Biomarkers for Lewy body diseases and other alpha-synucleinopathies in biofluids: current evidence and future directions

Synucleinopathies, including Parkinson’s disease (PD), dementia with Lewy bodies (DLB), and multiple system atrophy (MSA), are a group of proteinopathies characterized by neuronal and glial aggregated alpha-synuclein (α-syn) inclusions. Pathologically, these disorders are typically classified into two categories based on the distribution of α-syn: Lewy body diseases (LBDs), such as PD and DLB, which are characterized by α-syn aggregates in neuronal perikarya and neurites in the form of Lewy bodies (LBs) and Lewy neurites (LNs), and MSA, which shows α-syn aggregates in oligodendrocytes as glial cytoplasmic inclusions (GCIs). The clinical distinction between these disorders is challenging, especially in the early stages, due to the overlap of symptoms. This highlights the urgent need for reliable biomarkers to enable more accurate diagnosis and to guide the development of targeted therapeutic strategies. Current research focuses on α-syn which is recognized as a key protein in the pathology of PD, although its potential as a biomarker remains debated due to inconsistent findings. This review provides an overview of recent advancements in biomarkers research for synucleinopathies, focusing on α-syn, neurofilament light chain (NfL), tau, synapsin III (SynIII), and extracellular vesicles (EVs). These biomarkers have been identified in various biofluids and non-invasive sources, including cerebrospinal fluid (CSF), blood, olfactory mucosa (OM), and urine, suggesting promising avenues for the development of future diagnostic tools.

Highlight

α-syn SAAs provide high sensitivity and specificity for distinguishing LBDs from other neurodegenerative disorders.

α-syn, tau, and NfL correlate with clinical features, disease severity, and progression.

SynIII shows potential as a biomarker for synucleinopathies due to its interaction with α-syn and its accumulation in affected brain regions.

Biomarkers panel combining α-syn, NfL, tau, SynIII, and EVs offer improved diagnostic accuracy over single markers.

Urine and OM are promising new non-invasive and potential sources for biomarkers.

Introduction

Proteinopathies are neurodegenerative diseases characterized by abnormal accumulation, aggregation, and unregulated spreading of specific proteins having a physiological function in the central nervous system (CNS) [1]. Synucleinopathies are a group of proteinopathies, characterized by neuronal and glial aggregated alpha-synuclein (α-syn) inclusions [2]. Pathologically, synucleinopathies can be divided in two major groups: Lewy body diseases (LBDs) and multiple system atrophy (MSA) (Fig. 1). LBDs are characterized by the presence of α-syn aggregates in neurons perikarya and neurites in the form of Lewy bodies (LBs) and Lewy neurites (LNs). In contrast in MSA, α-syn typically accumulates in oligodendrocytes, in glial cytoplasmic inclusions (GCIs). Clinical manifestations of LBDs include Parkinson’s disease (PD), PD with dementia (PDD), and dementia with Lewy bodies (DLB) [3–5].

Fig. 1.

Schematic representation of Parkinsonian disorders. Synucleinopathies include PD and DLB, where α-syn fibrils accumulate in neurons of the substantia nigra forming LBs, and MSA, characterized by α-syn aggregates in oligodendrocytes as GCIs. Tauopathies include PSP and CBD, which are characterized by tau fibril deposition in distal astrocytes. LBDs: Lewy body diseases; MSA: multiple system atrophy; PD: Parkinson’s disease; DLB: dementia with Lewy body; LBs: Lewy bodies; GCIs: glial cytoplasmic inclusions; PSP: Progressive supranuclear palsy; CBD: corticobasal degeneration

PD is a progressive neurodegenerative disease affecting both the CNS and the peripheral nervous system [6]. It is estimated to impact 3% of people over the age of 80 years [7], making it the most common neurodegenerative movement disorder globally and the second most common neurodegenerative disease after Alzheimer’s disease (AD) [8, 9]. The clinical features of PD comprise motor symptoms, including resting tremor, bradykinesia, rigidity, postural disturbance and non-motor symptoms such as gastrointestinal dysfunction, hyposmia, depression, and rapid eye movement sleep behavior disorder (RBD) [10]. Notably, PD patients may show non-motor symptoms many years before the onset, during the prodromal phase, while the neurodegenerative process is ongoing [8, 9]. Indeed, motor symptoms mostly appear when patients have lost about 60–80% of dopaminergic neurons in the midbrain substantia nigra [11], that is one of the neuropathological hallmarks of PD associated with the formation of LBs and LNs [12, 13]. Interestingly, structural and functional cerebrovascular alterations and dysregulation of angiogenic factors such as angiogenin (ANG) have emerged as potential contributors in accelerating PD progression [14, 15]. Autopsy studies have shown a higher frequency of cerebrovascular lesions in PD compared with controls [16]. Loss-of-function variants of ANG are more frequent in PD patients compared with controls [15], indeed experimental evidence indicates that ANG supports neuronal survival, reduces toxin-induced dopaminergic cell death, and exerts protective effects in PD models [17], suggesting its potential therapeutic role for PD [17, 18]. Up to 80% of PD patients may develop dementia more than one year after the onset of motor symptoms, a condition known as PDD [19, 20]. Meanwhile, patients who develop dementia at or before the onset of parkinsonism are classified as DLB [21].

DLB is the second most common cause of neurodegenerative dementia after AD. The clinical features of DLB more similar to PDD, are dementia, fluctuating cognition, hallucinations, history of RBD, and motor symptoms [21], which may be absent in up to 25% of patients [22]. In addition, PDD and DLB have a similar age at onset >70 years, whereas PD onset is typically earlier with a mean of 60 years [22, 23].

Neuropathologically, DLB patients share features with both PD and AD, since, they present LBs inclusions in striatal and cortical neurons [24, 25], as well as senile plaques [26].

MSA is a progressive neurodegenerative movement disorder characterized by cerebellar ataxia, the initial motor symptom, along with autonomic failure, urogenital dysfunction, corticospinal disorders, and other parkinsonian features [27–30]. It is mainly characterized by the degeneration of striatonigral and olivopontocerebellar structures and by the presence of GCIs formed by fibrillized misfolded α-syn [31–33]. Additionally, misfolded α-syn is thought to propagate in a prion-like manner, spreading to other brain regions and contributing to the multisystem neuronal involvement typical of MSA [25, 34]. Clinically, MSA patients are typically classified into two subtypes, MSA with predominant cerebellar ataxia (MSA-C) and predominant parkinsonism (MSA-P) [29]. Although dementia is not a typical feature of MSA, a subset of MSA patients manifests mild cognitive impairment (MCI) followed by disfunction in memory and information processing speed [2, 35].

DLB and MSA belong to the group of atypical parkinsonian disorders (APDs) together with progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD). APDs are a heterogeneous group of movement disorders characterized by early dementia, gait instability, ataxia, pyramidal and cerebellar alterations, and also abnormal eye movement. While parkinsonism is a shared feature among APDs, each syndrome has signs that are atypical of PD. Additionally, these syndromes do not often respond to traditional PD treatments and also have a shorter survival time [36].

In contrast to DLB and MSA, which are classified as synucleinopathies, PSP and CBD are distinguished by an abnormal deposition of tau protein, classifying them as tauopathies (Fig. 1) [37].

PSP is a rare and rapidly progressive neurological disorder first described by Steele, Richardson and Olszewski in 1964 [38]. It is characterized by vertical supranuclear gaze palsy, degeneration of nerve cells in the brainstem and in the nucleus raphe interpositus [39]. Similar to PD, PSP presents with tremor, early asymmetric bradykinesia, and axial rigidity [40]. Additionally, PSP pathology affects gait, balance, speech, swallowing, vision, eye movements, behavior, and cognition [38]. The neuropathological hallmark is the aggregation of tau protein, which occurs as inclusions in astrocytes and oligodendroglia [41].

CBD has been also identified as a tauopathy, as evidenced by the presence of the characteristic “astrocytic plaques”, in which tau accumulates in the distal astrocytes [42]. The clinical presentation of CBD, known as corticobasal syndrome (CBS), presents an asymmetric movement disorder, which is characterized by bradykinesia, rigidity, and dystonia. Cognitive impairment is common and includes aphasia, executive dysfunction, visuospatial deficits, and behavioral change [43]. However, the majority of patients do not have tremor and are unresponsive to levodopa treatment [44, 45].

