Premium
P3‐149: Diagnosis of Diabetes Mellitus Does Not Differentially Impact the Proteomic Profile of Mild Cognitive Impairment Among Mexican Americans
Author(s) -
Edwards Melissa,
Hall James R.,
Johnson Leigh Ann,
O'Bryant Sid
Publication year - 2016
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2016.06.1809
Subject(s) - medicine , diabetes mellitus , biomarker , oncology , neuropsychology , dementia , interleukin 18 , gerontology , cohort , disease , clinical psychology , cognition , endocrinology , psychiatry , biology , cytokine , biochemistry
Background: Methods for establishing early detection of Mild Cognitive Impairment (MCI) have supported the use of blood based biomarkers. Among Mexican Americans, prior work has shown the biomarker profile for MCI to differ from that of the profile for Alzheimer’s disease(AD), which is comprised of predominantly metabolic proteins. As Mexican Americans experience higher rates of metabolic disorders such as diabetes mellitus (DM), the impact of metabolic processes on cognitive changes requires further exploration. This study sought to address this need and identify among Mexican Americans the impact of DM on the proteomic profile of MCI. Methods:Data were analyzed on 414 cases (DM n1⁄4 175; without DM n1⁄4 239) from the HABLE (Health and Aging Brain among Latino Elders) cohort, an epidemiological study of aging among Mexican Americans. Biobank analysis of CRP, SAA, sICAM1, sVCAM1, A2M, B2M, FVII, TNC, Eotaxin3, IL5, IL6, IL7, IL10, IL18, TARC, TNFa, FABP, I309, PPY, and THPO were conducted via electrochemiluminescence (ECL) using the MESO Scale Discovery Platform. Random forest analyses were conducted with diagnosis (MCI vs NC) as the outcome variable and serum biomarkers as the predictor variables split by DM diagnosis. Models further included select demographics (age, gender, education) and one neuropsychological measure (FAS) to enhance diagnostic accuracy, utilizing the Molecular Neuropsychology approach. Results: Among those with DM, seven biomarkers (IL10, TNFa, I309, thrombopoietin, IL18, FABP, and TARC) plus age and one measure of cognitive functioning (FAS) predicted MCI with an accuracy level of 83% (sensitivity [SN]1⁄40.96; specificity [SP]1⁄40.51). When examining those without DM, six biomarkers including TNFa, I309, IL10, thrombopoietin, TARC, and FABP out of the seven identified in the proteomic profile among those with DM, plus education, age, and one measure of cognitive functioning (FAS) predicted MCI with an overall accuracy level of 90% (SN1⁄40.96; SP1⁄40.63). Conclusions:No difference was identified among the proteomic profiles of those with and without DM who also experience MCI. The results suggest that the impact of impaired metabolic processes during the MCI state does not significantly impact the overall biological profile for Mexican-Americans. Additional work needs to examine if a similar pattern is present among the AD proteomic profile.