Over 200 publications, including the composite endpoints regulators now recognize. We publish our methodology in the open so reviewers meet it before your submission does.
Alzheimer's Association International Conference (AAIC) · 2024:Poster Tue-17 · Hendrix, Dickson et al.
Companies recently identified clinical study sites engaging in fraudulent practices, reducing success in Alzheimer's disease (AD) clinical trials. Successful trials are reliable…
When the sample size a conventional design demands is larger than the trial you can run, a Global Statistical Test lets the clinical-endpoint hypotheses be tested anyway. The paper sets out when GST applies, how it is pre-specified, and how it has been received in review.
14 publications
Sorted newest first
2026 · Alzheimers Dement
Nicodemus-Johnson, Jessie · Christensen, Joshua
Introduction: Alzheimer's disease (AD) study participants may present with cognitive impairment who do not have brain amyloid deposits (Aβ-). Identifying predictive biomarkers for non-amyloid-related CI may provide better screening tests for trials seeking only CI Aβ+ participants and new therapy targets. Methods: Analysis of the Bio-Hermes biomarker database identified subpopulations of clinically normal, CN Aβ- (n = 313), CI Aβ- (n = 296), and CI Aβ+ (n = 258), and CN Aβ+ (n = 84) participants. Comparative analysis of demographics, clinical assessments, biomarkers, cytokines, and proteomics results was conducted. Results: Subgroup comparison of CI Aβ- versus CN Aβ- found that neurofilament light most clearly differentiated CI Aβ- from CN Aβ- participants. No other biomarker analysis reached a level of differential significance. Discussion: Analyses showed many novel biomarkers do not differentiate CI Aβ- from CN Aβ-. New biomarkers are needed to best determine the neuropathology of the clinical presentation of AD. Highlights: NfL differentiated CN Aβ- versus cognitively impaired Aβ-. Proteomics (two platforms) did not differentially assess cognitively impaired Aβ--. Many novel biomarkers did not differentially assess cognitively impaired Aβ-. New biomarkers are needed to determine the neuropathology of AD clinical presentation.
2026 · The Journal of Prevention of Alzheimer's Disease
Hendrix, Suzanne · Mallinckrodt, Craig
Background: Soluble amyloid β1–42 (Aβ42) signals via the α7 nicotinic acetylcholine receptor to hyperphosphorylate tau in Alzheimer's disease (AD). Simufilam disrupts this pathogenic signaling by binding filamin A and disrupts its linkages with inflammatory receptors to reduce neuroinflammation. We assessed simufilam in two Phase 3 clinical trials in mild-to-moderate AD. Methods: Participants were age 50–87 with Stage 4 or 5 CE, a mini-mental state exam (MMSE) ≥16 and ≤27 and a Clinical Dementia Rating Global Score (CDR-GS) of 0.5, 1 or 2. The criterion supporting AD pathology was plasma phosphorylated (p)-tau181 or prior amyloid PET. RETHINK randomized participants to simufilam 100 mg or placebo for 52 weeks. REFOCUS evaluated simufilam 50 and 100 mg versus placebo for 76 weeks. Co-primary endpoints were change from baseline on ADAS-Cog12 and ADCS-ADL. Sub-studies assessed exploratory plasma biomarkers and, in REFOCUS only, CSF and imaging biomarkers. Results: Both trials failed to meet co-primary, secondary or exploratory biomarker endpoints. REFOCUS was terminated early, with 22% of participants still active in the trial. In the predefined mild subgroup in REFOCUS, simufilam was associated with slower cognitive decline than placebo through Week 64 (p = 0.019). This finding disappeared at Week 76 with 45% missing data and did not replicate in RETHINK. Favorable nominal exploratory post-hoc findings amongst participants with the highest half of screening plasma p-tau181 levels occurred in RETHINK but not REFOCUS. The plasma p-tau181 entry criterion did not reliably exclude amyloid PET negativity in the sub-study. Conclusions: Simufilam did not meet co-primary or secondary endpoints in these Phase 3 trials. Simufilam was safe and well tolerated. Trials registered at clinicaltrials.gov: NCT04994483 and NCT05026177
2025 · Journal of Alzheimer’s Disease
