Tag: nevin manimala

Alzheimers Dement. 2024 Dec;20 Suppl 6:e089155. doi: 10.1002/alz.089155.

ABSTRACT

BACKGROUND: CT1812 is an experimental therapeutic sigma-2 receptor modulator in development for Alzheimer’s disease (AD) and dementia with Lewy bodies. CT1812 reduces the affinity of Aβ oligomers to bind to neurons and exert synaptotoxic effects. This phase 2, multi-center, international, randomized, double-blind, placebo-controlled trial assessed safety, tolerability and effects of CT1812 on cognitive function in individuals with AD.

METHOD: Consenting participants with confirmed brain amyloid and baseline MMSE scores of 18 to 26 were randomized equally to receive oral placebo, 100 or 300mg of CT1812 daily for 6 months. Exploratory efficacy outcomes included ADAS-Cog11 and 13, Neurophysiological Test Battery, ADCS-Clinical Global Impression of Change and ADCS-Activities of Daily Living. The primary statistical analysis is treated (combined 100 and 300mg CT1812) versus placebo change from baseline in ADAS-Cog11. Canonical AD biomarkers were assessed in CSF and plasma.

RESULT: One hundred fifty-three participants were randomized. An interim analysis of the first 24 participants was performed where baseline demographics included mean age of 71.7 years, 15 females and a mean MMSE of 21.1. In the interim analysis there were twenty-two TEAEs (7, 8, 7 in the 100, 300mg CT1812 and placebo groups, respectively) reported and in the CT1812 groups, all TEAEs were mild or moderate and there were no treatment related SAEs. There was a trend within these 24 participants for slower progression at six months in the CT1812 treated group with ADAS-Cog11 improved 3.1 points relative to placebo according to the prespecified statistical model. Complete data from all participants including AEs, cognitive performance and canonical biomarkers will be presented with the ADAS-Cog11 change from baseline relative to placebo for the pooled CT1812 treated group prespecified as the first cognitive assessment with ApoE4 status as a covariate.

CONCLUSION: CT1812 is an oral sigma-2 modulator that displaces Aβ oligomers from neurons. Encouraging trends have been observed in previous studies (NCT03493282, NCT03522129, NCT04735536) and in an interim analysis of this phase 2 trial (NCT03507790). This trial will provide proof of concept data indicating if CT1812 can slow progression in people with mild to moderate AD. This study was supported by a grant from the NIH (AG058660).

PMID:39782655 | DOI:10.1002/alz.089155

Alzheimers Dement. 2024 Dec;20 Suppl 6:e088870. doi: 10.1002/alz.088870.

ABSTRACT

BACKGROUND: Clinical trial sponsors rely on research sites to identify and enroll appropriate study participants and to correctly and reliably assess symptom severity and function over the course of the trial. Low-recruiting sites represent a large financial and operational burden and may negatively impact trial success either by selecting inappropriate participants and/or high prevalence of data quality issues. We previously reported that >60% of sites in schizophrenia clinical trials recruited ≤5 participants. Here we analyze 3 large dementia trials to assess the proportion of low-recruiting sites and compare their data quality with the remaining sites.

METHOD: Data were obtained from 3 large early dementia clinical trials totaling 834 sites. Sites were divided into two groups based on the number of randomized subjects: 1. sites with more than 5 randomized subjects (high-recruiting sites) and 2. sites with 5 or less randomized subjects (low-recruiting sites). Data quality issues were defined as administration and scoring errors on relevant scales. These errors were summed per site and then compared by site size using Poisson regression with the site size, study and their interaction as predictors and the number of possible hits as exposure.

RESULT: 71 (8.5%) sites did not randomize a subject, 43 (5.2%) sites randomized one subject, and 377 (45.2%) sites randomized ≤5 subjects. Overall, administration and scoring errors were more frequent at low-recruiting sites (seen at 41.9% visits) compared to 35.3% at the high-recruiting sites. 2 studies show significantly higher IRRs, 1.59 (1.46, 1.74) and 1.2 (1.15, 1.25) respectively, while the third study has an IRR close to 1 but not statistically significant.

CONCLUSION: Our results indicate that low-recruiting sites (arbitrarily defined as randomizing ≤5 participants) are frequent in large dementia trials. These sites pose considerable cost to sponsors, but more importantly, are more likely to provide questionable data quality. While a single site like this only represents a small risk, in aggregate, they can pose a serious challenge. Clinical trial sponsors should therefore consider strategies to minimize the impact of low-recruiting sites on study outcomes.

