Rapid eye movement (REM) sleep behavior disorder (RBD) is a type of parasomnia characterized by dream enactment behaviors due to a loss of muscle atonia during REM sleep [1-3]. Robust evidence indicates that patients with isolated RBD (iRBD) have a greater risk of developing α-synucleinopathies, including Parkinson disease (PD) and dementia with Lewy bodies (DLB) [4]. The conversion rate from iRBD to overt α-synucleinopathy is approximately 33% at 5 years, 76% at 10 years, and 91% at 14 years [5]. Accordingly, exploring iRBD offers an opportunity to identify the prodromal stages of α-synucleinopathies and to assess the efficacy of early neuroprotective strategies.

While the pathomechanism of α-synucleinopathies remains incompletely understood, a mounting body of evidence suggests a role of neuroinflammation in the development and progression of these diseases. Widespread microglial activation in the brain has been found in PD [6-8] and DLB [9], and increased levels of inflammatory mediators have been documented in both the brain and cerebrospinal fluid of such patients [9-12]. Of note, epidemiological studies have indicated that the occurrence of PD is lower in those who use nonsteroidal anti-inflammatory drugs, thereby supporting the idea of a potential link between neuroinflammation and disease [13]. In support of this hypothesis, a recent study using mouse models has shown that spread of α-synuclein aggregate was preceded by inflammatory responses, and the degree of spreading correlated with the extent of inflammation [14]. In addition, these changes were suppressed by an anti-inflammatory drug [14]. Recently, it was discovered that iRBD patients show activated microglia in the substantia nigra along with reduced nigrostriatal dopaminergic function [15], which provide insight into the role of neuroinflammation in prodromal stages of α-synucleinopathies.

Systemic inflammation has been linked to inflammation within the brain, which can contribute to neurodegeneration [16-18]. More specifically, both α-synuclein-dependent and independent inflammatory responses in the enteric nervous system promote further α-synuclein pathology and augmented circulatory inflammatory signatures, which could potentially cause propagation to the central nervous system through the vagus nerve [19]. Conversely, T cells and other immune cells could extravagate via the blood-brain barrier in response to enhanced inflammation originating from the central nervous system [19]. The ease and feasibility of using blood samples have facilitated multiple investigations evaluating peripheral inflammatory markers, particularly cytokines, in patients with PD and DLB. These studies have frequently reported a significant link between inflammatory markers and disease severity, although not consistently. However, the impact of peripheral inflammatory markers on the development and progression of iRBD remains ambiguous. In this review, we focused on the association between inflammatory cytokines from peripheral blood and α-synucleinopathies. We first discuss previous studies investigating peripheral blood inflammatory cytokines and their associations with clinical outcomes in patients with PD and DLB. We then provide an overview of the current evidence available for the role of peripheral blood inflammatory cytokines in patients with iRBD.

Numerous studies have explored the connections between concentrations of inflammatory cytokines in peripheral blood and PD, but the findings regarding individual cytokines and the relationships between the studies were inconsistent. In a recent meta-analysis of 25 studies, inflammatory cytokine levels in plasma and serum were compared between 1,547 PD patients and 1,107 healthy controls [20]. The findings revealed significantly elevated concentrations of several cytokines in patients with PD, including interleukin (IL)-1β, IL-2, IL-6, IL-10, tumor necrosis factor alpha (TNF-α), C-reactive protein (CRP), and RANTES (regulated upon activation, normal T cell expressed and presumably secreted). However, not all cytokines exhibited such differences, and some other inflammatory cytokines, including interferon gamma (IFN-γ), showed comparable concentrations in the two groups. Furthermore, the cytokines elevated in PD patients in comparison to controls were not exclusively associated with one immune system function. Instead, they included non-specific acute-phase reactants like CRP and proinflammatory cytokines such as IL-1β and IL-2, as well as the anti-inflammatory cytokine IL-10, which plays a more complex role in the immune system [21].

Previous attempts have been made to examine the connection between inflammatory cytokines in serum or plasma and the severity of motor symptoms in PD patients. However, the findings have been contradictory, with both positive and negative results [22-29]. These conflicting observations may be partially attributed to the cross-sectional study design, as only one prior study has employed a longitudinal design [22]. Williams-Gray et al. [22] conducted a longitudinal study that found higher levels of IL-1β, IL-2, IL-10, TNF-α, and CRP, among 10 cytokines tested in PD patients (n = 230) compared with controls (n = 93). Notably, a principal-component (PC) analysis of the serum immune marker data revealed a higher “proinflammatory” component score, mainly influenced by IFN-γ and TNF-α, and lower “anti-inflammatory” component score, mainly influenced by IL-4, IL-13, and IL-12p70, with more rapid motor progression over 36 months in PD patients.

