Multiomics machine learning identifies inflammation molecular pathways in prodromal Alzheimer’s Disease

Mild Cognitive Impairment (MCI) is a phase that can precede Alzheimer’s Disease (AD). To better understand the molecular mechanisms underlying conversion from MCI to AD, we applied a battery of machine learning algorithms on 800 samples from the EMIF-AD MBD study. The cohort comprised participants diagnosed as 230 normal cognition (NC), 386 MCI (with longitudinal data on AD conversion or remaining stable) and 184 AD-type dementia. Data consisted of metabolites (n=540) and proteins (n=3630) measured in plasma coupled to clinical data (n=26). Multiclass models selected oleamide, MMSE and the priority language as the most confident features while MCI conversion models selected pTau, tTau and JPH3, CFP, SNCA and PI15 proteins. These proteins selected for MCI conversion have been previously associated with AD-related phenotype. Oleamide, a possible anti-inflammatory, prompted in-vitro experiments in rodent microglia. The results demonstrated that disease-associated microglia synthesize oleamide which were excreted in vesicles. In addition, plasma vesicles extracted from participants with AD showed elevated oleamide levels compared to controls (P<0.05). This study uncovered MCI conversion pathways that involve inflammation, neuronal regulation and protein degradation. ![Figure][1] ### Competing Interest Statement SL is named as an inventor on biomarker intellectual property protected by Proteome Sciences and Kings College London unrelated to the current study and within the past five years has advised for Optum labs, Merck, SomaLogic and been the recipient of funding from AstraZeneca and other companies via the IMI funding scheme. HZ has served at scientific advisory boards and/or as a consultant for Abbvie, Alector, ALZPath, Annexon, Apellis, Artery Therapeutics, AZTherapies, CogRx, Denali, Eisai, Nervgen, Novo Nordisk, Pinteon Therapeutics, Red Abbey Labs, reMYND, Passage Bio, Roche, Samumed, Siemens Healthineers, Triplet Therapeutics, and Wave, has given lectures in symposia sponsored by Cellectricon, Fujirebio, Alzecure, Biogen, and Roche, and is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program (outside submitted work). AL has served at scientific advisory boards of Fujirebio Europe, Eli Lilly, Novartis, Nutricia and Otsuka and is the inventor of a patent on synaptic markers in CSF (all unrelated to this study). JP has served at scientific advisory boards of Fujirebio Europe, Eli Lilly and Nestle Institute of Health Sciences, all unrelated to this study. SE has received unrestricted research grants from Janssen Pharmaceutica and ADx Neurosciences and has served at scientific advisory boards of Biogen, Eisai, Novartis, Nutricia / Danone, all unrelated to this study. FB is a steering committee or iDMC member for Biogen, Merck, Roche, EISAI and Prothena. Consultant for Roche, Biogen, Merck, IXICO, Jansen, Combinostics. Research agreements with Merck, Biogen, GE Healthcare, Roche. Co-founder and shareholder of Queen Square Analytics LTD, all unrelated to this study. CLQ has received funding related to this study from Pfizer and other companies via the IMI funding scheme, unrelated to this study and in the past five years, has been the recipient of funding from Novo Nordisk, served as scientific advisory board in Fondation Alzheimer's, Institute Pasteur, Academy of Finland. ### Funding Statement This research was conducted as part of the EMIF-AD MBD project which has received support from the Innovative Medicines Initiative Joint Undertaking under EMIF grant agreement no 115372, resources of which are composed of financial contribution from the European Union’s Seventh Framework Programme (FP7/2007-2013) and EFPIA companies’ in-kind contribution. The DESCRIPA study was funded by the European Commission within the 5th framework program (QLRT-2001-2455). The EDAR study was funded by the European Commission within the 5th framework program (contract # 37670). The Leuven cohort was funded by the Stichting voor Alzheimer Onderzoek (grant numbers #11020, #13007 and #15005). RV is a senior clinical investigator of the Flemish Research Foundation (FWO). The San Sebastian GAP study is partially funded by the Department of Health of the Basque Government (allocation 17.0.1.08.12.0000.2.454.01.41142.001.H). We acknowledge the contribution of the personnel of the Genomic Service Facility at the VIB-U Antwerp Center for Molecular Neurology. The research at VIB-CMN is funded in part by the University