We propose a method for recruiting asymptomatic Amyloid positive individuals in clinical trials, using a two-step process. We first select during a pre-screening phase a subset of individuals which are more likely to be amyloid positive based on the automatic analysis of data acquired during routine clinical practice, before doing a confirmatory PET-scan to these selected individuals only. This method leads to an increased number of recruitments and to a reduced number of PET-scans, resulting in a decrease in overall recruitment costs. We validate our method on three different cohorts, and consider five different classification algorithms for the pre-screening phase. We show that the best results are obtained using solely cognitive, genetic and socio-demographic features, as the slight increased performance when using MRI or longitudinal data is balanced by the cost increase they induce. We show that the proposed method generalizes well when tested on an independent cohort, and that the characteristics of the selected set of individuals are identical to the characteristics of a population selected in a standard way. The proposed approach shows how Machine Learning can be used effectively in practice to optimize recruitment costs in clinical trials.
DuboisBHampelHFeldmanHH, et al.Preclinical Alzheimer’s disease: definition, natural history, and diagnostic criteria. Alzheimer’s Dementia2016; 12: 292–323.
2.
JackCRKnopmanDSJagustWJ, et al.Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade. Lancet Neurol2010; 9: 119–119.
3.
HardyJAllsopD. Amyloid deposition as the central event in the aetiology of Alzheimer’s disease. Trends Pharmacol Sci1991; 12: 383–388.
4.
HardyJSelkoeDJ. The amyloid hypothesis of Alzheimer’s disease: progress and problems on the road to therapeutics. Science2002; 297: 353–356.
5.
HardyJAHigginsGA. Alzheimer’s disease: the amyloid cascade hypothesis. Science1992; 256: 184–185.
6.
KarranEMerckenMStrooperBD. The amyloid cascade hypothesis for Alzheimer’s disease: an appraisal for the development of therapeutics. Nat Rev Drug Discov2011; 10: 698–712.
7.
DoodyRSThomasRGFarlowM, et al.Phase 3 trials of solanezumab for mild-to-moderate Alzheimer’s disease. New Engl J Med2014; 370: 311–321.
8.
SallowaySSperlingRFoxNC, et al.Two phase 3 trials of Bapineuzumab in mild-to-moderate Alzheimer’s disease. New Engl J Med2014; 370: 322–333.
9.
BeckerREGreigNH. A new regulatory road-map for Alzheimer’s disease drug development. Curr Alzheimer Res2014; 11: 215–220.
10.
O’BrienJTHerholzK. Amyloid imaging for dementia in clinical practice. BMC Med2015; 13: 163–2–163–2.
11.
WatsonJLRyanLSilverbergN, et al.Obstacles and opportunities in Alzheimer’s clinical trial recruitment. Health Affairs (Project Hope)2014; 33: 574–579.
12.
ChetelatGLa JoieRVillainN, et al.Amyloid imaging in cognitively normal individuals, at-risk populations and preclinical Alzheimer’s disease. Neuroimage Clin2013; 2: 356–365.
13.
WitteMMFosterNLFleisherAS, et al.Clinical use of amyloid-positron emission tomography neuroimaging: practical and bioethical considerations. Alzheimer’s Dementia: Diagnos Assess Dis Monitor2015; 1: 358–367.
14.
MielkeMMWisteHJWeigandSD, et al.Indicators of amyloid burden in a population-based study of cognitively normal elderly. Neurology2012; 79: 1570–1577.
15.
InselPSPalmqvistSMackinRS, et al.Assessing risk for preclinical β-amyloid pathology with APOE, cognitive, and demographic information. Alzheimer’s Dementia: Diagnos Assess Dis Monitor2016; 4: 76–84.
16.
TosunDJoshiSWeinerMW. Neuroimaging predictors of brain amyloidosis in mild cognitive impairment. Ann Neurol2013; 74: 188–198.
17.
ApostolovaLGHwangKSAvilaD, et al.Brain amyloidosis ascertainment from cognitive, imaging, and peripheral blood protein measures. Neurology2015; 84: 729–737.
18.
ten KateMRedolfiAPeiraE, et al.MRI predictors of amyloid pathology: results from the EMIF-AD Multimodal Biomarker Discovery study. Alzheimer’s Res Ther2018; 10: 100–100.
19.
FosterJCTaylorJMRubergSJ. Subgroup identification from randomized clinical trial data. Stat Med2011; 30: 2867–2880.
20.
QianMMurphySA. Performance guarantees for individualized treatment rules. Ann Stat2011; 39: 1180–1210.
21.
ZhaoYZengDRushAJ, et al.Estimating individualized treatment rules using outcome weighted learning. J Am Stat Assoc2012; 107: 1106–1118.
22.
MinskerSZhaoY-QChengG. Active clinical trials for personalized medicine. J Am Stat Assoc2016; 111: 875–887.
