Impact of demographics and comorbid conditions on plasma biomarkers concentrations and their diagnostic accuracy in a memory clinic cohort.

Plasma biomarkers have emerged as promising tools for identifying amyloid beta (Aβ) pathology. Before implementation in routine clinical practice, confounding factors modifying their concentration beyond neurodegenerative diseases should be identified. We studied the association of a comprehensive list of demographics, comorbidities, medication and laboratory parameters with plasma p-tau181, glial fibrillary acidic protein (GFAP) and neurofilament light chain (NfL) on a prospective memory clinic cohort and studied their impact on diagnostic accuracy for discriminating CSF/amyloid PET-defined Aβ status. Three hundred sixty patients (mean age 66.5 years, 55% females, 53% Aβ positive) were included. Sex, age and Aβ status-adjusted models showed that only estimated glomerular filtration rate (eGFR, standardized β -0.115 [-0.192 to -0.035], p = 0.005) was associated with p-tau181 levels, although with a much smaller effect than Aβ status (0.685 [0.607-0.763], p < 0.001). Age, sex, body mass index (BMI), Charlson comorbidity index (CCI) and eGFR significantly modified GFAP concentration. Age, blood volume (BV) and eGFR were associated with NfL levels. p-tau181 predicted Aβ status with 87% sensitivity and specificity with no relevant increase in diagnostic performance by adding any of the confounding factors. Using two cut-offs, plasma p-tau181 could have spared 62% of amyloid-PET/CSF testing. Excluding patients with chronic kidney disease did not change the proposed cut-offs nor the diagnostic performance. In conclusion, in a memory clinic cohort, age, sex, eGFR, BMI, BV and CCI slightly modified plasma p-tau181, GFAP and NfL concentrations but their impact on the diagnostic accuracy of plasma biomarkers for Aβ status discrimination was minimal.