PMID:39782642 | DOI:10.1002/alz.088870

Alzheimers Dement. 2024 Dec;20 Suppl 6:e089821. doi: 10.1002/alz.089821.

ABSTRACT

BACKGROUND: Only about 50% of the variance in cognitive decline occurring during Alzheimer’s pathogenesis is attributable to standard AD biomarkers (cerebrocortical Aβ, pathological tau, and atrophy) (Tosun et al., Alzheimer’s Dement. 18: 1370, 2022). Other factors contributing to such decline include brain insulin resistance (BIR). Levels of its biomarker, pathological IRS-1 pS616 (insulin receptor substrate-1 phosphorylated at S616), accounts for as much as 47% of the variance in episodic memory scores in an elderly population we studied (Talbot et al., JCI 122: 1316, 2012). To better understand BIR in Alzheimer’s disease (AD), its proximal causes, and if incretin receptor agonists (IRAs) can reduce BIR, we extended our earlier ex vivo work on the hippocampal formation (HF) of non-cognitively impaired (NCI) and AD dementia (ADd) cases to non-amnestic MCI (naMCI) and amnestic MCI (aMCI) cases METHOD: Diverse brain banks donated fresh frozen HF tissue from NCI, naMCI, aMCI, and ADd cases (n = 10 per diagnostic group) who were age- and -sex matched whites with no history of diabetes and low postmortem intervals (m ± SD = 6.75 ± 2.2 h). Using our ex vivo stimulation protocol (Talbot et al., 2012), tissue sections were tested for insulin-induced phosphorylation of insulin signaling and regulating enzymes after 30 min incubations in 0, 1, or 10 nM insulin with or without 30 min preincubation in 100 nM of liraglutide activating GLP-1 (glucagon-like peptide-1) receptors, [D-Ala²]-GIP activating GIP (gastric inhibitory peptide) receptors, or Peptide 19, a dual IRA activating both GLP-1 and GIP receptors. Statistical significance was defined as p < 0.05.

RESULT: HF responsiveness to 1 and 10 nM insulin was significantly reduced in both the MCI and ADd samples. Pretreatment with each of the IRAs tested virtually restored normal insulin responsiveness in both MCI groups even though the proximal cause of insulin resistance in naMCI differed from aMCI and ADd. Only the dual IRA, however, significantly increased insulin responsiveness in ADd samples.

CONCLUSION: IRAs reaching brain neurons can virtually eliminate hippocampal insulin resistance in naMCI and aMCI cases, but significantly reducing such resistance in ADd cases appears to require a dual IRA.

PMID:39782620 | DOI:10.1002/alz.089821

Alzheimers Dement. 2024 Dec;20 Suppl 6:e087728. doi: 10.1002/alz.087728.

ABSTRACT

BACKGROUND: The well-accepted statistical efficacy inference approach for Alzheimer’s disease (AD) clinical trials compares the absolute difference in change from baseline at the last study visit using MMRM (henceforth referred to as MMRM-Last-Visit). Recent AD clinical trials have shown that treatment effects may be manifested prior to 18 months. The objective is to evaluate models estimating an overall treatment effect across all post-baseline visits that may characterize disease modifying effects in contemporary early AD clinical trials.

METHOD: We will evaluate the performance of: (1) MMRM-Last-Visit, (2) MMRM-All-Visits, estimating an overall treatment effect across all post-baseline visits, (3) Linear mixed effects (LME) model with time being continuous, comparing the slopes across treatment groups, (4) Constrained LME, comparing the slopes which constraining the treatment groups to have the same mean baseline, (5) Longitudinal data analysis model with natural cubic spline in time (LDA-NCS), modeling the raw data and comparing the absolute difference in change from baseline at the last visit (LDA-NCS -Last-Visit), (6) Constrained LDA-NCS, similar to LDA-NCS while constraining treatment groups to have the same mean baseline, and (7) proportional/percentage MMRM (pMMRM), which models the percentage reduction across all post-baseline visits. Simulated data mimicking disease modifying effect in contemporary early AD trials were used for the evaluation. Subsequently, we evaluate the performance using bootstrapping techniques.

RESULT: We will present simulation results that mimic a range of assumptions from contemporary early AD trials, including sample size, power, and type I error results.

CONCLUSION: Models estimating an overall treatment effect across all post-baseline visits have shown promise in characterizing disease modifying effects seen in contemporary early AD clinical trials and may help accelerating drug development. This study was funded by Biogen, JO and TC are employees and shareholders of BIogen.