To date, early diagnosis of these proteinopathies is challenging, particularly in the synucleinopathies group. Hence, the research has focused on the development of biomarkers for early and accurate diagnosis, as well as for following disease progression and monitoring response to treatment.

Therefore, the goal of this review is to present the current state of knowledge on potential biomarkers in various biofluids for the diagnosis of synucleinopathies, with a particular focus on novel source biomarkers.

Biomarkers

Nowadays, biomarkers are mainly obtained from body fluids such as cerebrospinal fluid (CSF), blood, olfactory mucosa (OM), and urine. The identification of an ideal biomarker for synucleinopathies is a challenging process, particularly in the early stages due to the considerable overlap in symptoms. Many studies have highlighted the potential of α-syn species, neurofilament light chain (NfL), tau with its phosphorylated forms (p-tau), and recently synapsin III (SynIII), as candidate biomarkers in these pathologies (Fig. 2) [46].

Fig. 2.

Biomarkers across biofluids. (a) Overview of biomarkers currently available or under investigation for synucleinopathies, including various forms of α-syn, NfL, different tau protein isoforms, SynIII, and EVs. (b) Potential biological sources for biomarkers detection, including CSF, OM, and urine, indicating where each biomarker has been identified to date. α-syn: α-synuclein; NfL: neurofilament light chain, p-tau: phosphorylated tau; SynIII: synapsin III; EVs: extracellular vesicles; CSF: cerebrospinal fluid; OM: olfactory mucosa

Lately, several studies have focused on the role of extracellular vesicles (EVs), particularly exosomes, in the pathogenesis of various diseases, including synucleinopathies. Since EVs naturally play an important role in carrying information to distant cells and tissues, through the transfer of nucleic acids, lipids, and proteins, they have also been implicated in protein oligomers and aggregates spreading between CNS cells [47]. Notably, the presence of α-syn in its toxic form has been demonstrated in CNS-originating EVs, highlighting their potential as an important source of novel biomarkers for synucleinopathies [48–51].

α-Synuclein

α-syn, the main component of LBs and LNs, is a small 140 amino-acid soluble protein, encoded by SNCA gene. It is a neuronal protein, predominantly present in the CNS and located in the presynaptic terminal [52, 53]. α-syn has been defined as an intrinsically disordered protein or natively unfolded monomer [54, 55], that alternatively may exist as a tetramer [56, 57]. This suggests a balance between the monomeric and the tetrameric forms in healthy neurons. The majority of studies suggest that high levels of soluble oligomeric α-syn, rather than the insoluble fibril form, may be the pathogenic species in PD [58, 59]. Even though its function has not been fully elucidated, α-syn plays a role in regulating neurotransmitter release, synaptic function, synaptic plasticity, vesicular packaging and trafficking, and brain lipid metabolism [60, 61].

Several studies have shown that α-syn plays a crucial role in the etiology of PD and other synucleinopathies, in particular its phosphorylated form at residue Serine-129 (pSer129) that is the most prevalent form detected in PD brains [62]. Since the formation of α-syn oligomers precedes neuronal death in the synucleinopathies, the detection of early α-syn aggregates, including oligomers and phosphorylated forms, may improve early diagnosis [63].

The presence of α-syn has been detected in various biological samples, including CSF, plasma, saliva, urine, tears, and skin tissues [64–73]. Up to date, the research has focused on its potential role as a biomarker for PD and other synucleinopathies, as well as which form (total, oligomeric/aggregated, or post-translational modified) may represent a reliable biomarker for early diagnosis, patient stratification, and disease progression. Although the first attempt to detect α-syn in the CSF of PD patients and healthy controls (HC) was unsuccessful [74], further studies have confirmed its presence in the CSF by means of ELISA or other assays [66, 75–77].

Many studies have shown that CSF levels of monomeric α-syn are significantly lower in PD than in HC, while others have reported no changes between groups, probably related to the method used for quantification [64–66, 76, 78–82]. In a study by Foulds et al., the presence of the phosphorylated form in CSF could be a more direct marker of α-syn pathology in the brain. Notably, they demonstrated that the oligomeric phosphorylated form of α-syn could differentiate MSA patients from HC and other proteinopathies as PD, DLB, and PSP [83]. Majbour et al. have reported that combining measurements of different α-syn species (oligomeric/total, p-Ser129) in CSF improved the ability to discriminate between PD and HC [84].

Over the past decade, Seed Amplification Assays (SAAs) have emerged as powerful tools for detecting misfolded α-syn in CSF, offering high diagnostic accuracy. These methods, which take advantage of the prion-like properties of misfolded α-syn, which can induce conformational changes in its normal counterpart, forcing the formation of similar pathological structures. Briefly, in SAAs, the samples are generally incubated with a recombinant protein as a reaction substrate. The presence of pathological proteins, in the samples, triggers the real-time formation of amyloid fibrils (seeding activity), observable using a fluorescent dye [85, 86]. Several studies have reported a high accuracy in differentiating LBDs from APDs using SAAs [87–89], as well as distinguish PD and DLB from controls [89–91]. However, CSF α-syn SAAs showed a reduced diagnostic performance to detect MSA compared to PD and DLB [77, 92].

When considering total CSF α-syn as a potential diagnostic biomarker for PD, several limitations hinder its clinical use. First, the invasive nature of lumbar puncture, which requires an experienced and specialized physician, limits its widespread use. In addition, pre-analytical and analytical factors, as blood contamination, could significantly modify CSF total α-syn. This is attributable to the fact that erythrocytes are one of the main sources of this protein, thus any residual erythrocytes or hemolysis could increase its levels [93].

In this context, plasma and serum represent alternative biological fluids, which are easier to obtain using less invasive methods. Several studies have reported the identification and quantification of total plasma α-syn by ELISA or other similar techniques, but the results have been conflicting. Specifically, total plasma α-syn has been reported to be either significantly higher [94–100], lower [101–103] or not different [46, 48, 101–106] in PD patients compared to HC. Similary to CSF, conflicting results emerged in several studies on plasma and serum, most likely due to the different quantification methods used (Western Blot, ELISA). Consequently, an alternative line of research has emerged focusing on the examination of specific α-syn toxic forms, including oligomeric and post-translationally modified α-syn which may provide more reliable biomarkers [107–110].

The first study to examine oligomeric α-syn, was carried out by El-Agnaf et al., who developed an ELISA method that specifically recognized only oligomeric form of α-syn in human CSF and plasma. This assay was able to detect oligomers in postmortem CSF of PD and DLB patients, with a very low signal in the HC subjects tested. They found significant higher levels of oligomeric α-syn in PD compared to HC [107]. Similarly, Foulds et al. were the first to detect phosphorylated forms of α-syn in plasma, reporting increased plasma levels of pSer129 α-syn in PD patients compared to HC. Subsequent studies have confirmed the same results, supporting the potential of pSer129 as PD biomarker [107, 111–114]. Accordingly, Zhao et al. showed that pSer129, total, and oligomeric α-syn levels in plasma were significantly higher in PD patients compared to HC with no significant differences between PD and other APDs. Moreover, they demonstrated that these biomarkers together could differentiate PD from HC with an AUC of 0.8552 [115]. These results suggest that while total α-syn alone may not be sufficient to distinguish PD patients from HC, a combination of specific α-syn forms holds greater promise.