The association between peripheral inflammatory cytokines and non-motor symptoms in PD patients has not been extensively investigated. While several studies have explored how serum or plasma cytokine markers are related to cognitive function in PD patients, their results have been inconsistent, possibly due to the limitations of cross-sectional study design [22,25-31]. Although a longitudinal study by Williams et al. [22] found significant associations between IL-6, TNF-α, and CRP and lower Mini-mental State Examination scores at baseline, these associations were not observed over the follow-up period. A more recent longitudinal study by Kim et al. [31] explored whether serum inflammatory marker profiles, including IL-1β , IL-2, IL-6, IL-10, TNF-α, and high-sensitivity CRP, are associated with the progression of non-motor symptoms in early PD patients. PC analysis showed that only PC3 scores, primarily driven by IL-2 and IL-6, were significantly higher in PD patients compared with healthy controls. Notably, higher PC3 scores in PD patients were associated with faster progression of Non-Motor Symptoms Scale total and mood/apathy domain scores over the 3-year follow-up period. This indicates that elevated peripheral inflammation, mainly associated with IL-2 and IL-6, is linked to the evolution of overall non-motor symptoms, particularly mood symptoms, in the early stages of PD. However, there were no significant associations of PC scores with longitudinal changes in autonomic and cognitive function.

A previous cross-sectional cohort study assessed peripheral immunophenotype in 31 patients with DLB, 31 patients with Alzheimer disease (AD), and 31 healthy controls, in which serum concentrations of IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12, IL-13, TNF-α, and IFN-γ were measured [32]. After adjusting for age and sex, among these cytokines, only IL-6 level was significantly upregulated compared with controls. In addition, although IL-1β was detected only in about half of the DLB patients, the detection rate was remarkably higher in DLB patients compared with the AD patients and controls. However, none of the cytokine concentrations were associated with the severity of DLB-related clinical features. Another cross-sectional study examined relationships of serum IL-6 and TNF-α with cognitive and neuropsychiatric symptoms in 33 probable or possible DLB patients, in which there were modest positive correlations between IL-6 and the Alzheimer’s Disease Assessment Scale Cognitive subscale scores and between TNF-α and the Neuropsychiatric Inventory total score [33]. More importantly, several neuropsychiatric symptoms, including depression, anxiety, sleep disturbance, and eating disturbance, were linked to significantly higher TNF-α level, suggesting that peripheral blood TNF-α may contribute to neuropsychiatric features during the disease course. A more recent study analyzed the concentrations of selected plasma cytokines including IL-4, IL-6, IL-10, TNF-α, IFN-γ, and monocyte chemoattractant protein-1 (MCP-1) in 16 DLB patients, 19 PD patients with dementia, 28 PD patients without dementia, and 19 controls [34]. The levels of TNF-α and IL-6 were significantly higher in the DLB and PD patients with dementia compared with the controls. Furthermore, all patient groups exhibited elevated MCP-1 level compared with controls. These observations suggest that the peripheral immune responses in both DLB and PD are associated with cognitive impairment. A multimodal imaging study with peripheral cytokine analysis explored central and peripheral inflammation and their correlation with clinical severity in DLB [9], where 3T magnetic resonance imaging and positron emission tomography imaging with 11C-PK11195, a marker of microglial activation, were performed in 19 patients with probable DLB and 16 age-matched controls. The levels of several peripheral inflammatory cytokines, mainly linked to T cell activation, were altered in the DLB patients compared with the controls. Specifically, the DLB patients had higher IL-2, IL-17A, and macrophage inflammatory protein-3 levels and lower IL-8 level. Of note, the results showed that microglial activation was elevated in DLB patients with mild impairment compared with those with moderate/severe impairment, suggesting that DLB is related to early microglial activation, which declines as cognitive impairment progresses. On the other hand, King et al. [35] reported the results of a study investigating inflammatory cytokines from plasma samples in 37 DLB patients, 20 AD patients, 38 patients with the DLB-mild cognitive impairment (MCI) profile, 20 patients with the AD-MCI profile, and 20 healthy controls. Interestingly, significant cytokine level changes were found only in the prodromal phase of DLB and AD patients. The DLB-MCI and AD-MCI groups displayed elevated levels of certain cytokines, specifically IL-1β, IL-2, IL-4, and IL-10, and lower level of TNF-α compared with control or dementia groups. Moreover, lower IL-1β, IL-2, and IL-4 levels and higher IL-6 and TNF-α levels correlated with more severe cognitive impairment across patient groups, and with greater motor severity in DLB and DLB-MCI groups. These findings provide a novel perspective of increased peripheral inflammation at the MCI stages that is attenuated with disease progression in both DLB and AD. In support of this hypothesis, a prospective longitudinal study over 3 years confirmed that six cytokines, IL-1β, IL-2, IL-4, IL-6, IL-10, and IFN-γ, exhibited significant decreases over time during the MCI phase of both DLB and AD, and these decreases were mostly correlated with cognitive decline as measured by Addenbrooke’s Cognitive Examination-Revised scores [36].

