How close are we to a breakthrough? The hunt for blood biomarkers in Parkinson's disease diagnosis

Parkinson's disease (PD), being the second largest neurodegenerative disease, poses challenges in early detection, resulting in a lack of timely treatment options to effectively manage the disease. By the time clinical diagnosis becomes possible, more than 60% of dopamine neurons in the substantia nigra (SN) of patients have already degenerated. Therefore, early diagnosis or identification of warning signs is crucial for the prompt and timely beginning of the treatment. However, conducting invasive or complex diagnostic procedures on asymptomatic patients can be challenging, making routine blood tests a more feasible approach in such cases. Numerous studies have been conducted over an extended period to search for effective diagnostic biomarkers in blood samples. However, thus far, no highly effective biomarkers have been confirmed. Besides classical proteins like α‐synuclein (α‐syn), phosphorylated α‐syn and oligomeric α‐syn, other molecules involved in disease progression should also be given equal attention. In this review, we will not only discuss proposed biomarkers that are currently under investigation but also delve into the mechanisms underlying the disease, focusing on processes such as α‐syn misfolding, intercellular transmission and the crossing of the blood–brain barrier (BBB). Our aim is to provide an updated overview of molecules based on these processes that may potentially serve as blood biomarkers.

K E Y W O R D S blood biomarkers, Parkinson's disease, pathogenesis, α-synuclein

| INTRODUCTION
With the increase in life expectancy, there is a growing emphasis on age-related diseases, particularly neurodegenerative disorders such as Parkinson's disease (PD) and Alzheimer's disease (AD).Early diagnosis and proactive treatment play a crucial role in ensuring a good quality of life for patients affected by these diseases.Over the past three decades, extensive research has been dedicated to discovering reliable, minimally invasive and cost-effective diagnostic and monitoring methods to achieve this objective (Surguchov, 2022).Biomarkers found in the cerebrospinal fluid (CSF) of AD patients provide such a precedent (Parnetti et al., 2019) because of its preclinical stage (Dubois et al., 2016), enabling early diagnosis (Porsteinsson et al., 2021).Similarly, biomarkers in the CSF of PD patients have also demonstrated reliable diagnostic capabilities (Parnetti et al., 2019).Effective biomarkers should not only support early diagnosis but also enable monitoring and prediction of the disease's progression and prognosis.
The diagnosis of PD is primarily based on the presence of symptoms, particularly cardinal motor symptoms such as bradykinesia, rest tremor, rigidity and postural instability (Grimes et al., 2019).However, by the time these symptoms manifest, approximately half of dopaminergic neurons in the substantia nigra (SN) have been damaged (Schapira, 1999).Before that, effective medications may achieve better therapeutic effects.Additionally, it is challenging to differentiate PD from secondary Parkinsonism or other neurodegenerative forms of Parkinsonism, such as multiple system atrophy (MSA), normal pressure hydrocephalus (NPH) and progressive supranuclear palsy (PSP), based solely on clinical symptoms (Grimes et al., 2019;Rizzo et al., 2016).Although imaging techniques like magnetic resonance imaging (MRI), single-photon emission computed tomography (SPECT) and positron emission tomography (PET) can assess the density of dopaminergic neurons, these methods lack specificity, are costly, inconvenient and involve radiation exposure (Bidesi et al., 2021).Therefore, there is a need for suitable biomarkers.Molecules in CSF are closer to the site of pathology but lumbar puncture is highly invasive.For asymptomatic patients and those requiring long-term follow-up, it is crucial to have minimally invasive and convenient monitoring methods.Thus, compared with CSF, biomarkers in the blood may be more practical and serve as our goal of research.
Indeed, significant progress has been made in the field of biomarkers for PD over the past decade.Several systematic reviews and meta-analyses have provided comprehensive assessments of these biomarkers (Jiménez-Jiménez et al., 2020;Nie et al., 2022;Nila et al., 2022;Rizzo et al., 2016;Santos et al., 2023).In this review, we will delve into the pathogenesis of PD, including the processes involved in the formation and transport of misfolded α-synuclein (α-syn), lysosomal dysfunction and metabolite profiling.By exploring these various aspects, we aim to provide an updated overview of blood biomarkers for PD, shedding light on their diagnostic and prognostic potential.Furthermore, familial cases are often more homogenous in terms of genetic mutations and can offer valuable clues about the molecular mechanisms involved.