Recently, OM has also been investigated as potential source of biomarkers for synucleinopathies as it can be collected with less invasive procedures and without blood contamination. Since, olfactory dysfunction is present in at least 80% of PD patients, due to α-syn deposition in olfactory nuclei, mucosa, or epithelia. Thus, OM sampling provides a direct access to one of the potential initiation points of seeding and spread of pathological α-syn aggregates in PD [116–118]. Earliest analyses of OM ware performed by Beach et al., who used immunohistochemical to detected α-syn and pSer129 in olfactory bulb samples from post-mortem individuals as PD, DLB, incidental LBD, AD with Lewy-type synucleinopathies, and AD without synucleinopathies. This study demonstrated positive staining for synucleinopathy in the olfactory bulb, which can predict the presence of neuropathologically confirmed PD and DLB with greater than 90% sensitivity and specificity. The results also demonstrated that the severity of olfactory bulb synucleinopathy may serve as an approximate guide for the severity of Lewy pathology throughout the CNS, as it correlates with the severity of Lewy pathology in the brainstem, limbic regions, and neocortex, as well as with cognitive and motor dysfunction [119]. In contrast, Witt et al. demonstrated through immunohistochemical analysis that the total α-syn expression and distribution exhibited comparable levels across PD patients, non-PD patients with hyposmia, non-PD patients with anosmia, and HC [117]. More recently, α-syn in OM has been detected using the SAAs. The first study with SAAs in OM of patients affected by different neurodegenerative parkinsonian syndromes associated with tau pathology (PSP, CBD) or not (PD, MSA), aimed to assess the presence of abnormal α-syn and its seeding activity. They found that a significant proportion of MSA and PD samples exhibited seeding activity for α-syn compared to PSP and CBD [120]. OM was later used for α-syn SAAs analysis in PD, MSA and DLB patients, with relevant diagnostic accuracy in MSA (82%) and DLB (86.4%), but not in PD patients (range 44–48%) [120–122]. Therefore, in a study by Bongianni et al., the authors wanted to explore the performance of SAAs on samples collected from different regions of OM, particularly the agger nasi (AN) in the top of the nasal vault, and the middle turbinate (MT). In PD patients, SAAs sensitivity was significantly increased (from 45% to 84%) when the swab was performed at AN, which demonstrates that the aggregates of α-syn are more readily detected in areas with higher concentration of olfactory neurons [123]. Moreover, another study has shown that SAAs can be used to differentiate between MSA-P and MSA-C, as they have opposite seeding effects. Specifically, the cohort of patients enrolled in the study has showed seeding activity in the OM of 69% patients with PD and 90% patients with MSA-P, whereas no seeding of α-syn was detected in patients with MSA-C and HC [124]. These findings suggested that, while MSA-P and MSA-C are classified within the same disease group, they may be triggered by distinct α-syn strains that exhibit differential tropism for the OM. Besides PD and MSA, α-syn aggregates in the OM of DLB patients were first detected using SAAs by Perra et al., demonstrating the presence of pathological α-syn in olfactory neuroepithelium [122, 125]. Overall, these data suggest that through a careful combination of aggregation kinetics, biochemical and morphological assays, the OM α-syn SAAs can significantly improve the clinical diagnosis of synucleinopathies.

Urine represents an excellent non-invasive sample because can be easily and accurately collected during a routine procedure with large volume. Importantly, urine includes information from proximal tissues and from blood perfusing distant organs [126, 127]. The first study to examine α-syn forms in urine of PD and non-PD patients was conducted by Nam et al., who developed and validated an ELISA method to detect various α-syn oligomers, finding increased levels of fibrillar α-syn in PD compared to non-PD patients [128]. Recently, Müller et al. using a different method of analysis have reported a significantly increased concentration of α-syn aggregates in the urine of PD patients i compared to HC [129] A recent pilot study by SAAs has detected α-syn species in urine samples, showing low sensitivity (22% in DLB), but exceptionally high specificity (100%) [130]. These findings highlight that the identification and quantification of different forms of α-syn in urine can be an innovative method to discriminate synucleinopathies from HC or other neurological disorders, helping in the clinical diagnosis and development of early intervention to delay disease progression.

Neurofilament light chain

NfL is one of the subunits of neurofilaments, a family of neuronal cytoplasmic proteins highly expressed in large-caliber myelinated axons, where they provide their structural support and stabilization. Neurofilaments are composed of a triplet ofproteins: heavy (NfH), medium (NfM) and light, which is the essential component of the neurofilament core. In physiological conditions, low levels of NfL are released from axons, probably in an age-dependent manner, with higher levels of NfL released at older ages [131, 132]. However, in response to CNS axonal damage caused by neurologic disorders, including acute conditions such as vascular or traumatic brain injury and stroke, NfL release can be altered mainly in CSF but also in blood [133–136]. Elevated levels of NfL in CSF and blood can be observed in many neurodegenerative diseases such as AD, frontotemporal dementia, and amyotrophic lateral sclerosis [137–143].

Moreover, many researches have been investigating the role of NfL in PD, MSA, and DLB as marker of axonal injury in synucleinopathies [144–146].

Holmberg et al. described elevated CSF levels of NfL in MSA patients compared to PD suggesting more extensive neuronal damage in MSA than in PD [147]. Accordingly, a meta-analysis about CSF NfL levels in MSA and PD patients reported an increase of NfL levels in MSA compared to PD, suggesting that NfL may be a promising biomarker to differentiate these disorders [148]. Sara Bech et al. also confirmed a significant increase in CSF NfL levels in DLB, MSA, and PSP patients compared to PD group, and also shown that NfL concentrations in DLB appear to be at an intermediate level between PD and the other groups, probably reflecting the more moderate rate of neurodegeneration in this group [149]. Although elevated CSF NfL levels provide a high diagnostic accuracy in differentiating PD from other parkinsonisms, lumbar puncture is an invasive procedure and CSF is not suitable for repetitive sampling or long-term follow-up. For the first time, Hansson et al. have reported a strong correlation between blood and CSF NfL concentrations by means of a SIMOA based homebrew assay. They studied the diagnostic utility of blood NfL to differentiate PD patients from APDs, such as MSA, PSP, and CBD. They analyzed in three different cohorts, including one cohort with early-stage disease (disease duration ≤ 3 years), and found increased blood levels of NfL in MSA, PSP, and CBS patients compared with PD and HC. demonstrating high diagnostic accuracy (AUC values between 0.81 and 0.91) [150]. Several groups of research have established that blood NfL is clinically useful in differentiating APDs from PD [151–153]. Specifically, Schmitz et al. analyzed plasma NfL concentration in PD, DLB, and MSA patients. They showed a significant increase of plasma NfL levels of DLB and MSA compared to PD or HC with a high diagnostic accuracy for distinguishing non-PD synucleinopathies from PD and HC (AUCs 0.88–0.90), proving that NfL is a promising marker to discriminate between these conditions. Similarly, Ashton et al., in the largest study investigating plasma NfL in two multicenter cohorts with different neurodegenerative diseases, demonstrated an increase in plasma NfL in APDs (CBS, PSP, MSA) compared to PD with a high diagnostic accuracy for both cohorts (AUC: >86% and >95%) [151]. Marques et al. investigated the diagnostic value of serum NfL in patients with evident signs of parkinsonism but without a certain diagnosis at the time of the inclusion in the study. They showed that serum NfL levels were increased in APDs (MSA, PSP) compared to PD or HC and discriminated APDs from these groups with a high diagnostic accuracy (AUC: 0.91, 0.88, respectively). Moreover, in the patient cohort of this study, high levels of serum NfL were associated with a very high risk to have APDs while low levels were associated with a high chance to develop PD [152]. Recently, a study demonstrated an association between CSF and plasma NfL and clinical disease progression in MSA. Moreover, in line with previous studies, in both CSF and plasma, NfL levels were significantly elevated in MSA compared to PD and HC, with no significant differences in PD compared to HC [154]. It has also been observed that elevated NfL levels are present in PD patients with rapid disease progression, in contrast to those with slow progression [46, 150, 155–157]. In conclusion, typical PD is associated with the lowest elevation in plasma NfL levels, compared to APDs or HC. A study of Lin et al. found higher plasma NfL levels in PDD compared with non-demented PD to indicate that in PD patients, plasma NfL is associated with the cognitive impairment [156].

Furthermore, NfL levels cannot differentiate between neurological diseases with similar degrees of axonal loss. For these reasons, the potential diagnostic role of NfL proteins in the clinical setting should be complemented by other neurological assessments and a combination of multiple biomarkers.

