Donohue, Michael

AbstractBackgroundMixed model repeated measures (MMRM) is the most common analysis approach in clinical trials for Alzheimer’s disease. The model treats time as a categorical variable, which can be problematic when assessments occur off‐schedule, as including off‐schedule visits can induce bias, and excluding them ignores valuable information and violates the intention to treat principle. This problem has been exacerbated by clinical trial visits which have been delayed due to the COVID19 pandemic. Since MMRM entails a mean parameter for each group and visit, the model can be inefficient for longer trials of Preclinical Alzheimer’s. As an alternative to MMRM, we propose a constrained longitudinal data analysis with natural cubic splines that treats time as continuous and uses test version effects to model the mean over time.MethodWe use model selection techniques to compare the predictive value of spline models to slope models, MMRM, and models that assume a proportional treatment effect using data from several completed trials. These trials include the ADCS trial of donepezil in MCI, trials of intranasal insulin and a fyn kinase inhibitor in mild Alzheimer’s dementia, and the A4 trial of solanezumab in Preclinical Alzheimer’s. Simulation studies assess the power and Type I error of these models for trials in Preclinical Alzheimer’s, MCI, and mild dementia.ResultThe spline model with ADAS‐Cog version effect attained the best predictive value for the trials in mild dementia. The proportional model was preferred for the trial in MCI, but the proportional effect assumption was violated, and the treatment effect estimate was in the wrong direction. Considering that the spline with two degrees of freedom and version effect provide fits that are similar to MMRM in Figure 1, we concluded that two degrees of freedom appear to be sufficient to capture enough of the general trend to assess the treatment effect. AIC results from the A4 Study will be considered and included after the study is completed in Spring 2023.ConclusionThe spline model is shown to have better predictive value and precision in real datasets and has better power and Type I error control in a variety of simulation scenarios.

ImportanceIncreased white matter hyperintensity (WMH) volume is a common magnetic resonance imaging (MRI) finding in both autosomal dominant Alzheimer disease (ADAD) and late-onset Alzheimer disease (LOAD), but it remains unclear whether increased WMH along the AD continuum is reflective of AD-intrinsic processes or secondary to elevated systemic vascular risk factors.ObjectiveTo estimate the associations of neurodegeneration and parenchymal and vessel amyloidosis with WMH accumulation and investigate whether systemic vascular risk is associated with WMH beyond these AD-intrinsic processes.Design, Setting, and ParticipantsThis cohort study used data from 3 longitudinal cohort studies conducted in tertiary and community-based medical centers—the Dominantly Inherited Alzheimer Network (DIAN; February 2010 to March 2020), the Alzheimer’s Disease Neuroimaging Initiative (ADNI; July 2007 to September 2021), and the Harvard Aging Brain Study (HABS; September 2010 to December 2019).Main Outcome and MeasuresThe main outcomes were the independent associations of neurodegeneration (decreases in gray matter volume), parenchymal amyloidosis (assessed by amyloid positron emission tomography), and vessel amyloidosis (evidenced by cerebral microbleeds [CMBs]) with cross-sectional and longitudinal WMH.ResultsData from 3960 MRI sessions among 1141 participants were included: 252 pathogenic variant carriers from DIAN (mean [SD] age, 38.4 [11.2] years; 137 [54%] female), 571 older adults from ADNI (mean [SD] age, 72.8 [7.3] years; 274 [48%] female), and 318 older adults from HABS (mean [SD] age, 72.4 [7.6] years; 194 [61%] female). Longitudinal increases in WMH volume were greater in individuals with CMBs compared with those without (DIAN: t = 3.2 [P = .001]; ADNI: t = 2.7 [P = .008]), associated with longitudinal decreases in gray matter volume (DIAN: t = −3.1 [P = .002]; ADNI: t = −5.6 [P < .001]; HABS: t = −2.2 [P = .03]), greater in older individuals (DIAN: t = 6.8 [P < .001]; ADNI: t = 9.1 [P < .001]; HABS: t = 5.4 [P < .001]), and not associated with systemic vascular risk (DIAN: t = 0.7 [P = .40]; ADNI: t = 0.6 [P = .50]; HABS: t = 1.8 [P = .06]) in individuals with ADAD and LOAD after accounting for age, gray matter volume, CMB presence, and amyloid burden. In older adults without CMBs at baseline, greater WMH volume was associated with CMB development during longitudinal follow-up (Cox proportional hazards regression model hazard ratio, 2.63; 95% CI, 1.72-4.03; P < .001).Conclusions and RelevanceThe findings suggest that increased WMH volume in AD is associated with neurodegeneration and parenchymal and vessel amyloidosis but not with elevated systemic vascular risk. Additionally, increased WMH volume may represent an early sign of vessel amyloidosis preceding the emergence of CMBs.

