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.
15 publications
Sorted newest first
2011 · Book
Brown, Bruce · Hendrix, Suzanne
An insightful guide to understanding and visualizing multivariate statistics using SAS®, STATA®, and SPSS®. Multivariate Analysis for the Biobehavioral and Social Sciences: A Graphical Approach outlines the essential multivariate methods for understanding data in the social and biobehavioral sciences. Using real-world data and the latest software applications, the book addresses the topic in a comprehensible and hands-on manner, making complex mathematical concepts accessible to readers. The authors promote the importance of clear, well-designed graphics in the scientific process, with visual representations accompanying the presented classical multivariate statistical methods . The book begins with a preparatory review of univariate statistical methods recast in matrix notation, followed by an accessible introduction to matrix algebra. Subsequent chapters explore fundamental multivariate methods and related key concepts, including: Factor analysis and related methods, multivariate graphics, canonical correlation, hotelling's T-squared, multivariate analysis of variance (MANOVA), multiple regression and the general linear model (GLM). Each topic is introduced with a research-publication case study that demonstrates its real-world value. Next, the question "how do you do that?" is addressed with a complete, yet simplified, demonstration of the mathematics and concepts of the method. Finally, the authors show how the analysis of the data is performed using Stata®, SAS®, and SPSS®. The discussed approaches are also applicable to a wide variety of modern extensions of multivariate methods as well as modern univariate regression methods. Chapters conclude with conceptual questions about the meaning of each method; computational questions that test the reader's ability to carry out the procedures on simple datasets; and data analysis questions for the use of the discussed software packages. Multivariate Analysis for the Biobehavioral and Social Sciences is an excellent book for behavioral, health, and social science courses on multivariate statistics at the graduate level. The book also serves as a valuable reference for professionals and researchers in the social, behavioral, and health sciences who would like to learn more about multivariate analysis and its relevant applications.
2011 · Progress in neurobiology
Abstract: Despite enormous financial and scientific efforts, still no approved disease-modifying therapies exist for Alzheimer's disease (AD). During the last decade all Phase III clinical trials on disease modifiers in AD have failed. The dementia stage of AD being probably too late in order to allow for successful disease modification has been identified as a possible culprit that could explain the failure of so many clinical trials. In parallel, a major development in the diagnostic research field of AD was achieved by the recent proposal of new diagnostic criteria for AD, which also specifically incorporate the use of biomarkers as defining criteria for preclinical stages of AD, thus extending the traditional definition of disease to very early stages that may be a more feasible target for various disease modifying therapeutic interventions. This ongoing paradigm shift in AD definition and diagnosis represents a fundamental basis for redefinition of interventional trials in AD, allowing to specifically focus on preventative measures during very early pathophysiologically confirmed stages of disease. This consensus paper reflects the outcome from a European Union and North American Task Force meeting comprised of experts from academia, industry, private foundations, and regulatory agencies that was convened in Toulouse, France on November 5, 2010 and that focused on prevention trials in AD. This position paper thoroughly analyzes prerequisites for successful preventative trials in AD and concludes with concrete recommendations on biomarkers, statistical tools and other variables important for improved study designs suitable for preventative as well as for early therapeutic interventional trials in AD.
2011 · American Statistical Association 2002 Proceedings of the Section on Statistical Graphics (1430-1454)
Brown, Bruce · Hendrix, Suzanne
Canonical correlation is one of the least used of the multivariate methods. It is the thesis of this paper that it is seldom used because it is incomplete and gives little information by itself. However, it can be highly illuminating when used to create a linked context of two or more multivariate spaces that in and of themselves have interesting internal structure. That internal structure can be experimental (with a MANOVA analysis within each space), a time series pattern linked across the spaces, or even just loose internal groupings by a collection of exploratory categorical variables. Also, the linking across the multidimensional spaces can be causal (e.g., mutual fund performance as predicted by market indices), or merely parallel (e.g., convergent behavioral and physiological measures of performance on cognitive tasks). A demonstration is given of canonical correlation graphs for a time series internal structure in a causally linked space.
2011 · Alzheimer's & Dementia
Background: Top-line results of a 24-week, multinational, randomized, placebo-controlled trial in patients with moderate to severe Alzheimer's disease (AD) receiving stable concurrent cholinesterase inhibitor treatment (MEM-MD-50, NCT00322153) demonstrated the efficacy of a new, extended-release (ER) formulation of memantine (28 mg, once daily) on primary outcome measures (SIB and CIBIC-plus), as well as on the NPI and a verbal fluency test. In this post-hoc analysis, we examined the effects of memantine ER on individual SIB domains, as well as on aggregated domains defined previously (Schmitt et al., 2006). Methods: Treatment groups were compared in terms of mean change from Baseline at Endpoint for nine SIB domains (Social Interaction, Memory, Orientation, Language, Attention, Praxis, Visuospatial Ability, Construction, and Orienting to Name) and combinations of domains aggregated using a face-valid approach into three higher-order subscales: MEMORY (memory, attention, orientation, orienting to naming), LANGUAGE (language, social interaction), and PRAXIS (praxis, visuospatial ability, construction). Between-group comparisons were based on the intent-to-treat population (placebo: n = 328; memantine ER: n = 333) and performed by means of an ANCOVA model with treatment group and study center as factors and baseline value as covariate, using observed cases (OC) and the last observation carried forward (LOCF) approach to missing data. In addition, a mixed-effects model with repeated measures (MMRM) that included terms for treatment group, visit, treatment-by-visit interaction, baseline score, baseline-by-treatment interaction, and center was used to compare the groups across the entire trial. No adjustments for multiple comparisons were made. Results: Significant advantage of memantine ER over placebo was observed for the domains of Memory (OC, P = 0.021; LOCF, P = 0.016; MMRM, P = 0.008), Language (OC, P = 0.003; LOCF, P = 0.004; MMRM, P = 0.001), Attention (OC, P = 0.014; LOCF, P = 0.003; MMRM, P = 0.004), Praxis (OC, P = 0.015; LOCF, P = 0.002; MMRM, P = 0.002), Orientation (LOCF, P = 0.043; MMRM, P = 0.028), and Construction (OC, P = 0.042), and for all three higher-order subscales (MEMORY: OC, P = 0.002; LOCF, P = 0.003; MMRM P < 0. 001; LANGUAGE: OC, LOCF, P = 0.003; MMRM, P = 0.001; PRAXIS: OC, P = 0.012; LOCF, P = 0.004; MMRM, P = 0.004). Conclusions: In this post-hoc analysis, memantine ER was associated with significant improvement relative to placebo on several cognitive domains, including memory, language, praxis, and attention.