Tau

Tau is a microtubule-associated protein [158, 159], highly expressed in neurons and encoded by MAPT gene [160]. It is abundant in the axons of neurons where it promotes the polymerization and stability of microtubules and regulates axonal transport [161–163]. Tau is also present in dendrites and postsynaptic compartments where regulates synaptic plasticity [164–166]. Tau localization in neurons and its role in synaptic function is regulated by various forms of post-translational modification, including phosphorylation (p-tau), which also sets tau function and pathogenesis of tauopathies [167]. Pathological tau can cause synaptic loss and dysfunction, high levels of total tau (t-tau) are correlated with neuronal degeneration, and high levels of p-tau with neurofibrillary tangle pathology in AD [168]. To date, large studies have shown that blood phosphorylated tau at residue threonine-181 (p-tau181), threonine-217 (p-tau217), and threonine-231 (p-tau231) may be able to distinguish AD from other neurodegenerative diseases [169–171]. Less is known about the use of these biomarkers in distinguishing the common AD-copathology in DLB. Gonzales et al., demonstrated that plasma p-tau181 and p-tau231 levels in DLB were intermediate between the AD and HC with significantly higher levels in DLB patients compared to HC but lower than AD. Moreover, in DLB, they described a significant association between both p-tau levels and cognitive decline, suggesting that both p-tau181 and p-tau231 may serve as valuable indicators for monitoring disease progression. In addition, they observed that plasma p-tau levels were increased in DLB patients with AD as comorbidity [172]. Accordingly, Hall et al. have also demonstrated that plasma p-tau is elevated in DLB with AD co-pathology [173]. Recently, Vrillon et al., confirmed these observations, showing that plasma p-tau181 were higher in DLB compared to HC but lower than AD. Moreover, they confirmed elevated p-tau181 levels in DLB patients with AD co-pathology, suggesting a potential use of this biomarker for patient selection in amyloid-beta (Aβ) targeting therapeutics [174]. Chouliaras et al., found a p-tau181 moderate diagnostic accuracy in distinguishing between MCI-AD and DLB (AUC = 0.67) [175]. Several studies investigated the association between plasma p-tau181 and cognitive performance in PD, with contrasting results. Chen et al., demonstrated a correlation between high plasma t-tau and cognitive evaluation scores and severity of PD, with increased plasma t-tau significantly associated with cognitive impairments in PD patients [176]. In another study, plasma p-tau181 was elevated in PD patients compared to HC, but it was not confirmed the association with cognitive decline [177].

Numerous studies have identified tau pathology in the anterior olfactory nucleus of PD patients, suggesting that tau may contribute to olfactory impairment in this disease. Interestingly, CBS and PSP, which have little or no olfactory loss, did not show tau pathology in this region. In conclusion, although tau pathology has been observed in olfactory-related brain regions of PD patients, direct quantification of tau protein in the OM has not been not reported yet [178, 179].

Synapsin III

SynIII is a member of the synapsins, a family of neuronal phosphoproteins, localized at the synapses, on the cytoplasmic surface of the synaptic vesicles [180–182]. SynIII is considered a key regulator of early neurodevelopment and dopamine neurotransmission, playing an important role in neurogenesis and axon formation, as well as being involved in slow synaptic transmission, particularly dopamine release [183]. For the first time, Zaltieri et al. have demonstrated an interaction between α-syn and SynIII which regulates the arrangement of releasable vesicle pools and synaptic function in dopaminergic neurons [184]. Recently, Faustini et al. found that SynIII deficiency prevents α-syn aggregation, nigral neuron degeneration, and the onset of synaptic changes in SynIII knockout PD mouse model. They have also shown, in vitro, that both absence or aggregation of α-syn in dopamine neurons significantly affects the expression and subcellular distribution of SynIII, and that the silencing of this protein in vitro is able to prevent α-syn aggregation and the associated redistribution of synaptic proteins. Therefore, it is possible that SynIII negatively regulates dopamine release and is involved in α-syn aggregation and toxicity. Taken together, these investigations about SynIII have yielded novel insights into the mechanisms of neural plasticity as they pertain to a range of disorders, suggesting that SynIII could serve as a potential biomarker for synucleinopathies [185, 186]. Further human studies are essential to clarify its translational potential, as current evidence is largely derived from animal and in vitro models, which, while informative, cannot fully recapitulate human pathophysiology [187]. Additionally, a specific assay for both the identification and quantification of SynIII has not yet been found. However, validating SynIII as a human biomarker is challenging due to the difficulty of reliably measuring its levels in accessible biofluids, given its low concentration. In summary, SynIII shows potential as a biomarker for synucleinopathies due to its interaction with α-syn and its accumulation in affected brain regions.

Extracellular vesicles

EVs represent a collective term to describe a large group of small, membranous particles released by cells. Based on their origin, EVs are generally classified into three main types: exosomes, derived from endosomes; microvesicles, formed by the outward budding of the plasma membrane; and apoptotic bodies, released during cell death via apoptosis [188]. Up to date, the term exosome refers to a group of small EVs with a diameter ranging from 50 to 150 nm, released into the extracellular space following thethe fusion of multivesicular bodies with the plasma membrane. Whereas, microvesicles represent a heterogeneous in size, ranging from 100 nm to 1 μm, and are shed directly from the plasma membrane [189]. EVs serve as carriers of nucleic acids, lipids, and proteins, facilitating intercellular communication and playing essential roles in various physiological processes [190]. Specifically, in the CNS, EVs contribute to myelination, neurotrophic support, and synaptic plasticity [191]. Regarding synucleinopathies, EVs could act as scavengers, facilitating the removal of pathological proteoforms when cellular clearance mechanisms become insufficient [192, 193]. EVs can transport key pathological proteins among cells, acting also as mediators of the pathology spread [194].

Importantly, EVs can be isolated from different biofluids, including CSF, blood, saliva, and urine, providing a direct prospective on the biological processes occurring within CNS [195–198]. To date, many studies have focused on EV cargo, especially proteins involved in proteinopathies such as α-syn, Aβ, and tau [192, 199, 200].

Stuendl et al. compared α-syn levels in CSF-derived EVs between PD, APDs, and HC. Interestingly, DLB patients exhibited lower α-syn concentration in EVs than PD and PSP patients, enabling differentiation between these groups with high sensitivity and specificity. However, the quantification of α-syn revealed that CSF-derived EVs from PD samples had lower concentration of α-syn compared to HC. They also found an increased EV number in PD respect to DLB and PSP [201]. Similarly, Hong et al. have demonstrated that levels of both total and aggregated α-syn-positive EVs were lower in the CSF of PD patients compared to HC. Interestingly, combining the two EV subpopulations, significantly improved diagnostic accuracy, discriminating PD patients from HC (AUC 0.82, sensitivity 80%, specificity 71%) [202].

Among the first studies of blood-derived EVs, Shi et al. reported that both plasma EV-associatedα-syn [203] and tau [204] correlated with PD severity, performing even better than the quantification of α-syn or tau in CSF. In particular, they have found a robust increase of EVα-syn and tau in PD patients compared to HC, suggesting an increased peripheral release or efflux of these pathological proteins. Accordingly, other several studies demonstrated that in plasma EVs of PD patients were detected higher levels of α-syn compared to HC [205, 206], as well as to APDs [48, 207]. Moreover, plasma EV-derived α-syn could be a potential diagnostic biomarker for PD, discriminating early-stage PD from HC (AUC 0.8, sensitivity 100%, specificity 57%) [206]. However, the studies on EV concentration are contradictory, thus it has been reported that plasma EV concentration was higher in PD versus HC and APDs [207–209], lower in PD compared to HC and APDs [48], or not significant between PD and HC [205]. Interestingly, Longobardi et al. observed a decrease in EV concentration and an increase in size in the DLB patients compared to the HC [210], with an opposite trend in the CSF of the same group [211].

These contradictory results could be influenced by methodological factors such as the isolation/extraction and quantification of EVs. Among the many studies on serum-derived EVs, two relevant studies have demonstrated the ability of EVs to discriminate between PD and APDs. Jiang et al. showed that exosomal α-syn was increased twofold in prodromal and clinical PD compared to HC, APDs (MSA, PSP), and RBD [49]. Likewise, Ishiguro et al. found a significant increase in filamentous α-syn in PD EVs compared to HC, MSA, and PSP, along with a notable reduction in overall EV concentration in PD compared to HC [212].

In the context of the non-invasive biofluids, such as urine, two studies have shown that EVs isolated from urine of PD patients contains higher levels of leucine-rich repeat kinase 2 (LRRK2) or deglycase protein 1 (DJ-1) proteins whose coding genes are known to carry causative mutations for PD than non-PD controls. Interestingly, α-syn was not detected in EVs from urine samples of both PD and non-PD samples [198, 213].

