PT - JOURNAL ARTICLE AU - Hahn, Sage AU - Mackey, Scott AU - Cousijn, Janna AU - Foxe, John J. AU - Hester, Robert AU - Hutchinson, Kent AU - Korucuoglu, Ozlem AU - London, Edythe AU - Lorenzetti, Valentina AU - Luijten, Maartje AU - Momenan, Reza AU - Orr, Catherine AU - Paulus, Martin AU - Schmaal, Lianne AU - Sinha, Rajita AU - Sjoerds, Zsuzsika AU - Stein, Dan J. AU - Stein, Elliot AU - van Holst, Ruth J. AU - Veltman, Dick AU - Wiers, Reinout W. AU - Yucel, Murat AU - Thompson, Paul M. AU - Conrod, Patricia AU - Allgaier, Nicholas AU - Garavan, Hugh TI - Predicting Alcohol Dependence from Multi-Site Brain Structural Measures AID - 10.1101/2020.01.17.20016873 DP - 2020 Jan 01 TA - medRxiv PG - 2020.01.17.20016873 4099 - http://medrxiv.org/content/early/2020/01/23/2020.01.17.20016873.short 4100 - http://medrxiv.org/content/early/2020/01/23/2020.01.17.20016873.full AB - Background The search for neuroimaging biomarkers of alcohol use disorder (AUD) has primarily been restricted to significance testing in small datasets of low diversity. To identify neurobiological markers beyond individual differences, it may be useful to develop classification models for AUD. The ever-increasing quantity of neuroimaging data demands methods that are robust to the complexities of multi-site designs and are generalizable to data from new scanners.Methods This study represents a mega-analysis of previously published datasets from 2,034 AUD and comparison participants spanning 27 sites, coordinated by the ENIGMA Addiction Working Group. Data were grouped into a training set including 1,652 participants (692 AUD, 24 sites), and test set with 382 participants (146 AUD, 3 sites). A battery of machine learning classifiers was evaluated using repeated random cross-validation (CV) and leave-site-out CV. Area under the receiver operating characteristic curve (AUC) was our base metric of performance.Results Multi-objective evolutionary search was conducted to identify sparse, generalizable, and high performing subsets of brain measurements. Cortical thickness in the left superior frontal gyrus and right lateral orbitofrontal cortex, cortical surface area in the right transverse temporal gyrus, and left putamen volume, appeared most frequently across searches. Restricting a regularized logistic regression model to these four features yielded a test-set AUC of .768.Conclusions Developing classification models on multi-site data with varied underlying class distributions poses unique challenges. Supplementing datasets with controls from new sites and performing feature selection increases generalizability. Four features identified by evolutionary search may serve as specific biomarkers for individuals with current AUD.Competing Interest StatementDr. Sinha has served on the scientific advisory board of Embera Neuro-therapeutics. DJS has received research grants and/or consultancy honoraria from Lundbeck and Sun. Prof. Yücel has received funding from several law firms in relation to expert witness reports. PT received grant support from Biogen, Inc., for research unrelated to this study. All other authors have no financial disclosures.Funding StatementThis work was made possible by NIDA grant R01DA047119 awarded to Dr. Garavan. Sage Hahn was supported by NIDA grant T32DA043593. Data collection was made possible through the following grants: Dr. Korucuoglu received support for the Neuro-ADAPT study from VICI grant 453.08.01 from the Netherlands Organization for Scientific Research (NWO), awarded to Reinout W. Wiers. Drs. Schmaal and Veltman received funding from Netherlands Organization for Health Research and Development (ZonMW) grant 31160003 from NWO. Drs. Sjoerds and Veltman received funding from ZonMW grant 31160004 from NWO. Dr. van Holst received funding from ZonMW grant 91676084 from NWO. Dr. Luijten and Veltman received funding from VIDI grant 016.08.322 from NWO, awarded to Dr. Cousijn received funding for the Cannabis Prospective study from ZonMW grant 31180002 from NWO. Drs. Garavan and Foxe received funds from NIDA grant R01-DA014100. Dr. London was supported by NIDA grant R01 DA020726, the Thomas P. and Katherine K. Pike Chair in Addiction Studies, the Endowment From the Marjorie Greene Family Trust, and UCLA contract 20063287 with Philip Morris USA. Data collection by Dr. Momenan was supported by the Intramural Clinical and Biological Research Program funding ZIA AA000125-04 DICB (Clinical NeuroImaging Research Core to RM) of the National Institute on Alcohol Abuse and Alcoholism (NIAAA). Dr. Paulus received funding from NIMH grant R01 DA018307. Dr. Stein was supported by the Intramural Research Program of NIDA and NIH. Dr. Sinha received funds from NIDA (PL30–1DA024859–01), the NIH National Center for Research Resources (UL1-RR24925–01), and NIAAA (R01- AA013892). Prof. Yücel was supported by National Health and Medical Research Council Fellowship 1117188 and the David Winston Turner Endowment Fund. Dr. Thompson was supported in part by NIH grant U54 EB020403.Author DeclarationsAll relevant ethical guidelines have been followed; any necessary IRB and/or ethics committee approvals have been obtained and details of the IRB/oversight body are included in the manuscript.YesAll necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived.YesI 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).YesI have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.YesData is not publically available but accessed through enigma addiction data pooling.