Kim et al. [37] first investigated peripheral inflammatory cytokine profiles in iRBD patients and explored whether these markers are related to prodromal symptoms of α-synucleinopathies. The authors collected plasma from polysomnography-confirmed iRBD patients without parkinsonism or dementia (n = 54) and healthy controls (n = 56) and measured the following cytokines: IL-1β, IL-2, IL-6, IL-10, and TNF-α. Unexpectedly, only the level of IL-10, the anti-inflammatory cytokine, were significantly upregulated in iRBD patients compared to the control group, but this difference did not withstand Bonferroni correction. Additionally, there was no correlation between the cytokine concentrations and any clinical variable. Taken together, the study did not find potential evidence supporting the role of peripheral inflammation in iRBD, but these observations were limited due to cross-sectional study design and the use of melatonin, which has anti-inflammatory properties [38], by many iRBD patients.

Recently, two longitudinal studies with comparable results have been published [39,40]. In one of these longitudinal studies, in 6 years of follow-up, the serum cytokine profiles of 30 iRBD patients were compared to those of 12 healthy controls using data collected under strict sample inclusion criteria [39]. The authors also explored whether such blood markers are related to phenoconversion risk. The IL-1β, IL-2, IL-6, IL-10, and TNF-α cytokine levels at baseline were not different between patients and controls, but a subgroup of patients with three or more of the five risk factors for developing α-synucleinopathies (mild parkinsonian signs, MCI, olfactory impairment, constipation, and reduced striatal dopamine transporter availability) had higher level of TNF-α than either those with fewer than three risk factors or controls. At longitudinal analyses, patients with higher TNF-α level were more likely to develop α-synucleinopathies over time than were those with lower level. Furthermore, there were no significant changes in cytokine levels during the 4-year follow-up period, which indicates that serum cytokine levels may be a reliable biomarker for predicting the risk of phenoconversion in iRBD rather than of disease progression. In another longitudinal study, the authors analyzed plasma levels of 10 cytokines including IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12p70, IL-13, TNF-α, and IFN-γ in 77 iRBD patients and 64 age- and sex-matched healthy controls [40]. They also studied 75 iRBD patients for up to 6 years of follow-up to determine phenoconversion. The results showed that TNF-α and IL-10 were significantly higher in iRBD patients than in controls, and that iRBD patients with higher TNF α/IL-10 levels were more likely to develop α-synucleinopathies during the follow-up period. Therefore, both longitudinal studies support the role of TNF-α in the neurodegeneration of iRBD. Interestingly, the potential importance of TNF-α in prodromal α-synucleinopathies was also emphasized in cohort studies on inflammatory bowel disease (IBD) [41,42]. IBD is one of the promising risk factors for PD [43], and anti-TNF agents are commonly used to control IBD-related symptoms [44]. In cohort studies, IBD significantly increased the incidence of PD, but the use of anti-TNF agents resulted in a remarkable reduction of PD risk [41,42]. Since such agents do not cross the blood-brain barrier, these findings may highlight peripheral TNF-α as one of the key factors in the pathogenesis of PD.

While it is becoming more evident that peripheral inflammatory cytokine changes exist in the early stages of α-synucleinopathies, the contribution of these changes to the progression of the disease is uncertain. This may be attributable to inconsistent findings across prior studies due to a lack of longitudinal study design. Thus, further longitudinal studies are warranted to better define the specific role of peripheral blood inflammatory cytokines in the clinical evolution of overt α-synucleinopathies. With respect to prodromal α-synucleinopathies, it remains unresolved whether peripheral inflammatory cytokine changes appear during the prodromal phases and how peripheral blood inflammatory cytokines are altered with disease progression. As the strongest prodromal marker of α-synuclein-specific neurodegeneration, iRBD is considered an optimal condition to address such questions. However, current evidence is lacking to support the association between peripheral blood inflammatory cytokines and iRBD as only a few studies with small sample sizes and limited cytokine analysis have examined their association. Additional large-scale studies with a broader range of cytokines are needed to reveal the role of peripheral blood inflammatory cytokines in the development of neurodegeneration during the prodromal stage of α-synucleinopathy and to validate the utility of these cytokines in predicting the conversion to overt α-synucleinopathy. Although emerging data indicate that TNF-α–associated neuroinflammation may promote neurodegenerative changes in prodromal α-synucleinopathy, this relationship needs further investigation.