IC‐P‐108: WHITE MATTER HYPERINTENSITIES IN RELATION TO PATTERNS OF ACCELERATED BRAIN AGING, AD‐LIKE ATROPHY AND AMYLOID BURDEN: RESULTS FROM THE ISTAGING CONSORTIUM ON MACHINE LEARNING AND LARGE‐SCALE IMAGING ANALYTICS

Pattern of Atrophy for Recognition of Brain Aging (SPARE-BA) index. Risk factors of interest included hypertension, obesity, depression, cognitive activity (e.g. book/newspaper reading), physical activity, smoking, diet, alcohol consumption and diabetes. Multiple linear regression tested the association between SES and brain volume. Sobel mediation analyses determined if this association was mediated by modifiable risk factors. All models were age, sex, and race adjusted. Results: Aim i) regression analysis revealed SES was positively associated with brain volume (Figure 1; b1⁄4.118 SE1⁄4.039; p1⁄4.002). Aim ii) from the nine risk factors investigated, smoking and depression were identified as two potential mediators due to their significant relationships with SES and brain volume (all p<.05). A Sobel mediation analysis determined smoking significantly mediated the relationship between SES and brain volume (Figure 2; p1⁄4.016). With respect to brain volume, smoking accounted for a significant amount of variance (b1⁄4-.116 SE1⁄4.065; p1⁄4.01) that was previously attributed to SES. Conclusions: Targeting SES disparities could be a promising means to improve brain health and may decrease vulnerability for AD. Notably, smoking appears to partly explain the adverse effects of low SES on brain volume. IC-P-108 WHITE MATTER HYPERINTENSITIES IN RELATION TO PATTERNS OF ACCELERATED BRAIN AGING, AD-LIKE ATROPHYAND AMYLOID BURDEN: RESULTS FROM THE ISTAGING CONSORTIUM ON MACHINE LEARNING AND LARGE-SCALE IMAGING ANALYTICS MohamadHabes, Raymond Pomponio, Haochang Shou, Jimit Doshi, Aristeidis Sotiras, Guray Erus, Lenore J. Launer, Elizabeth Mamourian, Murat Bilgel, Kristine Yaffe, Dhivya Srinivasan, Mark A. Espeland, Ilya M. Nasrallah, Christopher C. Rowe, Henry Voelzke, Sterling C. Johnson, Marilyn S. Albert, Nick Bryan, Hans J. Grabe, Susan M. Resnick, David A. Wolk, Christos Davatzikos, Department of Neurology and Penn Memory Center, University of Pennsylvania, Philadelphia, PA, USA; University of Pennsylvania, Philadelphia, PA, USA; Center for Biomedical Image Computing and Analytics and Department of Radiology, University of Pennsylvania, Philadelphia, PA, USA; Center for Biomedical Image Computing and Analytics, University of Pennsylvania, Philadelphia, PA, USA; National Institute on Aging, Bethesda, MD, USA; Center for Biomedical Image Computing and Analytics and Department of Radiology, University of Pennsylvania, Philadelphia, PA, USA; National Institute on Aging, NIH, Baltimore, MD, USA; Department of Epidemiology and Biostatistics, University of California San Francisco, San Francisco, CA, USA; Wake Forest School of Medicine, Winston-Salem, NC, USA; Department of Molecular Imaging and Therapy, Centre for PET, Austin Health, Heidelberg, Australia; Institute for Community Medicine, University Medicine Greifswald, Greifswald, Germany; Wisconsin Alzheimer’s Institute, University of Wisconsin School of Medicine and Public Health, Madison, WI, USA; Johns Hopkins University School of Medicine, Baltimore, MD, USA; Department of Radiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA; University of Texas at Austin, Austin, TX, USA; Department of Psychiatry, University Medicine Greifswald and German Center for Neurodegenerative Disease (DZNE) Rostock/ Greifswald, Greifswald, Germany; Penn Memory Center, University of Pennsylvania, Philadelphia, PA, USA. Contact e-mail: habesm@uphs.upenn.edu