To the best of our knowledge, only one study has so far characterized EVs from the urine of PD cases with urinary dysfunction and controls. Roy et al. did not find significant differences in urinary EV size and concentration; however, they found a significant positive association between urinary EV concentration and motor scores [214].

All these studies suggest that EV characterization, together with other synucleinopathy biomarkers, could represent a helpful analysis to take in account for the differential diagnosis and/or to evaluate the disease progress.

Discussion

This review synthesizes insights into potential biomarkers for α-synucleinopathy, focusing on α-syn, NfL, tau, SynIII, and EVs in biofluids (CSF, blood, urine) and other non-invasive sample types (OM) (main studies summarized in Table 1). To date, the accurate diagnosis of PD, MSA and DLB is challenging due to the frequent overlap of symptoms, highlighting the urgent need for reliable, and disease-specific biomarkers, or combinations of them.

Table 1.

Summary of the main studies described in this review focusing on α-syn, NfL, tau, SynIII, and EVs in CSF, blood (plasma/serum), urine, and OM

Author, year	Biomarker	Source	Subjects	Method	Main Findings
Zhen Hong et al., 2010 [64]	α-syn	CSF	PD:117, AD:50, Ctrl:13	Luminex assay	• ↓ α-syn in PD vs. AD, Ctrl • No significant correlation between biomarker levels and PD severity
Brit Mollenhauer et al., 2011 [65]	α-syn and tau	CSF	Training cohort: PD:51, DLB:55, MSA:29, AD:62, Ctrl:76 Validation cohort: PD:273, DLB:66, MSA:15, PSP:8, Ctrl:23	ELISA	Training cohort: • ↓ α-syn PD, MSA, DLB vs. AD, Ctrl • ↑ Tau in DLB and AD vs. PD Validation cohort • ↓ α-syn in PD, MSA, DLB vs. Ctrl • ↑ Tau in Ctrl vs. other groups
Tokuda T. et al., 2006 [66]	α-syn	CSF	PD:33, Ctrl:38	ELISA	• ↓ α-syn in PD vs. Ctrl
Ishii R. et al., 2015 [67]	α-syn	CSF, plasma	PD:30, Ctrl:58	ELISA	• ↓ α-syn in PD vs. Ctrl
Nam et al., 2021 [70]	α-syn	Urine	PD:13, non-PD:8	ELISA	• ↓ α-syn in PD vs. non-PD • ↑ oligomeric-α-syn/α-syn in PD vs. non-PD
Mollenhauer et al., 2008 [76]	α-syn	CSF	PD:8, DLB:38, AD:13 Ctrl:13	ELISA	• ↓ α-syn in DLB vs. Ctrl
van Rumund et al., 2019 [77]	α-syn seeding activity	CSF	PD:53, MSA:18, DLB:1, PSP:8, Ctrl:52	SAAs	• 84% α-syn diagnostic accuracy to distinguish α‐synucleinopathies from non–α‐synucleinopathies and controls
Ohrfelt et al., 2009 [79]	α-syn	CSF	PD:15, DLB:15, AD:66, Ctrl:55	ELISA	• No significant differences
Tokuda et al., 2010 [80]	α-syn, oligomeric α-syn	CSF	PD:32, PSP:18, AD:35, Ctrl:28	ELISA	• ↑ oligomeric α-syn PD vs. PSP, AD, Ctrl • ↑ oligomeric-α-syn/α-syn PD vs. PSP, AD, Ctrl
Mollenhauer et al., 2013 [82]	α-syn	CSF	PD:78, Ctrl:48	ELISA	• ↓ α-syn in PD vs. Ctrl
Foulds PG et al., 2012 [83]	α-syn, oligomeric α-syn, pSer129, oligomeric pSer129	CSF	PD:13, DLB:17, PSP:12, MSA:8, Ctrl:20	ELISA	• No significant differences in α-syn, oligomeric α-syn and pSer129 • ↑ oligomeric pSer129 in MSA vs. PD, DLB, PSP, and Ctrl
Majbour et al., 2016 [84]	α-syn, oligomeric α-syn, pSer129	CSF	PD: 46, Ctrl: 48	ELISA	• ↑ oligomeric α-syn and pSer129 in PD vs. HC • ↓ α-syn in PD vs. Ctrl
Quadalti et al., 2021 [87]	α-syn, NfL	CSF, plasma	PD:153, DLB:64, PSP/CBD:58, MSA:80, RBD:19, Ctrl:72	SAAs, Simoa	• α-syn SAAs reliably detected synucleinopathies (PD, MSA, DLB), distinguishing them from tauopathies (PSP, CBD) • ↑ NfL in MSA, PSP, CBD vs. PD, Ctrl • ↑ CSF NfL in MSA vs. PSP, CBD
Singer et al., 2020 [88]	Oligomeric α-syn, NfL	CSF	PD:16, DLB:13, MSA:38, Ctrl:15	SAAs, ELISA	• Oligomeric α-syn was detected in PD, DLB higher than MSA • ↑ NfL in MSA vs. PD, DLB, Ctrl
Rossi et al., 2020 [89]	α-syn	CSF	PD:71, DLB:34, MSA:31, PSP/CBD:30, AD:43, Ctrl:62	SAAs	• SAAs detected α-syn seeding with high sensitivity and specificity for PD and DLB, while MSA samples were mostly negative
Fairfoul G. et al., 2016 [90]	α-syn	CSF	DLB:12, PD:2, PSP:2, CBD:3, AD:30, Ctrl:20	SAAs	• SAAs detected PD and DLB with high sensitivity and seeding activity compared to tauopathies (PSP, CBD), AD, Ctrl
Kang et al., 2019 [91]	α-syn	CSF	PD:105, Ctrl:79	SAAs	• SAAs detected α-syn seeding only in PD samples
Shahnawaz M. et al., 2020 [92]	α-syn	CSF	PD:94, MSA:75, Ctrl:56	SAAs	• ↑ α-syn seeding in PD vs. MSA, Ctrl • PD and MSA have distinct seeding profiles, enabling differentiation between the two synucleinopathies
Bougea et al., 2020 [94]	α-syn, tau, p-tau181	CSF, plasma, serum	PD:30, DLB:29, PDD:18, Ctrl:30	ELISA	• ↑ CSF α-syn in DLB vs. PD, PDD and Ctrl • ↓ CSF α-syn in PD, PDD vs. Ctrl • ↑ serum α-syn in PD, PDD and DLB vs. Ctrl   • ↑ p-tau181 in DLB vs. PD
Duran et al., 2010 [95]	α-syn	Plasma	PD:95 (PD − L-Dopa:53, PD + L-Dopa:42), Ctrl:60	ELISA	• ↑ α-syn in PD vs. Ctrl
Horvath et al., 2017 [97]	α-syn	CSF, plasma	Plasma: PD:20, Ctrl:20 CSF: PD:30, Ctrl:30	ELISA	• ↑ α-syn in PD vs. Ctrl
Lee et al., 2006 [98]	α-syn	Plasma	PD:105, MSA:38, Ctrl:51	ELISA	• ↑ α-syn in PD, MSA vs. Ctrl • ↑ α-syn in PD vs. MSA
Ng et al., 2020 [99]	NfL	Plasma	PD:149, Ctrl:50	SIMOA	• ↑ NfL in PD vs. Ctrl
Wang et al., 2019 [100]	α-syn	Plasma	PD:45, Ctrl:45	ELISA	• ↑ α-syn in PD vs. Ctrl
Gorostidi et al., 2012 [101]	α-syn	Plasma	PD:134, Ctrl:109	ELISA	• ↓ α-syn in PD vs. Ctrl
Goldman et al., 2018 [106]	α-syn	CSF, plasma	PD:115, Ctrl:88	ELISA	• ↓ CSF α-syn in PD vs. Ctrl • Non-significant differences in plasma
El-Agnaf et al., 2006 [107]	Oligomeric α-syn	Plasma	PD:34, Ctrl:27	ELISA	• ↑ oligomeric α-syn in PD vs. HC
Foulds et al., 2011 [108]	α-syn, oligomeric α-syn, pSer129	Plasma	PD:32, Ctrl:30	ELISA	• ↑ pSer129 in PD vs. Ctrl • Non-significant differences in α-syn and oligomeric α-syn
Malec-Litwinowicz et al., 2018 [111]	α-syn	Plasma	PD:58, Ctrl:38	ELISA	• ↑ α-syn in PD vs. Ctrl
Cariulo et al., 2019 [112]	p-Ser129	Plasma	PD:5, Ctrl:5	ELISA	• ↑ pSer129 in PD vs. Ctrl
Foulds et al., 2013 [113]	α-syn, pSer129	Plasma	PD:189, Ctrl:91	ELISA	• ↑ pSer129 in PD vs. Ctrl • Non-significant differences in α-syn
Zhao et al., 2022 [115]	α-syn, pSer129, oligomeric α-syn	Plasma	PD:79, APDs:24, Ctrl:42	ELISA	• ↑ α-syn, pSer129, oligomeric α-syn in PD, APDs vs. Ctrl
