Kostadinova : Associations of GFAP, NfL, Aβ42-40, and pTau231 with global cognition in LEADS
Associations of GFAP, NfL, Aβ42/40, and pTau231 with global cognition in LEADS
Ralitsa V. Kostadinova, Dustin B. Hammers, Paige E. Logan, Mohit K. Manchella, Sára Nemes, Anne M. Fagan, Tatiana M. Foroud, Kaj Blennow, Henrik Zetterberg, Joel H. Kramer, Paul S. Aisen, Maria C. Carrillo, Gil D. Rabinovici, Brad C. Dickerson, Liana G. Apostolova, Jeffrey L. Dage
Alzheimer’s disease (AD) biomarkers have been used to identify the presence of pathology. While previous research has evaluated plasma pTau231 and its associations with cognition, little to none of this research has focused on early-onset AD (EOAD) populations. Here we investigate how select plasma biomarkers are associated with global cognitive measures in early-onset cognitive impairment.
The current sample included 367 Longitudinal Early-Onset AD Study (LEADS) participants (aged 41 to 65) categorized as amyloid PET-positive EOAD, amyloid PET-negative EOnonAD, or cognitively normal (CN). Each participant had baseline global cognitive (Mini-Mental State Examination [MMSE], Montreal Cognitive Assessment [MoCA], Clinical Dementia Rating Scale sum of boxes [CDR-SB], and ADAS-Cog13) and plasma biomarker assessments (Simoa-HDx N4PE kit: Aβ42/40, phosphorylated Tau [pTau231], Neurofilament light protein [NfL], and glial fibrillary acidic protein [GFAP]). Partial correlations were used to check for associations with cognitive performance controlling for age, sex, and years of education. Fisher r-to-z transformations were conducted to compare the performance across diagnostic groups.
Partial correlations in the pooled sample showed moderate associations between cognition and plasma pTau231, GFAP and NfL (r=.42-.50, p<.001) and weaker associations with Aβ42/40 (r=.25-.33, p<.001). When split into diagnostic groups, the NfL and GFAP correlations were significant in EOAD and EOnonAD only. The pTau231 correlations were significant only in EOAD. Aβ42/40 correlations were non-significant within specific diagnostic groups. No differences were observed in the magnitude of the cognitive and biomarker associations between EOAD and EOnonAD samples (ps>.05).
As expected, the neurodegenerative biomarkers pTau231 and NfL showed stronger association with cognition compared to the marker for brain amyloidosis (Aβ42/40). The nonspecific markers for neurodegeneration (NfL) and brain astrogliosis (GFAP) showed associations in both EOAD and EOnonAD while the markers specific to AD were only significant in EOAD. Plasma biomarkers show great promise for AD diagnosis and monitoring.