Profiling Plasma Biomarkers, Particularly pTau217 and pTau217/Aβ42, and Their Relation to Cognition in Memory Clinic Patients

Abstract

Alzheimer's disease (AD) is characterized by brain protein depositions, impaired synaptic transmission, and progressive cognitive decline. This clinical study, conducted at the tertiary memory clinic of Karolinska University Hospital in Stockholm, evaluates plasma pTau217 (in comparison to other plasma biomarkers) as a non-invasive marker for predicting brain amyloid load and cognitive impairment. Uniquely, it integrates plasma biomarkers with cognitive profiling and amyloid PET to assess diagnostic utility across disease stages in a real-world memory clinic setting. A total of 122 patients underwent extensive clinical examinations, including CSF analysis (used here for clinical diagnosis only), CT/MRI, neuropsychological (NP) testing (n = 80) and blood biomarker measurements. Prior to PET imaging, 74 patients were diagnosed with MCI among other diagnoses (AD, other dementia, no dementia). Following PET, patients were reclassified into diagnostic groups: MCI Aβ− (n = 29), MCI Aβ+ (n = 19), AD (n = 51), other dementias (n = 11). ROC analysis evaluated the ability of plasma biomarkers to predict Aβ-PET positivity. NP test z-scores were reduced into principal components (PCs) using PCA. Plasma pTau217 and pTau217/Aβ42 ratio were elevated in Aβ+ patients compared to MCI Aβ-patients. The ratio distinguished MCI Aβ+ from AD and, together with pTau217, showed the highest predictive value for Aβ positivity in the MCI group among the biomarkers analyzed (AUC 92.8% and 91.4%). Plasma pTau217/Aβ42 ratio was associated with principal component PC2 (“memory encoding and recall”) in MCI Aβ+ (ρ =0.64, p=0.01) and negatively correlated with RAVL retrieval (PC2) in the same group (ρ =−0.57 and −0.6, p=0.028 and 0.017, respectively). Additionally, pTau217 correlated with the “Information” z-score (PC4) in both AD (ρ = −0.50, p = 0.005) and MCI Aβ+ (ρ = 0.53, p = 0.042). Plasma pTau217/Aβ42 might be a valuable predictor of brain amyloid pathology and a potential marker of domain-specific cognitive impairment in AD.