PMID:39782610 | DOI:10.1002/alz.087728

Alzheimers Dement. 2024 Dec;20 Suppl 6:e087587. doi: 10.1002/alz.087587.

ABSTRACT

BACKGROUND: Participant retention is a key determinant for a successful clinical trial. In Alzheimer’s disease (AD) trials, participants are typically required to enroll with a study partner, which adds barriers to retention. Previous analyses of North American trial data found that most study partners were spouses and that such dyads had higher study completion rates than other study partner types. We sought to determine whether these findings could be confirmed in additional trials conducted across broader geographic regions.

METHOD: Our post-hoc analyses used data from two phase II trials of semorinemab, Tauriel (prodromal-to-mild AD; NCT03289143) and Lauriet (mild-to-moderate AD; NCT03828747), we examined the proportion of participants completing the trials by study partner type (Domestic Partner vs. Other) and geographical region (North America and Europe). We fit logistic regression models to assess whether participant retention differed by study partner type in the two trials (separately and combined) and adjusted the models for region, study partner type, participant age, sex, education, APOE4 status, and study partner sex. We considered potential effect modification by region by including an interaction term between study partner type and region. We conducted sensitivity analyses excluding participants who discontinued based on adverse events or death.

RESULT: We observed effect modification by region in the association between study partner type and retention (Tauriel: p = 0.011, Lauriet: p = 0.015; combined: p = 0.003). In Europe, Other study partner types had significantly lower odds of trial completion compared to domestic partners in Tauriel (OR: 0.21; CI: 0.06, 0.74) and in the combined analysis (OR: 0.28; CI: 0.11, 0.65). While not statistically significant, we observed similar trends in Lauriet (OR: 0.36; CI: 0.10, 1.22). No clear differences in retention were observed across study partner types in North America for either trial.

CONCLUSION: Data from the two semorinemab trials suggest that there may be regional differences in the impact of study partner type on retention. Further studies are needed to assess the generalizability of these findings, with an emphasis on cohorts with more representative samples of non-spousal dyads.

PMID:39782599 | DOI:10.1002/alz.087587

Alzheimers Dement. 2024 Dec;20 Suppl 6:e087702. doi: 10.1002/alz.087702.

ABSTRACT

BACKGROUND: The advent of disease-modifying therapies in Alzheimer’s disease (AD) necessitates a nuanced understanding of how therapies impact disease processes. Over the past decades, AD clinical trials have primarily relied on classical statistical analysis methodology such as the mixed model for repeated measures (MMRM) to estimate treatment effects. These conventional treatment effect quantifications are given as group differences in clinical outcome measures at a single visit. While this classical approach of estimating treatment effects is well established, the resulting quantifications have shortcomings in relation to data utilization, meaningfulness, cumulative benefit summarization, post-trial implications, and cross-trial comparability.

METHOD: Properties of conventional treatment effect quantifications from the MMRM were compared with two time-based quantifications from the progression model for repeated measures (PMRM)¹ and a latent time-disease progression model. Results were illustrated and compared using data from the TRAILBLAZER-ALZ 2 trial of donanemab.

RESULT: The MMRM had fewest assumptions, followed by PMRM and then the laten-time quantifications. PMRM and latent-time quantifications utilized information across visits better than the conventional MMRM quantification and produced greater power to detect treatment effects. Compared to conventional quantifications, the time-based quantifications of treatment effects offered several desirable properties in terms of meaningfulness, cumulative benefit summarization, post-trial implications, and cross-trial comparability.

CONCLUSION: Time-based estimates, particularly those derived from PMRM and latent time disease progression models, offer a set of desirable properties that complement conventional treatment effect quantifications. This study advocates for the inclusion of time-based measures in the evaluation of disease modification in AD, providing a more comprehensive and nuanced perspective for guiding future clinical trial methodologies and result interpretations. References: 1. Raket, L. L. (2022). Progression models for repeated measures: Estimating novel treatment effects in progressive diseases. Statistics in Medicine, 41(28), 5537-5557.

PMID:39782576 | DOI:10.1002/alz.087702

Alzheimers Dement. 2024 Dec;20 Suppl 6:e087120. doi: 10.1002/alz.087120.