Haaland, Benjamin · Dickson, Samuel · Christensen, Joshua · Mallinckrodt, Craig · Hendrix, Suzanne
Background The AFF006 trial (NCT01117818) provided unexpected evidence of benefits of the vaccine adjuvant AD04 (aluminum oxyhydroxide) in patients with early Alzheimer's disease (AD), compared with AD02, a vaccine consisting of a peptide that mimics the N-terminal region of human amyloid-β (Aβ) conjugated with keyhole limpet hemocyanin. Objective The objective of this post hoc analysis was to assess whether this unexpected benefit of AD04 was an artifact of multiple testing (i.e., type I error inflation) or a robust result. Methods In this post hoc assessment, we used permutation testing to estimate type I error inflation due to the evaluation of multiple outcomes in AFF006. Efficacy was assessed using a patient-level global statistical test combining composite endpoints of cognition, function, and global AD. In addition, we examined the observed treatment benefits of AD04 in the context of effects observed in trials of aducanumab, donanemab, and lecanemab, monoclonal anti-Aβ antibodies that received regulatory approval for AD. Results The global statistical test suggested a treatment benefit of AD04 versus ineffective AD02 arms, even after accounting for multiplicity (primary methodology p-value, 0.03; permutation test p-value, 0.02). The observed effect estimates for AD04 compared favorably with approved monoclonal antibodies. Conclusions Post-hoc analyses are hypothesis generating rather than confirmatory. Adjusting for multiplicity using permutation testing can determine whether post-hoc effects are worth pursuing, or unlikely to be confirmed. These analyses have motivated a follow-up prospective randomized controlled trial, ADVANCE (EudraCT 2022-003532-73), in which optimized AD04 dosing will be compared to placebo in early AD.
2024 · Alzheimers Dement (N Y)
Hendrix, Suzanne · Nicodemus-Johnson, Jessie · Knowlton, Newman
Introduction: Sodium phenylbutyrate and taurursodiol (PB and TURSO) is hypothesized to mitigate endoplasmic reticulum stress and mitochondrial dysfunction, two of many mechanisms implicated in Alzheimer's disease (AD) pathophysiology. Methods: The first-in-indication phase 2a PEGASUS trial was designed to gain insight into PB and TURSO effects on mechanistic targets of engagement and disease biology in AD. The primary clinical efficacy outcome was a global statistical test combining three endpoints relevant to disease trajectory (cognition [Mild/Moderate Alzheimer's Disease Composite Score], function [Functional Activities Questionnaire], and total hippocampal volume on magnetic resonance imaging). Secondary clinical outcomes included various cognitive, functional, and neuropsychiatric assessments. Cerebrospinal fluid (CSF) biomarkers spanning multiple pathophysiological pathways in AD were evaluated in participants with both baseline and Week 24 samples (exploratory outcome). Results: PEGASUS enrolled 95 participants (intent-to-treat [ITT] cohort); cognitive assessments indicated significantly greater baseline cognitive impairment in the PB and TURSO (n = 51) versus placebo (n = 44) group. Clinical efficacy outcomes did not significantly differ between treatment groups in the ITT cohort. CSF interleukin-15 increased from baseline to Week 24 within the placebo group (n = 34). In the PB and TURSO group (n = 33), reductions were observed in core AD biomarkers phosphorylated tau-181 (p-tau181) and total tau; synaptic and neuronal degeneration biomarkers neurogranin and fatty acid binding protein-3 (FABP3); and gliosis biomarker chitinase 3-like protein 1 (YKL-40), while the oxidative stress marker 8-hydroxy-2-deoxyguanosine (8-OHdG) increased. Between-group differences were observed for the Aβ42/40 ratio, p-tau181, total tau, neurogranin, FABP3, YKL-40, interleukin-15, and 8-OHdG. Additional neurodegeneration, inflammation, and metabolic biomarkers showed no differences between groups. Discussion: While between-group differences in clinical outcomes were not observed, most likely due to the small sample size and relatively short treatment duration, exploratory biomarker analyses suggested that PB and TURSO engages multiple pathophysiologic pathways in AD.
