probable dementia. However, adjusting for oversampling of Veterans with higher scores, the adjusted estimated prevalence was 4.1% for BA Veterans and 3.6% for WA Veterans. There was a strong positive relationship between risk scores and the prevalence of undiagnosed dementia (Figure 2). As anticipated, for Veterans with scores below the 90 th percentile, the percentages of undiagnosed dementia were low: 3.9% for BA and 2.9% for WA Veterans. Among Veterans with scores above the 90 th percentile, we found that a higher percentage of BA Veterans had undiagnosed dementia than WA Veterans: 25.6% vs. 15.3%. Conclusion Our findings suggest that race‐specific ML models can assist in the identification of undiagnosed dementia, particularly in BAs. Future studies should investigate implementing EHR‐based risk scores in clinics that serve both BA and WA Veterans.

Introduction In this pilot study, we explored sleep biomarker risk probabilities for different neurodegenerative disorder (NDD) phenotypes across a spectrum of NDD patients, compared with controls. Methods We analyzed a cohort of patients with different NDD phenotypes who underwent in-home recordings with Sleep Profiler, including Lewy body disease (LBD=20), Alzheimer’s disease dementia (AD=29), and isolated REM sleep behavior disorder (iRBD=19). Controls with an MMSE>28 (CG=61) and patients with Parkinson disease (PD=16) and mild-cognitive-impairment (MCI=41) also participated. We developed a machine-learning classifier that assigned NDD probabilities for LBD, AD, iRBD, and CG. The input variables included: time-REM, non-REM hypertonia, autonomic-activation index, spindle-duration, atypical-N3, time-supine, sleep-efficiency, relative-theta, and theta/alpha. Probabilities >50% were assigned “likely”, and for CG>=50% and a NDD group probability 20-50%, the assignment was normal “plus”. Probability assignments were then made for the NDD and CG groups, then further applied to the PD and MCI patient groups. Results The CG group participants were assigned Normal-Likely=74%, Normal+AD=11%, Normal+iRBD=5%, iRBD-Likely=5%, and AD-Likely=5%. LBD patient distributions were LBD-Likely=70%, iRBD-Likely=5%, AD-Likely=5%, Normal+LBD=5%, Normal+iRBD=5%, Normal+AD=5%, and Normal-Likely=5%. AD group distributions were AD-Likely=71%, LBD-Likely=4%, Normal+AD=14%, Normal+iRBD=4%, Normal-Likely=7%. iRBD patients were characterized with iRBD-Likely=37%, LBD-Likely=5%, AD-Likely=5%, Normal+iRBD=27%, Normal-Likely=26%. PD patients were assigned iRBD-Likely=29%, LBD-likely=14%, AD-Likely=14%, Normal+iRBD=21%, Normal+AD=7%, Normal-Likely=14%. MCI distributions were AD-Likely=45%, LBD-Likely=8%, iRBD-Likely=5%, Normal+AD=21%, Normal-Likely=13%. Conclusions For LBD, AD and CG groups, correct risk assignments were >70% while gross misclassifications were <10%. Classification patterns for PD, MCI and iRBD were disbursed in a manner consistent with the range of severities expected in each group.

Introduction In 2015, Lee et al. observed sleeping position influenced the efficiency of glymphatic clearance in rats. In 2018, Levendowski et al. reported supine sleep was independently associated with neurodegenerative disorders (NDD)(odds ratio=3.7). This update further evaluates the comparative frequency of the supine sleep position across a spectrum of cognitive impairment, including patients diagnosed with a Parkinsonian spectrum disorder (PSD), Alzheimer’s disease dementia (AD) and mild cognitive impairment (MCI), compared with a control group (CG) of patients without a known cognitive disorder. Methods After ethics review and with informed consent, a control group (CG: n=170), PSD (n=36), AD (n=29) and MCI group (n=41) were studied in-home with the Sleep Profiler; 89% across two-nights. Between-group comparisons were assessed with Mann-Whitney U and Chi-square tests. Results The hours of supine sleep were greater in PSD=3.0 + 1.9, AD=3.2 + 2.0 and MCI=3.0 + 2.0 vs. CG=1.9 + 1.8 (all p<0.002). The proportion of patients with >2 hours of supine sleep were PSD=69%, AD=69% and MCI=59% vs. CG=39%, with odds ratios of PSD=3.6, AD=3.5 and MCI=2.2 (all p<0.001). Conclusions Here we provide further evidence for a relatively strong association between supine sleep and neurodegenerative disease including PSD, AD and MCI cohorts. The nature of this association remains unclear from cross sectional study designs. Future prospective studies should test if there is a causal link between supine sleep and subsequent development of neurodegeneration.