De Luca CMG et al., 2019 [120]	α-syn	OM	PD:18, MSA:11, CBD:6, PSP:12	SAAs	• SAAs detected α-syn seeding with high efficiency, in PD and MSA than CBD and PSP
Stefani A. et al., 2021 [121]	α-syn	OM	PD:41, RBD:63, Ctrl:59	SAAs	• ↑ α-syn seeding in PD, RBD vs. Ctrl
Perra D et al., 2021 [122]	α-syn	OM, CSF	DLB:37, DLB/AD mixed dementia:6, non-α-synNDs:38 (non-α-syn-related neurodegenerative disorders)	SAAs	• SAAs detected α-syn seeding only in DLB and DLB/AD mixed dementia
Bargar C et al., 2021 [124]	α-syn	OM	MSA-P:20, MSA-C:10, PD:13, Ctrl:11	SAAs	• SAAs detected α-syn seeding in PD and MSA-P • MSA-C did not induce α-syn seeding activity
Nam D. et al., 2020 [128]	α-syn, oligomeric α-syn	Urine	PD:21, non-PD:11	ELISA	• ↑ oligomeric α-syn in PD vs. non-PD
Constantinescu R. et al., 2010 [145]	NfL	CSF	PD:10, MSA:21, PSP:14, CBD:11, Ctrl:59	ELISA	• ↑ NfL in APDs vs. PD, Ctrl
Herbert MK. et al., 2015 [146]	NfL, tau	CSF	Discovery cohort: PD:36, MSA:27, Ctrl:57 Validation cohor: PD:32, MSA: 25, PSP:15, CBD:5, Ctrl:56	ELISA	• ↑ NfL in MSA vs. PD, Ctrl • ↑ NfL in APDs vs. PD, Ctrl • ↑ tau in MSA vs. PD in discovery cohort
Holmberg B. et al., 1998 [147]	NfL	CSF	PD:19, MSA:10, PSP:12	ELISA	• ↑ NfL in APDs vs. PD
Bech S. et al., 2012 [149]	NfL	CSF	PD:22, DLB: 11, MSA:10, PSP:10	ELISA	• ↑ NfL in APDs vs. PD
Hansson O. et al., 2017 [150]	NfL	CSF, blood	Cohort 1: PD:171, MSA:30, PSP:19, CBD:5, Ctrl:53 Cohort 2: PD:20, MSA:30, PSP:29, CBD:12, Ctrl:26 Cohort 3: PD:53, MSA:28, PSP:22, CBD:6	ELISA, Simoa	• Blood NfL correlated strongly with CSF NfL • ↑ NfL in APDs vs. PD, Ctrl • ↑ NfL in PDs vs. Ctrl in Cohort 2
Marques TM et al., 2019 [152]	NfL	Serum	PD:55, MSA: 22, PSP:7, Ctrl:53	Simoa	• ↑ NfL in APDs vs. PD, Ctrl
Katzdobler S. et al., 2024 [154]	NfL	CSF, plasma	PD: 24, MSA: 47, Ctrl:25	Simoa	• ↑ NfL in MSA vs. PD, Ctrl
Lin Y. et al., 2018 [156]	NfL	Plasma	PD:26, PDD:23, AD:19, MCI:56, Ctrl:59	Simoa	• ↑ NfL in AD vs. PD, PDD, MCI, Ctrl • ↑ NfL in PDD vs. PD
Pilotto A. et al., 2021 [157]	NfL	Plasma	PD:92, Ctrl:45	Simoa	• ↑ NfL in PD vs. Ctrl
Gonzalez MC. et al., 2022 [172]	p-tau181, p-tau231	Plasma	PD:204, DLB:371, AD:207, Ctrl:205	Simoa	• ↑ p-tau181, p-tau231 in DLB vs. Ctrl • ↑ p-tau181, p-tau231 in AD vs. PD, DLB, Ctrl
Vrillon A. et al., 2024 [174]	p-tau181, NfL	Plasma	DLB:104, AD:76, Ctrl:27	Simoa	• ↑ p-tau181 in DLB vs. Ctrl • ↑ NfL in DLB vs. Ctrl • ↑ p-tau181, NfL in AD vs. DLB, Ctrl
Chouliaras et al., 2022 [175]	p-tau181, NfL	Plasma	DLB:117, MCI-AD:63, PSP:19, Ctrl:73	Simoa	• ↑ NfL in all dementias vs. Ctrl • ↑ p-tau-181 in MCI-AD vs. DLB
Batzu L. et al., 2022 [177]	p-tau181, NfL	Plasma	PD:136, Ctrl:63	Simoa	• ↑ p-tau181 in PD vs. Ctrl • NfL levels did not differ between the groups
Dutta S. et al., 2021 [197]	α-syn	Blood-derived EVs	PD:50, MSA:30, Ctrl:51	ELISA	• ↑ α-syn in MSA vs. PD, Ctrl • ↑ α-syn in PD vs. Ctrl
Stuendl A. et al., 2016 [201]	α-syn, EVs	CSF, CSF-derived EVs	Cohort 1: PD:159, Ctrl:110 Cohort 2: PD:32, DLB:32, PSP:25	NTA, Electrochemiluminescence immunoassay, ELISA	• ↓ α-syn in PD vs. Ctrl in both CSF and EVs • ↑ EV α-syn in PD, PSP vs. DLB in Cohort 2 • ↑ EV in PD vs. DLB, PSP
Hong Z. et al., 2021 [202]	α-syn, EVs	CSF-derived EVs	PD:170, Ctrl:131	NTA, ELISA	• No significant differences were observed in the number or size distribution of total EVs in PD vs. Ctrl • ↓ EV α-syn in PD vs. Ctrl
Shi M. et al., 2014 [203]	α-syn	Plasma-derived EVs, plasma, CSF	PD:267, Ctrl:215	Luminex assay	• ↑ EV α-syn in PD vs. Ctrl. • ↑ EV α-syn/plasma α-syn in PD vs. Ctrl • ↓ CSF α-syn in PD vs. Ctrl
Shi M., et al., 2016 [204]	tau	Plasma-derived EVs, plasma	PD:91, AD:106, Ctrl:106	Simoa	• EV tau in PD vs. Ctrl • ↑ plasma tau in AD vs. PD, Ctrl
Cerri S. et al., 2018 [205]	α-syn	Plasma-derived EVs, plasma	PD:39, HC:33	NTA, ELISA	• ↑ EV α-syn in PD vs. Ctrl • ↑ EV α-syn/plasma α-syn in PD vs. Ctrl • ↑ EV α-syn/ number of EVs in PD vs. Ctrl
Niu M. et al., 2020 [206]	α-syn	Plasma-derived EVs	PD:53, RBD:20, Ctrl:21	Electrochemiluminescence immunoassay	• ↑ EV α-syn in PD vs. Ctrl
Yu Z. et al., 2020 [207]	α-syn, EVs	Plasma-derived EVs, plasma	PD:34, MSA:32, Ctrl:31	NTA, Luminex assay	• ↑ EV α-syn in PD vs. MSA • ↑ EV α-syn/plasma α-syn in PD, Ctrl vs. MSA • ↑ EV concentration in PD, Ctrl vs. MSA
Vacchi E. et al., 2020 [208]	EVs	Plasma-derived EVs	PD:27, MSA:8, CBD/PSP:9, Ctrl:19	NTA	• ↑ EV concentration in PD vs. MSA, CBD/PSP
Ohmichi T. et al., 2019 [209]	EVs	Plasma-derived EVs	PD:15, MSA:15, PSP:7, Ctrl:15	ELISA	• ↑ EV concentration in PD vs. MSA, Ctrl
Longobardi A. et al., 2021 [210]	EVs	Plasma derived EVs	DLB:30, AD:30, Ctrl:30	NTA	• ↓ EV concentration in DLB vs. Ctrl • ↑ EV size in DLB vs. Ctrl
Longobardi A. et al., 2022 [211]	EVs	CSF derived EVs	DLB:30, AD:36, Ctrl:20	NTA	• ↑ EV concentration in DLB vs. Ctrl • ↓ EV size in DLB vs. Ctrl
Ishiguro Y. et al., 2024 [212]	EVs, α-syn	Serum derived EVs	PD:142, MSA:18, PSP:28, RBD:31, Ctrl:105	NTA, ELISA	• ↓ EV concentration in PD, RBD vs. Ctrl • ↑ EV α-syn in PD vs. MSA, PSP, Ctrl • ↑ EV α-syn in RBD vs. Ctrl
Roy S. et al., 2023 [214]	EVs	Urine derived EVs	PD:29, Ctrl:29	NTA	• No significant difference in both EV size and concentration
Gilboa et al. 2024 [215]	α-syn, pSer129	Plasma-derived EVs, plasma	PD:15, DLB:15, Ctrl:15	Simoa	• ↑ EV α-syn in PD vs. Ctrl • ↑ plasma α-syn in PD vs. Ctrl • ↑ EV α-syn/plasma α-syn in DLB vs. Ctrl • ↑ plasma pSer129 in PD vs. Ctrl
Wang Yu et al., 2012 [217]	α-syn, pSer129	CSF	Discovery cohort: PD:93, MSA:16, PSP:33, AD:26, Ctrl:78 Validation cohort: PD:116, MSA:25, PSP:27, Ctrl:126	Luminex assay	• ↓ α-syn in PD, MSA vs. Ctrl • ↑ pSer129 in PD vs. MSA, PSP • ↓ pSer129 in PSP vs. Ctrl • ↑ pSer129 in PD vs. Ctrl only in discovery cohort
Constantinides V. et al., 2021 [219]	α-syn, pSer129	CSF	PD:13, MSA:9, PSP:13, CBD:9, AD:51	ELISA	• ↓ α-syn in MSA vs. CBD • ↓ α-syn in PD, MSA vs. PSP, CBD • ↑ pSer129 /α-syn in PD, MSA vs. PSP, CBD
Poggiolini I. et al., 2021 [224]	α-syn	CSF	PD:74, MSA:24, RBD:45, Ctrl:55	SAAs	• ↑ α-syn seeding in PD, MSA, RBD vs. Ctrl
Orrù C. et al., 2021 [226]	α-syn	CSF	PD:108, Ctrl:85	SAAs	• ↑ α-syn seeding in PD vs. Ctrl
Cristiani C. et al., 2024 [232]	Oligomeric-α-syn, p-tau181	Serum	PD:43, PSP:27, Ctrl:39	ELISA, SIMOA	• No differences were found

