Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Jul 18:5:308-318.
doi: 10.1016/j.trci.2019.04.004. eCollection 2019.

The relative efficiency of time-to-progression and continuous measures of cognition in presymptomatic Alzheimer's disease

Dan Li  1 , Samuel Iddi  1   2 , Paul S Aisen  1 , Wesley K Thompson  3 , Michael C Donohue  1
Affiliations

The relative efficiency of time-to-progression and continuous measures of cognition in presymptomatic Alzheimer's disease

Dan Li et al. Alzheimers Dement (N Y). .

Abstract

Introduction: Clinical trials on preclinical Alzheimer's disease are challenging because of the slow rate of disease progression. We use a simulation study to demonstrate that models of repeated cognitive assessments detect treatment effects more efficiently than models of time to progression.

Methods: Multivariate continuous data are simulated from a Bayesian joint mixed-effects model fit to data from the Alzheimer's Disease Neuroimaging Initiative. Simulated progression events are algorithmically derived from the continuous assessments using a random forest model fit to the same data.

Results: We find that power is approximately doubled with models of repeated continuous outcomes compared with the time-to-progression analysis. The simulations also demonstrate that a plausible informative missing data pattern can induce a bias that inflates treatment effects, yet 5% type I error is maintained.

Discussion: Given the relative inefficiency of time to progression, it should be avoided as a primary analysis approach in clinical trials of preclinical Alzheimer's disease.

Keywords: Alzheimer's disease; Bayesian joint mixed-effect model; Clinical trial simulations; Common close design; Cox proportional hazards model; Longitudinal data; Mixed model of repeated measures (MMRM); Statistical power.

PubMed Disclaimer

Figures

Fig. 1
Fig. 1
Observed (upper panel) and predicted (lower panel) longitudinal profiles of the seven markers for all individuals. Bold lines are locally estimated scatter plot smoother. Abbreviations: ADASDWR, Alzheimer's Disease Assessment Scale delayed word recall; MMSE, Mini-Mental State Examination; CDRSB, Clinical Dementia Rating—Sum of Boxes; FAQ, functional assessment questionnaire.
Fig. 2
Fig. 2
Kaplan-Meier estimated rate of progression to MCI or dementia. Abbreviations: MCI, mild cognitive impairment; ADNI-PAD, Alzheimer's Disease Neuroimaging Initiative—presymptomatic (or preclinical) Alzheimer's disease.
Fig. 3
Fig. 3
Results of one simulated clinical trial with 20% treatment effect from (A) analysis of change from baseline using a categorical time MMRM of the PACC; (B) a cLDA model of PACC with linear time trends; (C) a cLDA model of PACC with quadratic time effects; and (D) Kaplan-Meier curves comparing the time-to-progression to mild cognitive impairment or dementia for the two groups. Abbreviations: MMRM, mixed models of repeated measures; PACC, Preclinical Alzheimer's Cognitive Composite; cLDA, constrained longitudinal data analysis.
Fig. 4
Fig. 4
Statistical power for the MMRM, cLDA, and Cox proportional hazards model for treatment effects 0% (type I error), 20%, 30%, and 40% for sample sizes of n = 1000 (left panel) and n = 1500 (right panel). Solid lines indicate power estimates for data observed after simulated nonignorable missingness, and dashed lines indicate power that would be achieved with complete data (including observations that would be unobserved in reality). The observed data show greater power with fewer observations because the nonignorable missingness induces a bias in favor of the treatment. Abbreviations: MMRM, mixed models of repeated measures; cLDA, constrained longitudinal data analysis; PH, proportional hazards.
See this image and copyright information in PMC

References

    1. Sperling R.A., Aisen P.S., Beckett L.A., Bennett D.A., Craft S., Fagan A.M. Toward defining the preclinical stages of Alzheimer's disease: Recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimer's Demen. 2011;7:280–292. - PMC - PubMed
    1. Sperling R.A., Rentz D.M., Johnson K.A., Karlawish J., Donohue M., Salmon D.P. The A4 study: Stopping AD before symptoms begin? Sci Translational Med. 2014;6:228fs13. - PMC - PubMed
    1. ClinicalTrials.gov An efficacy and safety study of atabecestat in participants who are asymptomatic at risk for developing Alzheimer's dementia (EARLY), Tech. rep., National Library of Medicine (US), Bethesda, MD (2015 Oct 6 - 2019 Jan 21) https://clinicaltrials.gov/ct2/show/NCT02569398 Accessed January 2019.
    1. Caputo A., Racine A., Paule I., Martens E.P., Tariot P., Langbaum J.B. Rationale for selection of primary endpoints in the Alzheimer Prevention Initiative Generation study in cognitively healthy APOE4 homozygotes, Alzheimer's & Dementia. J Alzheimer's Assoc. 2017;13:P1452.
    1. Donohue M.C., Sperling R.A., Salmon D.P., Rentz D.M., Raman R., Thomas R.G. The preclinical Alzheimer cognitive composite: Measuring amyloid-related decline. JAMA Neurol. 2014;71:961–970. - PMC - PubMed