Cerebrospinal fluid levels of growth associated protein 43 (GAP‐43) are associated with Alzheimer's disease biomarkers, cognition, and functional changes.

Background: Growth‐associated protein 43 (GAP‐43) is a presynaptic protein involved in synaptic plasticity implicated in Alzheimer's disease (AD). This study investigates cerebrospinal fluid (CSF) levels of GAP‐43 in relation to AD fluid biomarkers and clinical outcomes. To determine how associations change across disease states, interactions with cognitive and amyloid status are examined. Method: Vanderbilt Memory and Aging Project participants free of clinical dementia or stroke (n=153, 72±7 years, 37% mild cognitive impairment (MCI)) underwent fasting lumbar puncture, 3T brain MRI, neuropsychological assessment, and informant‐reported functional activities questionnaire (FAQ). CSF GAP‐43, b‐amyloid1‐42 (Ab1‐42), tau, and phosphorylated‐tau (p‐tau) were analyzed in batch. An AD‐signature imaging composite was calculated by summing cortical thickness measurements from brain regions commonly affected by AD (Schwarz et al., 2016). Linear mixed effects models cross‐sectionally related GAP‐43 to CSF AD biomarkers, structural neuroimaging, cognitive, and FAQ variables adjusting for baseline age, sex, education, race/ethnicity, apolipoprotein E (APOE)‐e4 status, modified Framingham Stroke Risk Profile, and cognitive status. Follow‐up models assessed GAP‐43 x cognitive status (normal cognition (NC), MCI) and GAP‐43 x amyloid positivity interactions on outcomes. Result: In main effect models, higher CSF GAP‐43 was associated with higher CSF tau (p<0.0001) and CSF p‐tau (p<0.0001); worse semantic fluency (p<0.02), visuospatial (p<0.007) and episodic memory performances (p=0.06); and worse FAQ scores (p=0.003). Remaining models were null (p‐values>0.23). GAP‐43 interacted with cognitive status on all CSF biomarkers, AD‐signature composite cortical thickness, visuospatial, executive functioning, and memory performances, and FAQ score (p‐values<0.06). CSF GAP‐43 interacted with amyloid positivity on CSF tau and CSF p‐tau, and memory composite score (p‐values<0.03). Stratified results suggested most associations were slightly stronger in participants with MCI (p‐values<0.02) versus NC and in amyloid positive (p‐values<0.02) versus negative participants. Conclusion: We found higher CSF GAP‐43 levels were associated with tau and p‐tau pathology, AD‐related neurodegeneration on MRI, and worse cognitive and functional scores. Results were more prominent in individuals at increased risk for AD‐type dementia (i.e., participants who were amyloid positive or have MCI). These findings extend prior work by highlighting the importance of synaptic dysfunction in the AD clinical syndrome. [ABSTRACT FROM AUTHOR]