Elsevier

Neurobiology of Aging

Volume 31, Issue 8, August 2010, Pages 1429-1442
Neurobiology of Aging

Optimizing the design of the clinical trials of the future
Boosting power for clinical trials using classifiers based on multiple biomarkers

https://doi.org/10.1016/j.neurobiolaging.2010.04.022Get rights and content

Abstract

Machine learning methods pool diverse information to perform computer-assisted diagnosis and predict future clinical decline. We introduce a machine learning method to boost power in clinical trials. We created a Support Vector Machine algorithm that combines brain imaging and other biomarkers to classify 737 Alzheimer's disease Neuroimaging initiative (ADNI) subjects as having Alzheimer's disease (AD), mild cognitive impairment (MCI), or normal controls. We trained our classifiers based on example data including: MRI measures of hippocampal, ventricular, and temporal lobe volumes, a PET-FDG numerical summary, CSF biomarkers (t-tau, p-tau, and Aβ42), ApoE genotype, age, sex, and body mass index. MRI measures contributed most to Alzheimer's disease (AD) classification; PET-FDG and CSF biomarkers, particularly Aβ42, contributed more to MCI classification. Using all biomarkers jointly, we used our classifier to select the one-third of the subjects most likely to decline. In this subsample, fewer than 40 AD and MCI subjects would be needed to detect a 25% slowing in temporal lobe atrophy rates with 80% power—a substantial boosting of power relative to standard imaging measures.

Section snippets

Subjects

Baseline neuroimaging and biomarker data were downloaded from the ADNI public database (www.loni.ucla.edu/ADNI/Data/) on or before 20 November 2009 and reflect the status of the database at that point; as data collection is ongoing. ADNI is a large 5 year study launched in 2004 with the primary goal of testing whether serial MRI, PET, other biological markers, and clinical and neuropsychological assessments at multiple sites (as in a typical clinical trial), can replicate results from smaller

AD and MCI classification based on MRI markers, ApoE genotype and demographic information

We first used the three MRI-derived summaries, ApoE genotype and demographic variables (age, sex and BMI) for AD and MCI classification with 635 ADNI subjects. SVM training was performed with all seven features using a linear kernel with C = 1, and the contributions of the different biomarkers were put into a rank order (best to worst) based on their SVM weights, assessed by wi2 in the notation of SVM described in the methods. The rank orders are shown in Table 2.

We then aimed to find the top N

Discussion

We explored the power of several baseline biomarkers for AD and MCI, used jointly for diagnostic classification and for predicting future (1 year) cognitive decline in MCI. We also showed how to apply the multimodality classifiers to choose subsamples of subjects for boosting power in clinical trials. We determined combinations of regional MRI numerical summaries with demographic variables and ApoE that best classified AD vs. control and MCI vs. control. The top set of complementary biomarkers

Disclosure statement

The authors have no potential financial or personal conflicts of interest including relationships with other people or organizations within 3 years of beginning the work submitted that could inappropriately influence their work.

Acknowledgements

Data collection and sharing for this project was funded by the Alzheimer's disease Neuroimaging Initiative (ADNI) (National Institutes of Health, Grant U01 AG024904). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: Abbott, AstraZeneca AB, Bayer Schering Pharma AG, Bristol-Myers Squibb, Eisai Global Clinical Development, Elan Corporation, Genentech, GE Healthcare,

References (71)

  • X. Hua et al.

    Optimizing power to track brain degeneration in Alzheimer's disease and mild cognitive impairment with tensor-based morphometry: an ADNI study of 515 subjects

    Neuroimage

    (2009)
  • X. Hua et al.

    3D characterization of brain atrophy in Alzheimer's disease and mild cognitive impairment using tensor-based morphometry

    Neuroimage

    (2008)
  • X. Hua et al.

    Tensor-based morphometry as a neuroimaging biomarker for Alzheimer's disease: an MRI study of 676 AD, MCI, and normal subjects

    Neuroimage

    (2008)
  • C.R. Jack et al.