Accurate prediction of progression in subjects at risk of Alzheimer's disease is crucial for enrolling the right subjects in clinical trials. However, a prospective comparison of state-of-the-art algorithms for predicting disease onset and progression is currently lacking. We present the findings of "The Alzheimer's Disease Prediction Of Longitudinal Evolution" (TADPOLE) Challenge, which compared the performance of 92 algorithms from 33 international teams at predicting the future trajectory of 219 individuals at risk of Alzheimer's disease. Challenge participants were required to make a prediction, for each month of a 5-year future time period, of three key outcomes: clinical diagnosis, Alzheimer's Disease Assessment Scale Cognitive Subdomain (ADAS-Cog13), and total volume of the ventricles. The methods used by challenge participants included multivariate linear regression, machine learning methods such as support vector machines and deep neural networks, as well as disease progression models. No single submission was best at predicting all three outcomes. For clinical diagnosis and ventricle volume prediction, the best algorithms strongly outperform simple baselines in predictive ability. However, for ADAS-Cog13 no single submitted prediction method was significantly better than random guesswork. Two ensemble methods based on taking the mean and median over all predictions, obtained top scores on almost all tasks. Better than average performance at diagnosis prediction was generally associated with the additional inclusion of features from cerebrospinal fluid (CSF) samples and diffusion tensor imaging (DTI). On the other hand, better performance at ventricle volume prediction was associated with inclusion of summary statistics, such as the slope or maxima/minima of patient-specific biomarkers. On a limited, cross-sectional subset of the data emulating clinical trials, performance of the best algorithms at predicting clinical diagnosis decreased only slightly (2 percentage points) compared to the full longitudinal dataset. The submission system remains open via the website https://tadpole.grand-challenge.org, while TADPOLE SHARE (https://tadpole-share.github.io/) collates code for submissions. TADPOLE's unique results suggest that current prediction algorithms provide sufficient accuracy to exploit biomarkers related to clinical diagnosis and ventricle volume, for cohort refinement in clinical trials for Alzheimer's disease. However, results call into question the usage of cognitive test scores for patient selection and as a primary endpoint in clinical trials.

AbstractIntroductionLumbar puncture (LP) to collect and examine cerebrospinal fluid (CSF) is an important option for the evaluation of Alzheimer's disease (AD) biomarkers but it is not routinely performed due to its invasiveness and link to adverse effects (AE).MethodsWe include all participants who received at least one LP in the Alzheimer's Disease Neuroimaging Initiative (ADNI) Study. For comparison between groups, two‐sample t‐tests for continuous, and Pearson's chi‐square test for categorical variables were performed.ResultsTwo hundred twenty‐seven LP‐related AEs were reported by 172 participants after 1702 LPs (13.3%). The mean age of participants who reported at least one AE was 69.79 (standard deviation (SD) 6.3) versus none 72.44 (7.17) years (p < 0.001) with female predominance (115/172 = 67.4% vs 435/913 = 48%), and had greater entorhinal cortical thickness and hippocampal volume (3.903 (0.782) vs 3.684 (0.775) mm, p = 0.002; 7.38 (1.06) vs 7.05 (1.15) mm3, p < 0.001), respectively.DiscussionWe found that younger age, female sex, and greater thickness of the entorhinal cortex were associated with a higher rate of LP‐related AE reports.

Abstract Mixed model repeated measures (MMRM) is the most common analysis approach used in clinical trials for Alzheimer's disease and other progressive diseases measured with continuous outcomes over time. The model treats time as a categorical variable, which allows an unconstrained estimate of the mean for each study visit in each randomized group. Categorizing time in this way can be problematic when assessments occur off‐schedule, as including off‐schedule visits can induce bias, and excluding them ignores valuable information and violates the intention to treat principle. This problem has been exacerbated by clinical trial visits which have been delayed due to the COVID19 pandemic. As an alternative to MMRM, we propose a constrained longitudinal data analysis with natural cubic splines that treats time as continuous and uses test version effects to model the mean over time. Compared to categorical‐time models like MMRM and models that assume a proportional treatment effect, the spline model is shown to be more parsimonious and precise in real clinical trial datasets, and has better power and Type I error in a variety of simulation scenarios.

