Associations of circulating c-reactive protein levels with central Alzheimer's disease biomarkers

C-reactive protein (CRP) is well-known marker of inflammation and immune response. Its potential role in Alzheimer's disease (AD) pathophysiology remains unclear, particularly in relation to central AD biomarkers, including beta-amyloid (Aβ), tau, and neurodegeneration.

To investigate the associations between circulating CRP levels and central AD biomarkers-including Aβ deposition, tau, and AD-signature neurodegeneration-in nondemented older adults.

This cross-sectional observational study analyzed data from a Korean Brain Aging Study for Early Diagnosis and Prediction of Alzheimer Disease conducted from 2014 to 2020. A total of 417 nondemented older adults underwent comprehensive evaluations, including blood sampling and multimodal neuroimaging for measuring of Aβ and AD-signature neurodegeneration. A subset of participants (N = 123) also underwent tau positron emission tomography (PET) scan.

The primary outcomes were A/T/N biomarkers of AD, including brain Aβ and tau deposition measured via amyloid and tau PET, as well as AD-signature neurodegeneration measured by fluorodeoxyglucose (FDG)-PET. Associations between CRP levels and these biomarkers were analyzed while adjusting for CRP-decreasing allele scores, as well as other confounders, including age, sex, vascular risk score, body mass index, nonsteroidal anti-inflammatory drug (NSAID) usage, smoking status, and APOE ε4 carrier status.

The mean (SD) age of participants was 70.57 (8.00) years, with 179 (42.9 %) females. Circulating CRP levels showed non-linear associations with A/T/N biomarkers of AD, showing a U-shaped relationship with Aβ and tau deposition and an inverted U-shaped association with neurodegeneration. Threshold effect analyses revealed that CRP was inversely associated with Aβ deposition (B = -0.081; 95 % CI, -0.153 to -0.007; p = 0.031) below 0.63 mg/L, after adjusting for all confounding variables. In contrast, higher CRP levels were associated with lower cerebral glucose metabolism in AD-signature regions, indicative of greater AD-related neurodegeneration, when above 2.15 mg/L (B = -0.056; 95 % CI, -0.112 to -0.001; p= 0.042).

Our study revealed differential associations between circulating CRP levels and central AD biomarkers that varied according to the CRP concentration. Further studies are necessary to elucidate the mechanisms underlying the inverse relationship between circulating CRP and brain Aβ within the clinically normal range, as well as potential aggravating effects of elevated CRP on Aβ-independent neurodegeneration.