Multivariate Protein Signatures of Pre-Clinical Alzheimer's Disease in the Alzheimer's Disease Neuroimaging Initiative (ADNI) Plasma Proteome Dataset

Priority Research Centre for Bioinformatics, Biomarker Discovery and Information-Based Medicine, The University of Newcastle, Callaghan, New South Wales, Australia.
PLoS ONE (Impact Factor: 3.23). 04/2012; 7(4):e34341. DOI: 10.1371/journal.pone.0034341
Source: PubMed


Recent Alzheimer's disease (AD) research has focused on finding biomarkers to identify disease at the pre-clinical stage of mild cognitive impairment (MCI), allowing treatment to be initiated before irreversible damage occurs. Many studies have examined brain imaging or cerebrospinal fluid but there is also growing interest in blood biomarkers. The Alzheimer's Disease Neuroimaging Initiative (ADNI) has generated data on 190 plasma analytes in 566 individuals with MCI, AD or normal cognition. We conducted independent analyses of this dataset to identify plasma protein signatures predicting pre-clinical AD.
We focused on identifying signatures that discriminate cognitively normal controls (n = 54) from individuals with MCI who subsequently progress to AD (n = 163). Based on p value, apolipoprotein E (APOE) showed the strongest difference between these groups (p = 2.3 × 10(-13)). We applied a multivariate approach based on combinatorial optimization ((α,β)-k Feature Set Selection), which retains information about individual participants and maintains the context of interrelationships between different analytes, to identify the optimal set of analytes (signature) to discriminate these two groups. We identified 11-analyte signatures achieving values of sensitivity and specificity between 65% and 86% for both MCI and AD groups, depending on whether APOE was included and other factors. Classification accuracy was improved by considering "meta-features," representing the difference in relative abundance of two analytes, with an 8-meta-feature signature consistently achieving sensitivity and specificity both over 85%. Generating signatures based on longitudinal rather than cross-sectional data further improved classification accuracy, returning sensitivities and specificities of approximately 90%.
Applying these novel analysis approaches to the powerful and well-characterized ADNI dataset has identified sets of plasma biomarkers for pre-clinical AD. While studies of independent test sets are required to validate the signatures, these analyses provide a starting point for developing a cost-effective and minimally invasive test capable of diagnosing AD in its pre-clinical stages.

Download full-text


Available from: Pablo Moscato
  • Source
    • "Table 1. List of analyte signature sets identified in [7] "

    Full-text · Dataset · Apr 2015
  • Source
    • "For example, the concentration ratio of two different analytes has been used previously to establish CSF biomarkers of AD from univariate analysis (e.g., A␤ 42 /tau) [34] [35]. We have previously demonstrated the benefit of considering meta-features for enhancing the accuracy of blood biomarker panels for AD [19] [20]. Again this is completely counter to conventional biomedical thinking, which assumes, usually without question, that only measures that show real and statistically significant differences between disease and control groups will be of importance and discards all other measures from consideration. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Interventions to delay or slow Alzheimer's disease (AD) progression are most effective when implemented at pre-clinical disease stages, making early diagnosis essential. For this reason, there is an increasing focus on discovery of predictive biomarkers for AD. Currently, the most reliable predictive biomarkers require either expensive (brain imaging) or invasive (cerebrospinal fluid collection) procedures, leading researchers to strive toward identifying robust biomarkers in blood. Yet promising early results from candidate blood biomarker studies are being refuted by subsequent findings in other cohorts or using different assay technologies. Recent evidence suggests that univariate blood biomarkers are not sufficiently sensitive or specific for the diagnosis of disorders as complex, multifactorial, and heterogeneous as AD. To overcome these present limitations, more consideration must be given to the development of 'biomarker panels' assessing multiple molecular entities. The selection of such panels should draw not only on traditional statistical approaches, whether parametric or non-parametric, but also on newer non-statistical approaches that have the capacity to retain and utilize information about all individual study participants rather than collapsing individual data into group summary values (e.g., mean, variance). These new approaches, facilitated by advances in computing, have the potential to preserve the context of interrelationships between different molecular entities, making them amenable to the development of panels that, as a multivariate collective, can overcome the challenge of individual variability and disease heterogeneity to accurately predict and classify AD. We argue that the AD research community should take fuller advantage of these approaches to accelerate discovery.
    Full-text · Article · Oct 2013 · Journal of Alzheimer's disease: JAD
  • Source
    • "Their hope was that identifying the disease in this preclinical stage will facilitate the development of new treatments to slow or halt the progression of the disease [2]. Across the globe, research on the predictive value of early medical tests used to detect Alzheimer's disease during this preclinical phase is underway, and is showing promising results [3,4]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Research is underway to develop an early medical test for Alzheimer's disease (AD). To evaluate potential demand for such a test, we conducted a cross-sectional telephone survey of 2,678 randomly selected adults across the United States and four European countries. Most surveyed adults (67%) reported that they are "somewhat" or "very likely" to get an early medical test if one becomes available in the future. Interest was higher among those worried about developing AD, those with an immediate blood relative with AD, and those who have served as caregivers for AD patients. Older respondents and those living in Spain and Poland also exhibited greater interest in testing. Knowing AD is a fatal condition did not influence demand for testing, except among those with an immediate blood relative with the disease. Potential demand for early medical testing for AD could be high. A predictive test could not only advance medical research, it could transform political and legal landscapes by creating a large constituency of asymptomatic, diagnosed adults. Key words: Alzheimer's disease, medical testing, predictive testing, medical decision-making, public attitudes, preclinical.
    Full-text · Article · Sep 2013 · Alzheimer's Research and Therapy
Show more