AbstractBackgroundWe developed the Institute on Methods and Protocols for Advancement of Clinical Trials in ADRD (IMPACT‐AD) to address the shortage of qualified investigators with adequate training and expertise to conduct Alzheimer’s Disease (AD) and AD Related Dementias (ADRD) clinical trials.MethodWe describe the experience of the first two years of funding of the IMPACT‐AD course, specifically the rates of applications to the course, descriptive summaries of the accepted trainees, and the measures of participant satisfaction and learning.ResultOver the first two years of the course, we received 195 eligible applications to the course. Most (n = 121, 62%) applicants were female and nearly half self‐reported as being African American/Black, Asian, or Pacific Islander race and/or Hispanic/Latino ethnicity. Forty‐one applicants (21%) indicated that they were the first in their family to attend collegeOf 69 total trainees accepted to the course (36% acceptance rate), a majority (n = 44, 64%) were female. Twenty‐eight (41%) self‐reported as being African American/Black, Asian, or Pacific Islander race and eleven (16%) were of Hispanic ethnicity. Fourteen trainees (20%) identified as being the first person in their family to attend college. Among trainees, 24 (35%) were MDs, 30 (43%) were PhDs, and 9 (13%) held master’s degrees.Course lecture topics were rated, on average, as “essential” or “valuable” by 98% of trainees. Across lecture topics, on average, 48% of trainees rated their prior knowledge of topics as “very strong” or “strong,” and 11% of trainees rated their prior knowledge as “weak.” Overall, 52% of trainees evaluated their change in knowledge based on the lectures as “very much increased,” 46% as “somewhat” or “slightly” increased, and 2% as “no change.” Results were similar for the two tracks. Based on pre‐ and post‐course assessments, trainees improved their knowledge of covered content (from 62% to 79% correct responses on knowledge assessments).ConclusionThe IMPACT‐AD course is achieving its mission of educating and diversifying the next generation of ADRD trialists.We thank Dr. Laurie Ryan and Dr. Kristina McLinden from the National Institute on Aging for their key contributions in leadership and scientific guidance of IMPACT‐AD.

AbstractThe poor generalizability of clinical research data due to the enrollment of highly educated, non‐Latinx White participants hampers the development of therapies for Alzheimer's disease (AD). Black and Latinx older adults have a greater risk for dementia, yet it is unclear how health‐care disparities and sociocultural factors influence potential AD therapies and prognosis. Low enrollment of under‐represented populations may be attributable to several factors including greater exclusion due to higher rates of comorbidities, lower access to AD clinics, and the legacy of unethical treatment in medical research. This perspective outlines solutions tested in the Brain Health Registry (BHR) and the Alzheimer's Disease Neuroimaging Initiative (ADNI), including culturally‐informed digital research methods, community‐engaged research strategies, leadership from under‐represented communities, and the reduction of exclusion criteria based on comorbidities. Our successes demonstrate that it is possible to increase the inclusion and engagement of under‐represented populations into US‐based clinical studies, thereby increasing the generalizability of their results.

AbstractIntroductionAn analysis of the ethnocultural and socioeconomic composition of Alzheimer's Disease Neuroimaging Initiative (ADNI) participants is needed to assess the generalizability of ADNI data to diverse populations.MethodsADNI data collected between October 2004 and November 2020 were used to determine ethnocultural and educational composition of the sample and differences in the following metrics: screening, screen fails, enrollment, biomarkers.ResultsOf 3739 screened individuals, 11% identified as being from ethnoculturally underrepresented populations (e.g., Black, Latinx) and 16% had <12 years of education. Of 2286 enrolled participants, 11% identified as ethnoculturally underrepresented individuals and 15% had <12 years of education. This participation is considerably lower than US Census data for adults 60+ (ethnoculturally underrepresented populations: 25%; <12 years of education: 4%). Individuals with <12 years of education failed screening at a higher rate.DiscussionOur findings suggest that ADNI results may not be entirely generalizable to ethnoculturally diverse and low education populations.

