CSF biomarkers predict dementia risk in PD patients

By Lucy Piper, Senior medwireNews Reporter

Early analysis of cerebrospinal fluid (CSF) could help diagnose parkinsonian disorders and enhance the prediction of dementia in Parkinson’s disease (PD) patients, study results indicate.

Specifically, the presence of high CSF levels of neurofilament light chain protein (NFL) and heart fatty acid-binding protein (HFABP) and low levels of β-amyloid (Aβ)1-42 predicted the likelihood of dementia accurately enough “to be clinically useful”, say the study researchers.

“Changes in these biomarkers, even at the time of the diagnosis of PD, may alert physicians to a patient’s risk of developing dementia”, they add.

As published in JAMA Neurology, David Bäckström (Umeå University, Sweden) and colleagues quantified the cerebrospinal fluid concentrations of a panel of biomarkers in 128 patients with new-onset PD or related parkinsonian disorders.

At baseline, the results showed distinct CSF patterns in 104 of the participants with PD compared with 13 with progressive supranuclear palsy and 30 healthy control individuals, but not relative to 11 patients with multiple system atrophy.

In PD patients, NFL levels were significantly increased compared with controls, while Aβ1-42 levels were slightly lower and the ratio of these biomarkers distinguished PD patients from controls with a diagnostic accuracy of 69%.

These CSF biomarkers also distinguished patients with progressive supranuclear palsy from those with PD. A baseline CSF NFL level above 2020 ng/L diagnosed progressive supranuclear palsy with an accuracy of 82%, a sensitivity of 75% and a specificity of 83%, while the ratio of NFL to Aβ1-42 improved diagnostic accuracy to 87%. At values exceeding 2.3, the ratio distinguished progressive supranuclear palsy from PD with 100% sensitivity and 68% specificity.

These results remained stable when CSF was analysed a year later following dopaminergic treatment.

“Marked elevation in NFL might reflect a more aggressive, accumulating subcortical axonal degeneration in [progressive supranuclear palsy] compared with PD”, the researchers suggest.

“If validated, this difference could serve as a supportive diagnostic criterion for [progressive supranuclear palsy] in early disease”, they say, adding that NFL may also serve as a biomarker for neurodegeneration in patients with the condition.

As well as diagnosing parkinsonian disorders, an early CSF pattern of high NFL and low Aβ1-42, along with high HFABP levels, predicted the risk of dementia in PD patients.

Baseline levels of these three CSF biomarkers correlated with the development of dementia in 35 of 99 PD patients over a follow-up of 5 to 9 years.

Patients with PD who, at baseline, had NFL levels exceeding 1100 ng/L, Aβ1-42 levels below 626 ng/L and HFABP levels above 500 ng/L were a respective 2.6, 2.8 and 2.8 times more likely to develop dementia than other patients.

A combination of the ratio of NFL and HFABP to Aβ1-42 provided the greatest diagnostic accuracy, at 83%, with a baseline ratio exceeding 2.1 associated with a hazard ratio for dementia of 11.8, giving a sensitivity of 90% and a specificity of 71%.

The researchers note that only two of the 25 PD patients with ratio values of 1.0 or below went on to develop dementia.

They also confirm that the predictive ability of the three CSF biomarkers remained after taking into account age, and was only slightly lowered after adjustment for baseline mild cognitive impairment.

Licensed from medwireNews with permission from Springer Healthcare Ltd. ©Springer Healthcare Ltd. All rights reserved. Neither of these parties endorse or recommend any commercial products, services, or equipment.