23.
SatlinAWangJLogovinskyV, et al.Design of a Bayesian adaptive phase 2 proof-of-concept trial for BAN2401, a putative disease-modifying monoclonal antibody for the treatment of Alzheimer’s disease. Alzheimer’s & Dementia: Translat Res Clin Intervent2016; 2: 1–12.
24.
Ansart M, Epelbaum S, Gagliardi G, et al. Prediction of amyloidosis from neuropsychological and mri data for cost effective inclusion of pre-symptomatic subjects in clinical trials. In: Cardoso J, Arbel T, Carneiro G, et al. (eds) Deep learning in medical image analysis and multimodal learning for clinical decision support. Springer, 2017, pp.357–364.
25.
ChupinMHammersALiuRSN, et al.Automatic segmentation of the hippocampus and the amygdala driven by hybrid constraints: method and validation. NeuroImage2009; 46: 749–761.
26.
ClarkCMPontecorvoMJBeachTG, et al.Cerebral PET with florbetapir compared with neuropathology at autopsy for detection of neuritic amyloid-β plaques: a prospective cohort study. Lancet Neurol2012; 11: 669–678.
27.
ShawLMVandersticheleHKnapik-CzajkaM, et al.Cerebrospinal fluid biomarker signature in Alzheimer’s disease neuroimaging initiative subjects. Ann Neurol2009; 65: 403–413.
28.
BreimanL. Random forests. Mach Learn2001; 45: 5–32.
29.
FriedmanJHastieTTibshiraniR. Regularization paths for generalized linear models via coordinate descent. J Stat Software2010; 33: 1–1.
30.
MullerKRMikaSRatschG, et al.An introduction to kernel-based learning algorithms. IEEE Transact Neural Networks2001; 12: 181–201.
31.
FriedmanJHastieTTibshiraniR. Additive logistic regression: a statistical view of boosting. Ann Stat2000; 28: 337–407.
32.
FriedmanJ. Greedy function approximation: a gradient boosting machine. Ann Stat2001; 29: 1189–1232.
33.
HyvarinenA. Fast and robust fixed-point algorithms for independent component analysis. IEEE Transact Neural Network1999; 10: 626–634.
34.
LopezCLCaputoALiuF, et al.The Alzheimer’s prevention initiative generation program: Evaluating cnp520 efficacy in the prevention of Alzheimer’s disease. J Prevent Alzheimer’s Dis2017; 4: 242–246.
35.
WolpertDHMacreadyWG. No free lunch theorems for optimization. IEEE Transact Evolution Comput1997; 1: 16–16.
36.
WolpertDHThe supervised learning no-free-lunch theorems. In: RoyRKöppenMOvaskaS, et al.(eds). Soft computing and industryLondon: Springer, pp. 25–42.
37.
HughesG. On the mean accuracy of statistical pattern recognizers. IEEE Transact Inform Theor1968; 14: 55–63.
38.
DomingosP. A few useful things to know about machine learning. Commun ACM2012; 55: 78–78.
39.
BatemanRJXiongCBenzingerTL, et al.Clinical and biomarker changes in dominantly inherited Alzheimer’s disease. New Engl J Med2012; 367: 795–804.
40.
AmievaHJacqmin-GaddaHOrgogozoJ-M, et al.The 9 year cognitive decline before dementia of the Alzheimer type: a prospective population-based study. Brain2005; 128: 1093–1101.
41.
AmievaHMokriHLe GoffM, et al.Compensatory mechanisms in higher-educated subjects with Alzheimer’s disease: a study of 20 years of cognitive decline. Brain2014; 137: 1167–1175.
42.
JackCRBennettDABlennowK, et al.NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimer’s Dementia2018; 14: 535–562.
43.
SperlingRAAisenPSBeckettLA, et al.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 Dementia: J Alzheimer’s Assoc2011; 7: 280–292.
44.
JackCRWisteHJWeigandSD, et al.Different definitions of neurodegeneration produce similar amyloid/neurodegeneration biomarker group findings. Brain2015; 138: 3747–3759.
45.
ArbabshiraniMRPlisSSuiJ, et al.Single subject prediction of brain disorders in neuroimaging: promises and pitfalls. NeuroImage2017; 145: 137–165.
46.
FrankeKZieglerGKlöppelS, et al.Estimating the age of healthy subjects from T1-weighted MRI scans using kernel methods: exploring the influence of various parameters. NeuroImage2010; 50: 883–892.
47.
EpelbaumSGenthonRCavedoE, et al.Preclinical Alzheimer’s disease: a systematic review of the cohorts underlying the concept. Alzheimer’s Dementia: J Alzheimer’s Assoc2017; 13: 454–467.
48.
NakamuraAKanekoNVillemagneVL, et al.High performance plasma amyloid-β biomarkers for Alzheimer’s disease. Nature2018; 554: 249–254.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