2017 · Alzheimer's & Dementia
Background: Edonerpic (T-817; 1-{3-[2-(1-benzothiophen-5-yl)ethoxy]propyl}azetidin-3-ol maleate, Toyama Chemical, Ltd) protects against Aβ42-induced neurotoxicity and memory deficits, promotes cortical and hippocampal neuron outgrowth, and preserves hippocampal synapses and spatial memory in tau transgenic mice, possibly via sigma receptor activation. The primary objective was to assess the efficacy and safety of T-817 in Alzheimer's disease (NCT 02079909). Methods: Outpatients, ages 55 to 85, meeting criteria for probable AD, MMSE 12–22, taking stable doses of donepezil or rivastigmine, taking or not memantine, were randomly assigned (1:1:1) to placebo, 224mg, or 448mg of T-817 once/day for 52 weeks. The primary outcomes were the ADAScog and ADCS-CGIC (CIBIC+) at week 52. Secondary outcomes were the MMSE, ADCS-ADL, FAQ, and NPI at week 52; and the outcomes at weeks 12, 24, 36, and 44. Biomarkers were MRI whole brain, lateral ventricular, and hippocampal volumes; and CSF Aβ40, 42, t-tau, and p-tau; and population pharmacokinetics. Results: 140 of 158 (88.6%) participants assigned to placebo, 117 of 166 (70.5%) to 224mg, and 120 of 158 (75.9%) to 448mg completed the trial, conducted from June 2014 to December 2016 at 52 US sites. LS mean ADAScog change was 7.9, 7.5, and 7.1 for the placebo, 224mg, and 448mg groups, respectively; difference, placebo vs. 448mg, -0.8 (95% CI: -2.8, 1.1; P=0.3919). Mean ADCS-CGIC scores were 5.2, 5.2, and 5.3; difference, 0.04 (95% CI: -0.19, 0.26; P=0.7588). There were no significant differences for the secondary outcomes. P-tau was nominally significantly lower in the 448mg group vs. placebo (n=24 and N=18, P=0.0338). Hippocampal volumes decreased less in the 224mg group than placebo (n=79 and N=89; -0.27 vs. -0.39 mL, P=0.0106) but not in the 448mg group (n=76; -0.31 vs. -0.39 mL, P=0.0996). 4.4%, 13.9% and 14.6% discontinued because of AEs, placebo, 224mg, and 448mg groups, respectively. Most frequent AEs ≥ 5%: diarrhea (12.7%, 20.5%, 31.0%), nausea (3.8%, 7.8%, 5.7%); infections, injuries, falls, agitation and anxiety were more common with placebo. Conclusions: T-817 appeared safe, tolerable, with expected GI symptoms occurring early, but without evidence for clinical effect in the protocol-specified primary and secondary outcomes. Decreased CSF p-tau and hippocampal volumes require confirmation.
2015 · The Journal of Prevention of Alzheimer's Disease
Background: Optimized scales and composite outcomes have been proposed as a way to more accurately measure Alzheimer's disease related decline. AFFITOPE® AD02, is an amyloid-beta (Aβ)-targeting vaccine to elicit anti-Aβ antibodies. IMM-AD04, commonly known as Alum, originally designated as a control agent, appeared to have disease-modifying activity in a multicenter, parallel group phase II study in early AD patients. Objectives: To develop adapted outcomes for cognition, function and a composite scale with improved sensitivity to decline and treatment effects in early AD (mild plus prodromal AD) based on historical data and to assess these adapted outcomes in this phase II study. Design: Data from public datasets was analyzed using a partial least squares model in order to identify an optimally weighted cognitive outcome, Adapted ADAS-cog, and an optimally weighted ADL outcome, Adapted ADCS-ADL which were prospectively defined as co-primary endpoints for the study and were also combined into a composite scale. Data from 162 patients in the placebo groups of ADCS studies and 156 mild patients in the ADNI I study were pooled for this analysis. The Adapted ADAS-cog scale considered 13 ADAS-cog items as well as several Neuropsychological test items and CogState items, the Adapted ADCS-ADL considered all ADCS-ADL items. After the pre-specified analyses were complete, additional adapted and composite scales were investigated in a post-hoc manner. Evaluation of the adapted and composite scales was performed on Phase II trial data for AFFITOPE® AD02 (AFF006, Clinical Trial Identifier: NCT01117818) and historic data in early AD. Least square means, standard deviations, and least squares mean to standard deviation ratios were compared among adapted and composite scales and traditional scales for the 5 treatment