ABSTRACT

BACKGROUND: Donepezil, an acetylcholinesterase inhibitor (AChEI), is an FDA-approved drug to treat these neurodegenerative diseases, e.g., Alzheimer’s Disease (AD) and Mild Cognitive Impairment (MCI). AChEIs are able to stabilize or slow decline in cognition, function, and behavior. Our objective is to investigate whether Donepezil is able to significantly reduce the rate of hippocampal (Hip) atrophy in neurodegenerative diseases.

METHODS: We followed the PRISMA statement guidelines during the preparation of this systematic review. We searched in MEDLINE (PubMed), CENTRAL (Cochrane Library, November 2023), SCOPUS, and Web of Science and included randomized clinical trials (RCTs) comparing 10 mg donepezil-treated with donepezil-untreated (placebo) and/without control in terms of magnetic resonance imaging (MRI) follow up visits’ results.

RESULTS: A total of four studies out of 174 met our inclusion criteria (599 participants; donepezil = 281, placebo = 318), two of them were ADs and the others were MCIs. 323 participants were female (representing 53.92% of included study population). Follow up between baseline and endpoint results was 12 months. Available outcome data cover reduction of hippocampal atrophy rate in patients with neurodegenerative diseases, but data on several outcome dimensions were either unavailable or not consistently reported across all studies. Results concluded from studies been conducted on MCI patients were statistically insignificant (P > 0.05) annual percentage of change (APC) of Hip volume at 12 months compared to placebo, but studies on AD patients indicated statistically significant APC of Hip volume at 24 and 50 weeks (P < 0.001), but one of these studies also reported no significant difference in neuropsychological performance between treatment groups.

CONCLUSION: The findings of this review suggest that donepezil reducing hippocampal atrophy rate was statistically insignificant for MCI and statistically significant for AD, but its clinical significance is questionable until further investigations. It is also important to note that while the data provided insights into the impact of donepezil, there were limitations, such as incomplete reporting of outcome dimensions in some studies.

PMID:39782575 | DOI:10.1002/alz.087120

Alzheimers Dement. 2024 Dec;20 Suppl 6:e086844. doi: 10.1002/alz.086844.

ABSTRACT

BACKGROUND: Clinical trials should strive to yield results that are clinically meaningful rather than solely relying on statistical significance. However, the determination of clinical meaningfulness of dementia clinical trials lacks standardization and varies based on the trial’s nature. To tackle this issue, a proposed approach involves assessing the time saved before reaching a specific threshold in cognitive status. In this study, we investigated the time saved in cognitive decline among the top responders based on the individual-level treatment responses (ITR) analysis, using data from the Internet-based Conversational Engagement Clinical Trial (I-CONECT; NCT02871921).

METHOD: I-CONECT is a randomized controlled trial to examine the effects of conversational interactions on cognition among socially isolated participants aged ≥ 75 years old. The experiment group engaged in video chats with study staff 4 times/week for 6 months; the control groups received weekly check-in phone calls. We focused on cognitive outcomes that exhibited significant treatment effects at 6-month follow-up: the Montreal Cognitive Assessment (MoCA) for global cognition and Category Fluency Animals (CFA) for semantic fluency. To assess ITR, we employed 300 iterations of 3-fold cross-validated random forest models. We estimated treatment heterogeneity by conducting permutation tests on the area between curves (ABC) statistics derived from the ITR scores. We estimated time saved in cognitive decline as the difference in the number of months required for the top responders (top 25%; 33%) and the remaining participants to reach the same cognitive level at 6-months follow-up.

RESULT: ABC statistics showed substantial heterogeneity in treatment response with MoCA but modest heterogeneity in treatment response with CFA. For global cognition, assuming a treatment effect size between 30-50%, the top 25% and 33% of responders exhibited potential cognitive delays ranging from 4.1 to 10.8 months and 5.9 to 13.9 months, respectively (Figure 1). For semantic fluency, large effect sizes (70-100%) are required to show potential cognitive delays ranging from 0.5 to 6.7 months.

CONCLUSION: Individual differences in the time saved for cognitive decline through the ITR analysis are meaningful outcomes in clinical trials. Future trial outcomes should consider both quantity and quality concepts such as quality-adjusted time saved.

PMID:39782573 | DOI:10.1002/alz.086844

Alzheimers Dement. 2024 Dec;20 Suppl 6:e086330. doi: 10.1002/alz.086330.