PD: Parkinson’s disease; DLB: dementia with Lewy body; MSA: multiple system atrophy; PSP: Progressive supranuclear palsy; CBD: corticobasal degeneration; AD: Alzheimer’s disease; APDs: Atypical parkinsonian disorders; MCI: Mild cognitive impairment; PDD: Parkinson’s disease with dementia; RBD: Rapid eye movement sleep behavior disorder; Ctrl: control; α-syn: α-synuclein; NfL: neurofilament light chain; p-tau: phosphorylated tau; SynIII: synapsin III; EVs: extracellular vesicles; pSer129: Phosphorylated alpha-synuclein at residue Serine-129; p-tau181: Phosphorylated tau at residue threonine-181; p-tau231: Phosphorylated tau at residue threonine-231; CSF: cerebrospinal fluid; OM: olfactory mucosa; SAAs: Seed amplification assays; NTA: Nanoparticle Tracking Analysis; ↑: increased; ↓: decreased

Given the central role of α-syn in the pathology and pathophysiology of the synucleinopathies, there has been growing interest in quantifying both total α-syn and its pathological species in these disorders. Nowadays, no single α-syn species provides definitive diagnostic accuracy across all biofluids. However, oligomeric and phosphorylated forms, particularly in CSF, have emerged as the most reliable for PD diagnosis [84]. Plasma pSer129 has shown the most consistent signal among peripheral biofluids, with multiple studies reporting elevated levels in PD compared to controls [113, 115, 215], although these findings remain heterogeneous [216, 217]. Interestingly, few studies have evaluated pSer129 for differentiating LBDs from other neurodegenerative disorders. Higher CSF pSer129 levels have been reported in PD compared to MSA in a validation cohort [217], while other studies have failed to replicate these findings, likely due to small sample sizes [46, 83, 218, 219]. Many studies also analyzed pSer129 to distinguish PD from PSP, CBD, and AD, but no consistent differences were observed [46, 83, 218–220].

Overall, these findings suggest that α-syn species alone are insufficient, whereas composite measures—such as pSer129/t-α-syn or oligomeric/t-α-syn ratios—offer greater diagnostic specificity across synucleinopathies [84, 221, 222].

In this context, the ideal biomarker would be highly sensitive and specific, and would distinguish α-synucleinopathies from each other and from non-synuclein-related disorders. Furthermore, biomarkers should be detectable in peripheral, minimally invasive, and easily accessible biofluids (e.g., blood, OM, urine), given the limitations and reduced patient compliance associated with invasive procedures like lumbar puncture. CSF-based assays currently offer the highest diagnostic accuracy, but their invasiveness is a practical barrier. Blood assays are appealing for ease of use but must overcome challenges like contamination and inconsistent results. Non-invasive options, such as OM and urine, show growing promise, especially when paired with highly sensitive techniques like SAAs. However, urine has low CNS relevance, therefore is mainly explored for complementary systemic biomarkers or in multiplex panels [223]. Overall, CSF remains the ‘gold standard’ for CNS as it reflects brain biochemistry and pathology. Biomarkers such as misfolded α-syn, t-tau, and p-tau are often elevated in CSF, showing high diagnostic sensitivity and specificity for disease detection [65, 224–226].

However, CSF α-syn shows high sensitivity 71–94% but limited specificity 25–53%, whereas plasma α-syn yields 48–53% of sensitivity and 69–85% of specificity [227]. Blood-derived α‑syn assays are promising due to their accessibility, but currently less reliable than CSF, with moderate discriminatory power and considerable variability. In a study comparing plasma and CSF NfL, CSF NfL had cutoff values giving ~ 95% specificity with ~ 71% sensitivity for discriminating neurodegenerative disease from psychiatric conditions; plasma performance was comparable [228]. The OM represents another promising source, as the olfactory system is early affected in synucleinopathies and sampling is minimally invasive relative to CSF. Several studies using α-syn SAAs in OM detection demonstrate moderate sensitivity 51–64% with consistently higher specificity 86–92%. In a cohort of RBD and PD versus controls, the sensitivity was ~ 45.2% and the specificity ~ 89.8% [121]. Another study that optimised sampling location found that OM from the anterior naris yielded a sensitivity of 78–84% compared with ~ 45% for the middle turbinate, meanwhile the specificity remained high. In the same study, CSF achieved ~ 92% sensitivity with similarly high specificity. Combining OM and CSF further improved diagnostic accuracy, occasionally approaching near 100% [123]. Regarding urine, research is still preliminary and focuses on the potential of α‑syn. A recent study reported low sensitivity (22% in DLB), but exceptionally high specificity (100%) for urinary α-syn [130].