groups in the phase II study and overall for the historic data. Treatment effect sizes and p-values were also compared for the phase II study. Results: Cognitive items that were selected for the adapted cognitive scale (aADAS-cog) and had the highest weights were Word Recall, Word Recognition, and Orientation. Delayed Word Recall and Digit Cancellation were among the items excluded due to lack of improved sensitivity to decline. Highly weighted ADL items included in the adapted functional scale (aADCS-ADL) were using the telephone, traveling, preparing a meal/snack, selecting clothing, shopping and using appliances. Excluded items were primarily basic ADLs such as eating, walking, toileting and bathing. Comparisons between traditional scales and primary outcome adapted scales show improved sensitivity to group differences with the adapted scales in the phase II trial. Most of the improvement in the sensitivity of the aADAS-cog and the aADCS-ADL is due to a larger treatment difference observed rather than the improved sensitivity to decline in the comparison groups. Conclusion: To our knowledge, this is the first study to prospectively use optimized scales as primary endpoints and to demonstrate the superior power of optimized scales and composites in early disease. Although it is possible that the treatment difference between randomized groups is due to a factor other than the treatment itself, for instance baseline imbalance, the improved power to detect these differences still argues in favor of the adapted scales. The issue of oversensitivity to detect treatment effects is controlled by selection of the alpha level for significance, and in our case will happen less than 5% of the time. Clinical relevance of the treatment difference should be assessed separately from statistical significance, and in this phase II study, is supported by significant or similar sizes of effect on function, behaviour and quality of life outcomes, which are important to patients and caregivers.
2012 · Alzheimer's & Dementia
Background: No well-established, validated endpoints exist that are sensitive to change in MCI populations in clinical trials. Our goal was to develop a tool based on standard clinical items that would demonstrate maximum responsiveness to progression and to treatment in an MCI population and would also perform well in a mild AD population (collectively referred to as Early AD). This analysis uses a novel approach by utilizing data from multiple studies to investigate responsiveness to progression and treatment effects rather than sensitivity to baseline deficits. Methods: This tool was built empirically with no a priori assumptions. A partial least squares (PLS) regression model used placebo data from 4 MCI studies over 12 months to select the combination of cognitive and functional items which is most sensitive to change over time, using items from a variety of well-established and validated scales. The PLS regression coefficients from the model were used to form a weighted composite score. The resulting composite score is comprised of ADAS-Cog, MMSE and CDR items. Performance of the composite score was assessed against the original scales in an MCI population, in enriched (CSF Aβ positive) MCI subgroups, with split sample validation, in the presence of a treatment effect and in a mild AD patient population combining data from 3 studies. Results: A composite clinical score was devised from 12 items from the ADAS-Cog, MMSE, and CDR-SB that assess both cognition and global function. This score demonstrates improved sensitivity to decline, as well as reduced heterogeneity, as compared with original scales, and is responsive to treatment effect in MCI, enriched MCI and mild AD populations. The composite score allows for substantial sample sizes reductions in non-enriched and enriched MCI populations for a 12-month study (see Figure). Conclusions: A new composite clinical score that utilizes relevant items from validated and well-established clinical tools provides a tool that can be used as a single clinical outcome in studies that target MCI, enriched MCI and mild AD populations. This tool will enable the use of substantially smaller sample sizes due to improved sensitivity to disease progression and treatment effects.