ABSTRACT

BACKGROUND: Iron is vital for metabolism but can act as a catalyst for oxidative damage. Elevated brain iron, determined from biomarkers of iron (CSF ferritin and quantitative susceptibility mapping MRI) and from post-mortem measurement of brain iron, has been associated with accelerated cognitive decline in multiple Alzheimer’s disease (AD) clinical, cohorts. These findings supported the hypothesis that treatment with the brain-permeable iron chelator deferiprone may be associated clinical benefit in AD.

METHODS: A multicentre, phase II, double-blind, randomized, placebo controlled clinical trial of deferiprone (15 mg/kg administered orally twice a day) was conducted in people with amyloid-confirmed MCI and mild AD (MMSE≥20) over 12 months. The primary outcome was cognition, assessed at baseline, 6 months, 12 months using neuropsychological test battery of memory (3), executive function (3) and attention (2). Secondary outcomes included adverse events (safety analysis), change in brain iron burden measured by quantitative susceptibility mapping MRI (target engagement), and brain volume changes (secondary efficacy measure).

RESULTS: 81 participants were recruited and randomized in a 2:1 ratio (53 drug:28 placebo). Baseline characteristics were equivalent between arms. 54 participants completed the study (withdrawals = 7 placebo and 20 deferiprone arm). Quantitative susceptibility mapping of the hippocampus at baseline and 12 months confirmed a significant (β[95%CI]:-6.7 [-11.0,-2.5]; P = 0.004) lowering of brain iron in the deferiprone arm (mean change:-3.71 ppb, 95% CI:-7.24, -0.18) compared to placebo, whose iron levels increased during this timeframe (+3.03 ppb, 95%CI: 0.27, 5.80). Participants in the Deferiprone arm showed large statistically significant cognitive decline (β[95%CI]:-0.295 [-0.456, -0.135], P<0.001, Cohen’s d: -0.704) compared to placebo (deferiprone mean change, [95%CI]:-0.86, [-1.12, -0.61]; placebo mean change, [95%CI]:-0.27, [-0.55, 0.00]).

CONCLUSION: Deferiprone (30 mg/kg/day) treatment markedly accelerated cognitive decline in people with amyloid-confirmed MCI and mild AD. These findings highlight the importance of iron for cognition in AD, and provoke new questions. It is possible that iron elevation in AD is protective, or represents iron being sequestered inappropriately (e.g. deposited in pathology) leading to functional iron deficiency; or the chosen dose of deferiprone was too high for this condition.

PMID:39782553 | DOI:10.1002/alz.086330

Alzheimers Dement. 2024 Dec;20 Suppl 6:e086486. doi: 10.1002/alz.086486.

ABSTRACT

BACKGROUND: Pivotal Alzheimer’s Disease (AD) trials typically require thousands of participants, resulting in long enrollment timelines and substantial costs. We leverage deep learning predictive models to create prognostic scores (forecasted control outcome) of trial participants and in combination with a linear statistical model to increase statistical power in randomized clinical trials (RCT). This is a straightforward extension of the traditional RCT analysis, allowing for ease of use in any clinical program. We demonstrate the application of these methods retrospectively on 3 pivotal Phase III clinical trials in mild-to-moderate AD (NCT00236431, NCT00574132, and NCT00575055).

METHOD: A probabilistic deep learning model was trained on the trajectories of nearly 7000 participants who had varying degrees of cognitive impairment, ranging from mild cognitive impairment (MCI) to moderate AD. These trajectories were collected observational studies and the control arms of RCTs. This trained model was used to forecast the control outcomes of participants in the three trials retrospectively, by entering their individual trial baseline data. The resultant forecasts are known as prognostic scores and represent comprehensive predictions across a broad range of AD outcomes. We evaluated the potential reduction in estimated variance and how this could translate to required sample size by incorporating the prognostic score as a covariate in the primary linear statistical model of each study, analyzing the 11-component Alzheimer’s Disease Assessment Scale-Cognitive Subscale (ADAS-Cog11) and the Clinical Dementia Rating Sum-of-Boxes (CDR-SB) endpoints as applicable.

RESULT: Prognostic scores have the potential to decrease estimated variance between 5% to 10% and placebo arm sample size between 7% and 17% in the 3 studies when comparing standard + prognostic score vs. standard adjustment.

CONCLUSION: Prognostic scores have the potential to increase the statistical power in clinical trials; this would enable a reduced number of subjects required to detect a significant treatment effect. Potential sample size reduction during trial planning must be carefully estimated using independent validation studies to reduce the risk of under-powering the trial.

PMID:39782540 | DOI:10.1002/alz.086486