| Formation of misfolded and aggregated α-syn
The aggregation of misfolded α-syn in Lewy bodies (LBs) was identified as a prominent feature in the autopsies of PD patients in the 1990s (Spillantini et al., 1997).Since then, α-syn has emerged as a key molecule in PD (Simonsen et al., 2016).The full-length α-syn protein consists of 140 amino acids and is encoded by six exons in the α-synuclein gene (SNCA).However, there are also α-syn variants with different sizes, containing varying numbers of amino acids (126,112,98) (Simonsen et al., 2016).Posttranslational modifications (PTMs) such as phosphorylation (at Ser129 and Tyr125), ubiquitination (at K12, K21 or K23) and oxidation can induce conformational changes in α-syn (Schmid et al., 2013;Vicente Miranda et al., 2017), transitioning it from an α-helical structure to a β-sheet structure.This misfolding of α-syn leads to the formation of aggregates (Figure 1), similar to the processes observed in the conversion of healthy prion (PrP C ) becoming toxic prion (PrP Sc ) (Vargas et al., 2019).Continuous aggregation of misfolded α-syn results in the formation of monomers, oligomers and fibrils.This prion hypothesis has paved the way for the development of mouse models using α-syn preformed fibrils (PFFs) (Volpicelli-Daley et al., 2014).Mutations (such as A53T, A30P, and E46K) and gene multiplications in SNCA have been identified in familial cases of PD and are known to contribute to the accumulation of misfolded α-syn (Sahay et al., 2017).Additionally, mutations in other genes, such as leucine-rich repeat kinase 2 (LRRK2), DJ-1, Parkin and PTENinduced putative kinase 1 (PINK1), which are associated with familial forms of PD, have also been implicated in the pathogenesis of the disease (Harper et al., 2018).LRRK2, encoded by the PARK8 gene, is associated with PD.The G2019S mutation in LRRK2 is a common mutation found in PD patients.Pathogenic mutations in the LRRK2 gene lead to increased kinase activity of LRRK2, and small-molecule LRRK2 kinase inhibitors have shown potential neuroprotective effects (Tolosa et al., 2020).DJ-1, encoded by the PARK7 gene, is another important protein associated with PD.It has been implicated in inflammatory disorders related to PD (Repici & Giorgini, 2019).Recent research has highlighted the role of the autophagy-lysosome pathway in PD (Alvarez-Erviti et al., 2011).Damage to this pathway can result in intracellular aggregation of α-syn and reduced intracellular degradation of α-syn (Sacino et al., 2017).Glucocerebrosidase (GCase), an important lysosomal hydrolase encoded by the GBA gene, plays a significant role in this process (Figure 1).Mutations (E326K, T369M) in GBA contributed to accumulation of α-syn (Maor et al., 2019).During the process of aggregation, both nonamyloid-β component (NAC) region and highly acidic COOH-terminal region in α-syn protein play important roles (McCann et al., 2016).These regions are involved in the formation and stabilisation of α-syn aggregates.(3) PTMs such as phosphorylation (Ser129, Tyr125), ubiquitination (K12, K21, or K23) and oxidisation can accelerate the formation of misfolded α-syn.Once misfolded α-syn is formed, the impairment of degradative pathways contributes to misfolded α-syn aggregation and α-syn oligomer (Oli-α-syn) formation.(4) αsyn produced by neurons themselves or obtained from outside the cells can be degraded by lysosome and (5) ubiquitin-proteasome system (UPS).Glucocerebrosidase (GCase) in lysosome is encoded by GBA gene.The mutation of GBA gene results in the decreased activity of GCase.Additionally, (6) tau and Aβ42 also participate in α-syn fibril formation.(7) Different α-syn species in neurons can be exchanged by TNTs.α-syn can also be released to extracellular space directly or by EVs and then taken up by other neurons.(8, 9) α-syn released by neurons in CSF or red blood cells (RBC) and platelets in peripheral blood can be exchanged through BBB.The interaction of α-syn with HDL may facilitate its transport, although the mechanisms are not yet clear.

