Other CSF biomarker candidates
There are a number of other candidates that might be of interest as cognitive biomarkers in PD. Genetic and molecular studies point towards several partly interrelated pathways that may directly or indirectly drive the abnormal protein accumulation and neurodegeneration observed in PD, such as impaired protein degradation, mitochondrial dysfunction, oxidative stress, and neuroinflammation . Given the central pathogenic role of aSyn, and probably Ap, pathology in cognitive impairment related to PD, CSF markers that reflect these pathways appear particularly promising and have been increasingly explored. However, few studies have investigated these markers with specific focus on their relation to cognitive impairment in PD.
Markers of protein degradation
Impaired protein degradation by the ubiquitin-proteosome system, the autophagy-lysosomal pathway, and possibly other proteolytic systems, is considered to contribute to the pathological cascade leading to accumulation of aSyn in PD [13, 60]. Indeed, differences between PD patients and controls in activity levels of several lysosomal enzymes in the CSF have been reported in cross-sectional studies [61, 62]. However, associations with cognitive measures or state were not explored. Another study found significantly reduced levels of neurosin, an enzyme proposed to be involved in the cleavage of extracellular aSyn , in patients with synucleinopathies compared with controls and AD, with the lowest levels observed in subjects with DLB . However, no differences were seen between PD-D patients versus PD without dementia, questioning its utility as cognitive marker in PD.
Markers of Ap production and clearance
CSF markers that reflect the production or clearance of Ap have been extensively studied in AD. Some have had promising results, such as the P-site APP cleaving enzyme-1 (BACE1)  and ne- prilysin , an Ap-degrading protein found at pre-synaptic terminals. Lowered levels of activity of neprilysin in the CSF of patients with DLB and PD-D compared with PD without dementia and normal controls have been reported in one cross-sectional study , but no longitudinal studies have yet been published. Visinin-like protein-1 (VILIP-1), a marker of neuronal injury, has also been shown to predict the rate of cognitive decline in cognitively normal individuals and those with very mild and mild AD [67, 68], but has not been investigated in PD.
Oxidative stress markers
Mitochondrial dysfunction and oxidative stress have been implicated in the pathogenesis of PD. DJ-1 is a multifunctional protein that is involved in mitochondrial regulation and protection against oxidative stress. Studies of DJ-1 in the CSF are inconsistent, showing either reduced  or elevated  levels in PD compared with controls, and have not explored possible associations with cognitive impairment in PD. CSF levels of 8-hydroxy-2'-deoxyguanosine (8-OHdG), a marker of oxidative stress, have been reported by several groups to be elevated in PD. One of these also reported an inverse correlation of this marker with Mini-Mental State Examination scores in PD-D but no significant differences in levels of 8-OHdG in the CSF of PD-D, DLB, and AD patients .
Neuroinflammatory changes including microglial activation are thought to play an important role in many neurodegenerative diseases, including PD . Microglial activation has potential beneficial effects by clearing aggregates of pathological protein and remodelling synapses, but may also damage neurons through the release of cytokines, proteases, and nitric oxide [72, 73]. Hence, inter-individual differences in microglial activation may have an impact on the spread of pathology , and therefore the rate of cognitive decline and risk for dementia in PD. Cross-sectional correlations of several inflammatory CSF markers, such as tumour necrosis factor-alpha (TNF-a)
 and brain-derived neurotrophic factor (BDNF) , with cognition have been described in PD but await further replication and exploration in longitudinal studies.