    Hypothetical model of dynamic biomarkers of the Alzheimer's pathological cascade

    Lancet Neurol

    (2010)
  • F. Kruggel et al.

    Impact of scanner hardware and imaging protocol on image quality and compartment volume precision in the ADNI cohort

    Neuroimage

    (2010)
  • J.P. Lerch et al.

    Automated cortical thickness measurements from MRI can accurately separate Alzheimer's patients from normal elderly controls

    Neurobiol Aging

    (2008)
  • L. Lukas et al.

    Brain tumor classification based on long echo proton MRS signals

    Artif Intell Med

    (2004)
  • J.H. Morra et al.

    Validation of a fully automated 3D hippocampal segmentation method using subjects with Alzheimer's disease mild cognitive impairment, and elderly controls

    Neuroimage

    (2008)
  • J.H. Morra et al.

    Automated mapping of hippocampal atrophy in 1-year repeat MRI data from 490 subjects with Alzheimer's disease, mild cognitive impairment, and elderly controls

    Neuroimage

    (2009)
  • J. Mourão-Miranda et al.

    Classifying brain states and determining the discriminating activation patterns: Support Vector Machine on functional MRI data

    Neuroimage

    (2005)
  • R.C. Petersen

    Alzheimer's disease: progress in prediction

    Lancet Neurol

    (2010)
  • H.D. Protas et al.

    FDDNP binding using MR derived cortical surface maps

    Neuroimage

    (2010)
  • J.L. Stein et al.

    Genome-wide analysis reveals novel genes influencing temporal lobe structure with relevance to neurodegeneration in Alzheimer's disease

    Neuroimage

    (2010)
  • D. Sun et al.

    Elucidating an MRI-based biomarker for psychosis: classification using probabilistic brain atlas and machine learning algorithms

    Biol Psychiatry 66

    (2009)
  • P. Vemuri et al.

    Alzheimer's disease diagnosis in individual subjects using structural MR images: validation studies

    Neuroimage

    (2008)
  • S.M. Wilson et al.

    Automated MRI-based classification of primary progressive aphasia variants

    Neuroimage

    (2009)
  • H. Zhang et al.

    Use of the Support Vector Machine and Sensitivity of an AD-Related Region-of-Interest Gray Matter Classifier in Identifying Amnestic MCI Subjects Who Convert t6o AD: Preliminary Findings From the AD Neuroimaging Initiative

    Presented at the International Conference on Alzheimer's Disease, Chicago, July 26–

    (2008)
  • G.E. Alexander et al.

    Six month MRI gray matter declines in Alzheimer's dementia evaluated by voxel-based morphometry with multivariate network analysis: Preliminary findings from the Alzheimer's Disease Neuroimaging Initiative

    Alzheimers Dement

    (2008)
  • Alzheimer's disease facts and figures

    Alzheimers Dement

    (2009)
  • L.G. Apostolova et al.

    3D comparison of hippocampal atrophy in amnestic mild cognitive impairment and Alzheimer's disease

    Brain

    (2006)
  • L.G. Apostolova et al.

    Conversion of mild cognitive impairment to Alzheimer disease predicted by hippocampal atrophy maps

    Arch Neurol

    (2006)
  • L.G. Apostolova et al.

    Three-dimensional gray matter atrophy mapping in mild cognitive impairment and mild Alzheimer disease

    Arch Neurol

    (2007)
  • N.A. Azad et al.

    Gender differences in dementia risk factors

    Gend Med

    (2007)
  • M.N. Braskie et al.

    Plaque and tangle imaging and cognition in normal aging and Alzheimer's disease

    Neurobiol Aging Nov 10

    (2008)
  • A.S. Buchman et al.

    Change in body mass index and risk of incident Alzheimer disease

    Neurology

    (2005)
  • Cited by (0)

    Data used in the preparation of this article were obtained from the Alzheimer's disease Neuroimaging Initiative (ADNI) database (www.loni.ucla.edu/ADNI). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. Complete listing of ADNI investigators available at: www.loni.ucla.edu/ADNI/Collaboration/ADNI_Manuscript_Citations.pdf.

    View full text