AbstractBackgroundThe overarching goal of the Longitudinal Early‐onset Alzheimer Disease study (LEADS) is to optimally characterize early‐onset AD (EOAD) and establish an EOAD clinical trials network. Here we report the baseline demographic and imaging biomarker comparisons of the LEADS cohort to late‐onset AD (LOAD) subjects from the Alzheimer’s Disease Neuroimaging Initiative (ADNI3).Method123 amyloid‐positive EOAD, 47 amyloid‐negative EOnonAD, 60 cognitively normal young controls were compared to 130 amyloid‐positive LOAD, 110 amyloid‐negative LOnonAD and 286 amyloid‐negative cognitively normal older controls. To account for the effect of cognitive aging between EO and LO populations, each cognitive measure was Z‐transformed. Cortical and hippocampal atrophy were quantified using W‐scores adjusted for age, sex and total intracranial volume. Z‐scores and W‐scores were compared using t‐test or Wilcoxson rank test as appropriate. All p‐values were corrected for multiple comparisons using the false discovery rate correction.ResultEOAD showed greater pathology burden and greater cortical atrophy (AD signature) relative to LOAD. EOAD also showed greater cognitive impairment across all cognitive tests. EOAD showed greater functional impairment, more depression but less neuropsychiatric behaviors overall compared to LOAD (Table 1 and Figure 1, all ps<0.05). Repeating the analyses stratified by cognitive stage (MCI/dementia) or CDR global rating (0.5/1) did not result in any major differences.EOnonAD differed from LOnonAD by also showing greater impairment on all cognitive and functional measures There were no significant differences in amyloid and tau burden, or atrophy W‐scores between these groups. EOnonAD were more depressed and showed more functional impairment compared to LOnonAD (Table 2 and Figure 2, all ps<0.05). Repeating the analyses split by cognitive stage (MCI/dementia) or CDR global rating (0.5/1) did not result in any major differences.ConclusionConsistent with our preexisting hypotheses, EOAD and EOnonAD perform much worse relative to their LO counterparts. EOAD also show greater pathological burden as expected. The reported analyses were done in chronological rather than disease time (time since disease onset). Benchmarking individuals along the disease spectrum might prove to be a better strategy especially when conducting analyses on rate of disease progression.

AbstractBackgroundAn analysis of the ethnoracial and educational composition of Alzheimer's Disease Neuroimaging Initiative (ADNI) participants, and the relationship between ethnoracial/education and screening, enrollment, dropout, and biomarkers is needed to assess the generalizability of ADNI data to diverse populations, and to inform efforts to increase diversity.MethodData from all 4 ADNI phases were used to determine ethnoracial and educational breakdown and differences of the following ADNI participation metrics: screening, screen fails, enrollment, as well as drop‐out within and between ADNI phases. Multivariable logistic regression was used to analyze the association between ethnoracial and educational group and either amyloid positivity or ApoE e4 status, adjusting for age, gender, and diagnostic group.ResultAcross the 4 ADNI phases (screened: n=3,739; enrolled: n=2,286), 11% of screened and enrolled participants identified as Latinx, non‐Latinx Black, or non‐Latinx‐Asian and 16% of the screened and 15% of the enrolled participants reported having =<12 years of education. In contrast, the 2019 US Census American Community Survey for adults 60+ reports 25% non‐White, and 44 % =< 12 years of education. There were no observed differences between participants from historically underrepresented ethnoracial groups and NLW participants in screen fail and dropout rates. We found that individuals with =<12 years of education failed screening at a significantly higher rate (p=.01) and rolled‐over significantly less to subsequent ADNI phases (p<.001) compared to individuals with an education >12 years. Identifying as non‐Latinx Asian was associated with lower odds of being amyloid positive (OR=0.36; p=.014) and an ApoE4 carrier (OR=0.36; p=.008) compared to NLW. Identifying as Latinx was associated with being less likely to be amyloid positive than NLW participants (OR=0.54; p=.031).ConclusionCompared to the US Census, ADNI underrepresents enrollment of Latinx, non‐Latinx Black, and non‐Latinx Asian participants and those with low educational attainment. Hence, ADNI reflects the on‐going challenges with recruiting and enrolling underrepresented populations into most AD research. Our findings highlight the need for ADNI to increase enrollment and engagement of underrepresented ethnoracial and educational populations. ADNI is actively undertaking steps to implement improved recruitment approaches aimed at increasing ethnoracial and socioeconomical representation in multi‐center cohort studies.