2008 · Alzheimer's & Dementia
Background: Tarenflurbil is a Selective Aβ42-Lowering Agent (SALA) that modulates γ-secretase activity to preferentially reduce production of Aβ42 in vivo and in vitro. Evidence for potential benefit of tarenflurbil 800 mg bid in subjects with mild AD was recently observed in a randomized, double-blind Phase 2 trial of up to 24 months of treatment. Methods: A target of 1600 subjects with mild AD (MMSE score 20–26) were to be randomized (1:1) to receive tarenflurbil 800 mg bid or placebo for 18 months. Randomization was stratified according to stable use/nonuse of acetylcholinesterase inhibitors (AChEIs) and/or memantine. The co-primary efficacy outcomes are the rate of change (slope) in ADAS-cog and the ADCS-ADL, with assessments conducted every 3 months. The secondary outcome is the CDR-sb, with additional exploratory outcomes including the NPI, Quality of Life-AD, and Caregiver Burden Inventory. ECGs are obtained every 3 months, with adverse events monitored throughout the study. Plasma samples are collected every 3 months for pharmacokinetic and exploratory biomarker analyses. Results: This trial enrolled 1684 subjects at 133 sites in the United States. The last patient visit occured in March 2008, with database lock occuring in June of 2008. At enrollment, 33% of subjects were using AChEIs alone (stable for ≥ 6 months), 6% of subjects were using memantine alone (stable for ≥ 3 months), 41% of subjects were using both AChE is and memantine and 19% of subjects were taking no AD therapy. A second Phase 3 trial of similar design, fully enrolled with 840 subjects, is ongoing in the US, Canada and Europe and is expected to be completed in the clinic in October of 2008. Conclusions: This Phase 3 protocol was developed based on Phase 2 trial results. It is powered to evaluate the efficacy of tarenflurbil with respect to the primary outcomes, and to examine its effect both as monotherapy and as add-on therapy with currently marketed AD medications.
2007 · Alzheimer Disease & Associated Disorders
Abstract: To evaluate the safety and tolerability and pharmacokinetic properties of R-flurbiprofen (Tarenflurbil) in normal elderly individuals and to determine the effect of the drug on amyloid beta 42 (Abeta42) levels, we conducted a double-blind, placebo-controlled study of 48 healthy subjects aged 55 to 80. Three successive cohorts were randomized to doses of 400, 800, or 1600 mg/d, or placebo, given as 2 divided doses for 21 days. Blood and cerebrospinal fluid were collected for pharmacokinetic studies and measurement of Abeta levels at baseline and on day 21. R-flurbiprofen was well-tolerated at all 3 doses. The compound penetrated the blood-brain barrier in a dose-dependent manner. From baseline to 21 days, comparisons between study groups revealed no significant differences in changes of cerebrospinal fluid Abeta42 levels and no significant differences in changes of plasma Abeta42 levels at the time of trough drug level at 21 days of treatment. Further analysis of drug concentration-response for plasma samples showed that at the time of peak plasma concentration, higher plasma drug concentration was related to lower Abeta42 plasma levels (P=0.016). R-flurbiprofen had an excellent safety profile and showed dose-dependent central nervous system penetration. Exploratory analyses of plasma Abeta and peak drug levels suggested a short-term effect in plasma that warrants independent verification. The safety, tolerability, and pharmacokinetic profile of R-flurbiprofen in these older individuals support the ongoing studies of this compound in patients with Alzheimer disease.
2007 · European Journal of Neurology
Background: Tarenflurbil is a Selective Ab42-Lowering Agent (SALA) that lowers brain levels of Ab42 in a mouse model of AD and chronic dosing in this model prevents defects in learning and memory. These data and the Phase-2 study indicating sustained benefit in activities of daily living, global function and cognition in mild AD patients, suggest the potential for tarenflurbil to have disease-modifying properties. Methods: This was a placebo-controlled, 1-year study evaluating tarenflurbil in 207 patients with mild-to-moderate AD. At randomization, 94% of subjects were on stable acetylcholinesterase inhibitor therapy. An exploratory post-hoc analysis was performed which compared time to adverse psychiatric events between treatment groups. Results: In subjects with mild AD (MMSE 20–26) was a significant delay in time to clinically significant adverse psychiatric events, 800 mg BID compared to placebo (p=0.011). Among 35% of the placebo group who had an event, the median time was approximately 106 days. In the 800 mg BID group, the median time to event was greater than 333 days with only 14% of this group having an event. The most common psychiatric events reported in the placebo group were agitation, aggression, confusional state and depression.