Reproducibility is another essential criterion for a biomarker, as consistent results must be obtained across different laboratories and studies. However, reproducibility issues have been noted with some techniques, such as α-syn SAAs, due to variability in technical conditions. Beyond diagnostic utility, an ideal biomarker should also enable early detection during the prodromal or preclinical stages, allow monitoring of disease progression, and assess therapeutic efficacy. In this regard, SAAs have shown the capacity to detect misfolded α-syn aggregates in CSF even before motor symptoms onset in PD, supporting preclinical diagnosis [24]. Moreover, α-Syn SAAs shows a strong association with LBDs, correlating with both LB burden and disease stage. It has also been linked to motor and non-motor symptoms, including cognitive decline and RBD [89]. In parallel, NfL levels increase in the early stages in rapidly progressive synucleinopathies, such as MSA, and continue to rise with disease progression, correlating with more rapid motor and global clinical deterioration which supports both early detection and longitudinal monitoring [229]. NfL is consistently higher in MSA than PD, aiding differentiation among synucleinopathies [152, 154]. NfL is also elevated in AD, particularly in the later disease stages, but generally lower than in MSA and may overlap with LBD depending on disease severity. However, NfL is non-specific biomarker that can be affected by age and other neurological conditions [230].

Tau protein provides additional diagnostic value. T-tau and p-tau are typically higher in AD than DLB and can detect early co-pathology that often influences the prodromal phase of LBDs [231]. These tau forms have been implicated in early diagnosis and monitoring of PD progression, with strong associations to cognitive impairment in both PD and DLB [231, 232].

Emerging biomarkers, including EVs and SynIII, show promise for identifying early molecular changes in synucleinopathies. EVs have demonstrated diagnostic potential in early disease stages [233] and have been evaluated in CSF, plasma, and serum for differentiating DLB or PDD from AD [210, 211, 234, 235]. Overall, diagnostic accuracy was generally lower in CSF than plasma, though publication bias has been noted across studies [236]. Notably, EV-derived α-syn has been shown to correlate with motor symptom severity and may reflect disease progression in PD [234]. SynIII remains in the experimental phase but could provide mechanistic insight into the early α-syn aggregation.

While no single biomarker is fully specific, a combinatorial panel—including α-syn, p-α-syn, NfL, tau, SynIII, and EV profiling—can improve diagnostic accuracy and support differentiation of LBDs from other neurodegenerative disorders with overlapping clinical features. However, despite this potential, translating these findings into validated diagnostic assays remains challenging. Technical variability, high assay costs, limited access to advanced platforms, and insufficient clinical validation continue to impede implementation. Population heterogeneity—including genetic, ethnic, and environmental factors—can further influence baseline biomarker levels, affecting both sensitivity and specificity.

The substantial variability in reported results, especially for α-syn, highlights the lack of standardized methods, as these biomarkers are not yet used in clinical practice. The main pre-analytical and analytical factors that potentially influence the measurement of these biomarkers, are (i) sample collection and storage, (ii) blood contamination, and (iii) selected assay. For instance, α-syn protein structures can be denatured with low temperature, significantly reducing the α-syn values; studies have shown that sample centrifugation at 4 °C has been associated with lower α-syn concentrations compared to room temperature processing [237]. Blood contamination during CSF collection is another critical concern, as α-syn is highly expressed in red blood cells [238, 239]. The assay choice is also crucial, as it involves selecting the appropriate antibody for capturing and detecting the target protein. Equally important is the choice of the assay, as it directly influences the accuracy and sensitivity of the result [238]. Although ELISA has been widely used, alternative technologies, such as Luminex and SIMOA, are also available, introducing variability in the detection range and sensitivity [71, 150, 202].

NfL levels are influenced by blood–brain barrier permeability, age, and are susceptible to non-specific increase in various neurological disorders [240]. Tau protein detection is complicated by multiple isoforms and post-translational modifications [241], while low-abundance proteins, such as SynIII, are highly sensitive to sample handling and assay selection. EVs measurements are influenced by isolation methodology. A recent meta-analysis reported higher diagnostic accuracy for EVs isolated by ultracentrifugation compared to polymer-based precipitation or size exclusion chromatography [236]. Taken together, while multi-biomarker panels show promise for improving diagnostic accuracy, significant challenges remain to be address for standardization, reproducibility, and clinical validation, highlighting the need for future research.

Conclusion and future direction

α-syn, tau, NfL, SynIII, and EVs hold significant promise for clinical use and their validation requires a concerted effort across multiple fronts to demonstrate their diagnostic, prognostic, and therapeutic value across different stages of synucleinopathies.

One critical requirement is the execution of large-scale and multi-center studies involving different populations since current biomarker research often suffers from small and homogenous cohorts. Moreover, longitudinal studies are essential to evaluate biomarker dynamics across the disease progression, from prodromal to advanced stages, and to determine their prognostic value. Equally important is the standardization and validation of assay protocols. Biomarker measurements can vary significantly across laboratories due to differences in sample handling, assay platforms, and detection techniques. Establishing universally accepted protocols will enhance reproducibility and comparability across studies, facilitating regulatory approval and clinical implementation. Given the multifactorial nature of neurodegenerative diseases, reliance on single biomarker is unlikely to provide sufficient diagnostic or prognostic accuracy. There is growing consensus around the development of multi-biomarker panels that integrate complementary pathological signatures. Among the most promising panels, α-syn SAAs combined with NfL capture both disease-specific protein aggregation and axonal injury, supporting early detection and differentiation of PD and related synucleinopathies from APDs [87]. Similarly, pairing tau with α-syn, especially when measured in CSF or EVs, enhances discrimination between LBDs and AD [242, 243]. Looking ahead, combining fluid biomarkers with imaging, digital phenotyping, and machine learning could provide the most reliable diagnostic tools. Machine learning can uncover non-linear relationships and interaction among biomarkers, clinical variables, and imaging data, improving diagnostic accuracy and patient stratification. Beyond technical validation, it is essential to establish the clinical utility and cost-effectiveness of these biomarkers. Their integration into routine clinical workflows will depend not only on their diagnostic accuracy but also on their ability to inform treatment decisions and improve patient outcomes. Additionally, future research should aim to clarify the mechanistic and pathological specificity of biomarkers.

An integrated approach that combines technological innovation, clinical applicability, and economic feasibility will be useful to overcoming the current limitations of biomarker-based diagnostics. Other essential steps would include standardizing methodologies, simplifying testing platforms, integrating multiple data streams, and fostering global collaboration.

Abbreviations

α-syn

Alpha-synuclein

Aβ

Amyloid-beta

AD

Alzheimer’s disease

AN

Agger nasi

APDs

Atypical parkinsonian disorders

CBD

Corticobasal degeneration

CBS

Corticobasal syndrome

CNS

Central nervous system

Ctrl

Control

CSF

Cerebrospinal fluid

DJ-1

Deglycase protein 1

DLB

Dementia with Lewy bodies

EVs

Extracellular vesicles

GCIs

Glial cytoplasmic inclusions

HC

Healthy controls

LBDs

Lewy body diseases

LBs

Lewy bodies

LNs

Lewy neurites

LRRK2

Leucine-rich repeat kinase 2

MCI

Mild cognitive impairment

MSA

Multiple system atrophy

MSA-C

Multiple system atrophy with cerebellar ataxia

MSA-P

Multiple system atrophy with predominant parkinsonism

MT

Middle turbinate

NfH

Neurofilament heavy chain

NfL

Neurofilament light chain

NfM

Neurofilament medium chain

NTA

Nanoparticle Tracking Analysis

OM

Olfactory mucosa

PD

Parkinson’s disease

PDD

Parkinson’s disease with dementia

pSer129

Phosphorylated alpha-synuclein at residue Serine-129

PSP

Progressive supranuclear palsy

p-tau

Phosphorylated tau

p-tau181

Phosphorylated tau at residue threonine-181

p-tau217

Phosphorylated tau at residue threonine-217

p-tau231

Phosphorylated tau at residue threonine-231

RBD

Rapid eye movement sleep behavior disorder

SAAs

Seed amplification assays

SynIII

Synapsin III

t-tau

Total tau

Author contributions

AR and CS contributed to the conception of the article and the composition of the original draft. AR, AL, and CS contributed to the collection of information. AL, AC, FAC, MBB, BB, RG, and FM contributed to the review and editing of the manuscript. All authors read and approved the final version of the manuscript.

Funding

This research was funded by the Italian Ministry of Health (GR-2021-12372019).

Data availability

Not applicable.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.