2011 · Alzheimer's & Dementia
Background: We have proposed an Alzheimer's Prevention Initiative (API) to relate a presymptomatic Alzheimer's disease (AD) treatment's biomarker effects to clinical outcome in cognitively normal people at the highest imminent risk to develop AD, including the world's largest kindred of early-onset AD (EOAD) causing mutation carriers (mean age at clinical onset=45), from Antioquia, Colombia. Here, we used longitudinal data from cognitively normal E280A Presenilin 1 (PS1) mutation carriers and non-carriers over age 35 to 1) characterize the combination of cognitive tests most sensitive to cognitive decline, 2) estimate the number of mutation carriers needed in randomized clinical trial (RCTs) to detect AD-slowing treatment effect on the composite cognitive endpoint, and 3) estimate the treatment effect that could be detected in 75 PS1 mutation carriers with 80% power and p=0.05. Methods: A battery of 19 cognitive tests, acquired every 2-5 years between 1995 and 2010 by the Neuroscience group at the University of Antioquia, was used to calculate the mean-to-standard-deviation ratios (MSDR) for each combination of one to six measurements. Measurements were adjusted for aging/practice effects using data from no carriers. The best combination was used to estimate statistical power in 24-60 month presymptomatic AD RCTs. Results: After practice/aging correction, the optimal combination to predict cognitive decline included CERAD word list delayed recall, category verbal fluency, MMSE Orientation and Time, Constructional Praxis and Ravens progressive matrices. (A similar pattern was observed in cognitively normal older APOE4 carriers [Langbaum et al, ICAD abstract 2011]). We estimate the need for 215/79 PS1 mutation carriers per group over the age of 35, respectively, to detect a 25% treatment effect in a 24/60-month RCT. We estimate that 75 carriers per group would permit us to detect 43/26% treatment effects, respectively in a 24/60-month RCT. Conclusions: We have identified a combination of cognitive tests to evaluate presymptomatic AD treatments in cognitive normal people at highest imminent risk for EOAD. We have found a similar combination in those at highest risk for late-onset AD. We will continue to develop this approach in preparation for the presymptomatic AD/surrogate marker development trials proposed in the API.
2011 · The American Journal of Human Genetics
2011 · Journal of the American Medical Directors Association
2011 · Alzheimer's & Dementia
Background: We have proposed an Alzheimer's Prevention Initiative (API) to relate a pre-symptomatic Alzheimer's disease (AD) treatment's biomarker effects to clinical outcome in cognitively normal people who, based on age and genetic background, are at highest imminent risk of symptomatic AD. Here, we used longitudinal data from two cohort studies at the Rush Alzheimer's Disease Center to identify a new cognitive composite score that is sensitive to cognitive decline associated with pre-symptomatic AD to inform on the design of a randomized clinical trial (RCT) in apolipoprotein E (APOE) e4 carriers. Methods: Using a battery of 19-21 cognitive tests, the annualized mean-to-standard-deviation ratios (MSDR) of the change over time were calculated for the un-weighted sum of every combination of two to six tests for those who developed cognitive impairment (MCI or AD) in the five years prior to diagnosis. Measurements were adjusted for aging/practice effects using data from participants who remained cognitively normal. The best combinations were evaluated for construct validity. This optimal test combination was then examined in APOE4 carriers. Results: The optimal combination of measurements that was selected to sensitively measure cognitive decline over time included Logical Memory-delayed recall, CERAD word list-delayed recall, category fluency, Ravens progressive matrices, and MMSE (annual MSDR =0.182847). A similar composite test was independently developed using data from a longitudinal cohort of cognitively normal PS1 E280A mutation carriers (Ayutyanont et al, ICAD abstract 2011). This composite score was also shown to be sensitive to decline in APOE4 carriers relative to non-carriers between the ages of 70 and 85. Using this composite cognitive endpoint, we estimate that 4,360/1,100 APOE4 carriers per group would permit us to detect a 30% treatment effect in a 24/60-month RCT, respectively. Conclusions: In this first phase of the process, we have identified a combination of cognitive tests to evaluate pre-symptomatic AD treatments in cognitive normal people at high imminent risk for late-onset AD, and have shown that a similar combination performs well in individuals at highest risk for early-onset AD. We will continue to develop and refine this approach in preparation for the pre-symptomatic AD/surrogate marker development trials proposed in the API.
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