2006 · Presented as a poster exhibit at the 10th International Conference on Alzheimer's Disease and Related Disorders, Madrid, Spain
2006 · Alzheimer's & Dementia
Background: MPC–7869 (R–flurbiprofen) is a Selective Aβ42–Lowering Agent (SALA). In a mouse model of AD (Tg2576), MPC–7869 lowers brain levels of Aβ42 and chronic dosing in this model reduces brain amyloid pathology and prevents defects in learning and memory. These data suggest a potential for MPC–7869 to have disease–modifying properties. Objective(s): This study evaluated the efficacy and safety of treatment with MPC–7869 for 12 months in subjects with mild–to–moderate AD, and includes data from a 1–year follow–on study. Methods: This was a placebo–controlled, double–blind, 1–year trial evaluating 400 mg BID and 800 mg BID of MPC–7869 in 207 patients with mild–to–moderate AD (MMSE 15–26, with an average MMSE score of 21). The mean age of the subjects at baseline was 75 years and 94% of subjects were on stable acetylcholinesterase inhibitor therapy. Primary outcomes included measures of cognition (ADAS–cog), activities of daily living (ADCS–ADL), and global function (CDR–sb). A population pharmacokinetic analysis was also performed. At the end of this study, over 80% of eligible patients were enrolled into a 1–year follow–on treatment study in which placebo patients were randomized into one of the two treatment groups and treated patients continued their stable dose. Treatment groups remained blinded to patient/investigator. Results: A prespecified interaction analysis revealed that mild and moderate AD patients responded differently to MPC–7869 (P= 0.03). In mild AD patients (MMSE 20–26) taking the 800–mg BID dose, statistically significant benefit was observed at 12 months in activities of daily living with a treatment effect size of d=0.44 (P= 0.033) and global function with a treatment effect size of d=0.42 (P= 0.042) with a positive trend observed in cognition. In addition, there was a significant plasma drug concentration to response relationship (ADCS–ADL, P= 0.037 and CDR–sb, P= 0.019). No benefit was observed in moderate AD patients. MPC–7869 was well tolerated and patients taking the 800–mg BID dose continued to show benefit in the 1–year follow–on study. Conclusions: MPC–7869 has an attractive therapeutic and safety profile in patients with mild AD. This study justifies larger–scale confirmatory trials of MPC–7869 as an amyloid–based intervention strategy for AD treatment.
2005 · Alzheimer's & Dementia
Background: Several epidemiological studies have suggested that longer-term use of NSAIDs may reduce the risk of developing Alzheimer's disease (AD). Although there were encouraging results from some earlier studies, more recent longer-term, larger, placebo-controlled clinical trials utilising drugs with an anti-inflammatory action have produced disappointing outcomes. Re-analysis of the epidemiological data has shown that the protective effect is limited to a subgroup of NSAIDs, including flurbiprofen, which have Aβ-42 lowering properties and which were not used in the previous longer-term placebo-controlled trials. Flurizan (MPC-7869 (R)-flurbiprofen) is a single enantiomer of flurbiprofen, which has been shown to lower brain levels of Aβ-42 in a mouse model of AD (Tg2576), with some evidence of an effect on learning and memory. There is little risk of gastric toxicity because of a lack of anti-COX activity, and the compound has been shown to be safe and well tolerated in healthy older volunteers (55-80yrs) at doses of up to 1600mg per day when administered for 21 days in a phase I study. It is therefore an exciting potential disease modifying treatment for AD. Objective(s): This presentation will provide efficacy and safety data from a clinical trial of Flurizan, which to our knowledge will be the first multi-centre, placebo-controlled, double-blind clinical study of a selective Aβ-42 lowering agent in patients with mild to moderate Alzheimer's disease. Methods: This is a one-year trial evaluating both 400mg BID and 800mg BID of (R)-flurbiprofen per day, in 210 patients (50% female) with mild to moderate AD (MMSE 15-26). It employed the ADAS-cog, ADCS-ADL, CDR-sb, CIBIC plus, NPI and MMSE as outcome measures. At baseline the mean age of the subjects was 75 years with an average MMSE score 21 and an average standard ADAS-cog score 23. Of the patients enrolled in the trial, 94 per cent were also taking stable doses of cholinesterase inhibitors. Concomitant treatment with memantine was not allowed. Conclusions: The study completes in March 2005, and the cognitive, functional, global and behavioural outcomes will be presented, together with the safety data.
Loading more publications…
Looking for older work, or a specific methodology paper? Our research team can point you to the right reference.
Ask our team
