Normal cerebrospinal fluid concentrations of PDGFRβ in patients with cerebral amyloid angiopathy and Alzheimer's disease

Abstract

Background: Cerebrospinal fluid (CSF) platelet‐derived growth factor receptor‐β (PDGFRβ) has been proposed as a biomarker of blood–brain barrier (BBB) breakdown. We studied PDGFRβ levels as a biomarker for cerebral amyloid angiopathy (CAA), amnestic mild cognitive impairment (aMCI), or Alzheimer's disease (AD). Methods: CSF PDGFRβ levels were quantified by enzyme‐linked immunosorbent assay in patients with CAA, patients with aMCI/AD, and in matched controls. In aMCI/AD we evaluated CSF PDGFRβ both by clinical phenotype and by using the AT(N) biomarker classification system defined by CSF amyloid (A), tau (T), and neurodegeneration (N) biomarkers. Results: PDGFRβ levels were similar in CAA patients and controls (P = .78) and in aMCI/AD clinical phenotype and controls (P = .91). aMCI/AD patients with an AD+ biomarker profile (A+T+[N+]) had increased PDGFRβ levels compared to (A–T–[N–]) controls (P = .006). Conclusion: Our findings indicate that PDGFRβ levels are associated with an AD+ biomarker profile but are not a suitable biomarker for CAA or aMCI/AD clinical syndrome.

Associations of CSF PDGFRβ With Aging, Blood-Brain Barrier Damage, Neuroinflammation, and Alzheimer Disease Pathologic Changes

“…In contrast to our findings, increases in CSF PDGFR concentrations have been observed in AD defined clinically or by A/T/N classification 3,9,10,24,30 . The lack of association between CSF PDGFR and A status and A or tau biomarkers was observed previously 3,5,9,10,24 , although one study showed that A burden modulated the association of PDGFR with Tau-PET 31 . Other authors also did not find an association between PDGFR and small vessel disease in cerebral amyloid angiopathy (CAA) subjects 9 .…”

“…The lack of association between CSF PDGFR and A status and A or tau biomarkers was observed previously 3,5,9,10,24 , although one study showed that A burden modulated the association of PDGFR with Tau-PET 31 . Other authors also did not find an association between PDGFR and small vessel disease in cerebral amyloid angiopathy (CAA) subjects 9 . The existing literature has important differences from our study that need to be considered.…”

“…Method-wise, some of these studies used a western blot method for detection of PDGFR in CSF instead of ELISA, which might have led to lower accuracy in the measurements 5,6,24,31 . The studies where ELISA was used had a smaller sample size than ours 3,9,10,25 . In the only study that compared the western blot and ELISA methods in parallel, the authors suggest that the two techniques measure different species of PDGFR, which might have led to discrepancies in the results between different studies 9 .…”

“…Ours is the largest PDGFR clinical study to date and was conducted in a cohort characterized with not only CSF but also imaging measures. Previously, clinical groups were mostly defined based on clinical diagnosis, and the only differences in CSF PDGFR in groups defined by biomarkers were between A+/T+/N+ and A-/T-/N-(i.e., a difference was only seen when amyloid, tau and neurodegeneration CSF biomarkers were pathological, but not when only core AD biomarkers were abnormal) or within cohorts defined by A/T/N that only included preclinical AD 3,5,9,10,24 . Most importantly, this is one of the few and the largest study using PET imaging and not only CSF biomarkers.…”

“…that BBB integrity loss (measured at dynamic contrast-enhanced MRI [DCE-MRI]) was agedependent and correlated with CSF levels of PDGFR levels, overall investigations in clinical cohorts are few, biased by a small samples size and often reporting conflicting results 24 . For instance, some (but not all) studies have shown correlations of PDGFR with age as well as with QAlb 3,5,9,10,24,25 . Here we report that in a large cohort of well-characterized participants, older age was consistently associated with higher CSF levels of PDGFR and that the association was unaffected by clinical diagnosis and possible concomitant AD pathology.…”