| Abstract: | ![]()
The pathophysiology of negative affect states in older adults is complex, and ahost of central nervous system and peripheral systemic mechanisms may playprimary or contributing roles. We conducted an unbiased analysis of 146 plasmaanalytes in a multiplex biochemical biomarker study in relation to number ofdepressive symptoms endorsed by 566 participants in the Alzheimer''s DiseaseNeuroimaging Initiative (ADNI) at their baseline and 1-year assessments.Analytes that were most highly associated with depressive symptoms includedhepatocyte growth factor, insulin polypeptides, pregnancy-associated plasmaprotein-A and vascular endothelial growth factor. Separate regression modelsassessed contributions of past history of psychiatric illness, antidepressant orother psychotropic medicine, apolipoprotein E genotype, body mass index, serumglucose and cerebrospinal fluid (CSF) τ and amyloid levels, and none ofthese values significantly attenuated the main effects of the candidate analytelevels for depressive symptoms score. Ensemble machine learning with RandomForests found good accuracy (∼80%) in classifying groups with andwithout depressive symptoms. These data begin to identify biochemical biomarkersof depressive symptoms in older adults that may be useful in investigations ofpathophysiological mechanisms of depression in aging and neurodegenerativedementias and as targets of novel treatment approaches. |