of Antwerp Research Fund. HZ is a Wallenberg Scholar supported by grants from the Swedish Research Council (#2018-02532), the European Research Council (#681712), Swedish State Support for Clinical Research (#ALFGBG-720931), the Alzheimer Drug Discovery Foundation (ADDF), USA (#201809-2016862), and the UK Dementia Research Institute at UCL. FB is supported by the NIHR biomedical research centre at UCLH. LS is funded by the Virtual Brain Cloud from European commission (grant no. H2020-SC1-DTH-2018-1). R.G. was supported by the National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London. This paper represents independent research part-funded by the National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health and Social Care. JX and CLQ thank Lundbeck Fonden for the support. Finally, this publication has also been made possible by the support of the Fundación Séneca-Agencia de Ciencia y Tecnología de la Región de Murcia (Spain), which finances the PhD of Alicia Gómez-Pascual (21259/FPI/19. Fundación Séneca. Región de Murcia, Spain). ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: The study obtained ethical approval from the local medical ethical committee in each centre involved in the EMIF-AD cohort (The medical ethical committee of the VU University Medical Center has given approval to the Amsterdam dementia cohort study; The medical ethical committee of the Beta amyloid oligomers in the early diagnosis of AD and as marker for treatment response (EDAR) has given approval to the EDAR cohort study; The medical ethical committee of the PharmaCog has given approval to the PharmaCog cohort study; The medical ethical committee of the Development of screening guidelines and criteria for predementia Alzheimer's disease (DESCRIPA) has given approval to the DESCRIPA cohort study; The medical ethical committee of the University of Antwerp has given approval to the Antwerp cohort study; The medical ethical committee of the Investigacions Biomediques Sant Pau - Hospital de Sant Pau has given approval to the Barcelona Sant Pau cohort study; The medical ethical committee of the Hospital Clinic Barcelona has given approval to the Barcelona Institut d'Investigacions Biomediques August Pi i Sunyer (IDIBAPS) cohort study; The medical ethical committee of the KU Leuven and University Hospitals Leuven has given approval to the Leuven cohort study; The medical ethical committee of the CITA-Alzheimer Foundation has given approval to the San Sebastian GAP cohort study; The medical ethical committee of the University of Gothenburg has given approval to the Gothenburg cohort study; The medical ethical committee of the Lausanne University Hospital has given approval to the Lausanne cohort study). Subjects had already provided written informed consent at the time of inclusion in the cohort for use of data, samples and scans. Research was conducted in accordance with the Declaration of Helsinki and all participants provided written informed consent. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes The generated metabolomic and proteomic data in EMIF is considered sensitive patient data and can therefore not be publicly available in compliance with the European privacy regulations governed by GDPR and according to limitations included in the informed consents signed by the study participants. Data are available upon request by contacting the EMIF-AD data hub steering committee via the academic EMIF-AD lead, Prof. Pieter Jelle Visser, and data access coordinator Dr. Stephanie Vos (s.vos{at}maasticuniversity.nl). Requests should include the following information, name and contact details of the person requesting the data, aims and objectives, study design and methods, requested molecular data and clinical variables, outcomes, timelines and cohorts of interest in the EMIF-AD catalogue (). Requests will be subject to consideration by the steering committees of the cohorts and the management board. Time frame for a response will be within 4 months. Data requests under agreement will be subject to appropriate confidentiality obligations and restrictions. The averaged protein and metabolites plasma levels for each diagnosis group (NL, sMCI, cMCI and NL) have been deposited on GitHub (). * AD : Alzheimer’s Disease MCI : mild cognitive impairment cMCI : MCI converted to Alzheimer’s sMCI : MCI stable MMSE : mini mental state examination NC : normal cognition ML : machine learning LR : logistic regression MLP : multi-layer perceptron network RF : random forest SVM : support vector machine. [1]: pending:yes