| Transport processes of α-syn fibres
The transmission pathways of α-syn are not yet fully understood, but there is evidence suggesting several mechanisms.α-syn can be transferred directly between cells or via extracellular vesicles (EVs).Initially, cells take up α-syn through receptors such as heparan sulfate proteoglycan (HSPG) and lymphocyte activation gene-3 (LAG-3) (Mao et al., 2016).Then, exogenous α-syn could interact with endogenous α-syn, providing an opportunity for aggregation to occur.Subsequently, different α-syn species could be released out of cells through tunnelling nanotubes (TNTs) or exosomes (Alvarez-Erviti et al., 2011) and then taken up by other cells (Rodriguez et al., 2018;Vargas et al., 2019).Interestingly, misfolded α-syn has also been observed to cross blood-brain barrier (BBB) directly although the underlying processes are not yet fully understood.Studies have indicated that α-syn may interact with high-density lipoproteins (HDL), which could potentially facilitate its transport across the BBB (Emamzadeh & Allsop, 2017).Another phenomenon is that astrocytes, a component of the BBB, also uptake α-syn fibres, providing a potential route for α-syn to enter the bloodstream.EVs, including exosomes and mitochondria-derived vesicles (MDVs), are released by virtually all cell types and play significant roles in intercellular communication with a diameter of 40-160 nm (average 100 nm) (Wang et al., 2023).Neuron-derived exosomes (NDEs) labelled with markers such as L1 Cell Adhesion Molecule (L1CAM), neural cell adhesion molecule (NCAM) and oligodendrocyte-myelin glycoprotein (OMG) have been identified.Other types of EVs, such as astrocyte-derived exosomes (ADEs) and oligodendrocytederived exosomes (ODEs), have also been reported (Dutta et al., 2021).In the central nervous system (CNS) (Wang et al., 2023), these EVs serve as another important pathway for the transport of α-syn among neurons and across BBB (Figure 1) (Rodriguez et al., 2018;Shi et al., 2014).

| α-syn species in blood
Expanding on the mechanisms of α-syn aggregation and transport, various studies have detected total α-syn, PTMs α-syn, oligomeric α-syn and exosomal α-syn in serum and plasma (Jiménez-Jiménez et al., 2023).Interestingly, no research has reported the presence of aggregated α-syn in the blood, which has been observed in CSF using the RT-QuIC Assay (Manne et al., 2019).Total α-syn has not demonstrated significant diagnostic potential, as its concentration in blood can vary, being higher, lower or indistinguishable.Different studies also examined α-syn in RBC; unfortunately, the results were also conflicted.However, in a meta-analysis incorporating 35 studies, total α-syn in plasma and serum showed the power in distinguishing PD and healthy controls (HCs) (Zubelzu et al., 2022).
Due to small diameter and double-membrane structure of NDE α-syn, they can transmit information to places that cells cannot reach, such as crossing BBB, making them potential biomarker candidates.A longitudinal study (Niu et al., 2020) and a pilot study (Wang et al., 2018) demonstrated a significant positive correlation between NDE total α-syn in plasma and the severity of PD, as assessed by the Movement Disorders Society Unified Parkinson's Disease Rating Scale (UPDRS) and Nonmotor Symptom Questionnaire (NMSQ) tests.Oligomeric α-syn and the Ser129 phosphorylated α-syn were also detectable in NDE.The ratios of α-syn oligomer/total α-syn and phosphorylated α-syn oligomer/total phosphorylated α-syn were helpful for diagnosing PD (AUC = 0.71 and AUC = 0.69, respectively) (Zheng et al., 2021).It is worth noting that α-syn and p-α-syn were found on the surface of exosomes (Zheng et al., 2021), which suggested a potential role in intercellular communication or disease propagation.Similarly, a meta-analysis reached the same conclusion that NDE α-syn could distinguish PD and HC (Xylaki et al., 2023).

| GCase
As discussed above, several studies have investigated lysosomes enzymes to be as biomarkers for PD diagnosis such as GCase, alpha-galactosidase (α-Gal) and cathepsin.These enzymes may have stronger diagnostic power in CSF (Parnetti et al., 2017).One study found an inverse correlation between GCase enzymatic activity in peripheral blood lymphocytes of PD patients and the exosomal α-syn/total α-syn ratio in PD.However, there was no significant difference in GCase activity between PD and HCs, suggesting that increased α-syn levels might inhibit GCase activity (Cerri et al., 2018).In PD patients with GBA mutations (GBA-PD), both GCase and α-Gal activity were reduced in dry blood spot samples but not sphingomyelinase, alpha-iduronidase or alpha-glucosidase activity (Pchelina et al., 2017).Interestingly, levels of GCase protein in PD without GBA mutations increased compared with HC, although it should be confirmed in a larger cohort (den Heijer et al., 2023).
Another important mechanism involved in the regulation of α-syn degradation is ubiquitination.The small ubiquitin-like modifier protein (SUMO) attached to the lysine residues of α-syn increases in the brains of PD patients, which may be one of the reasons for the reduced ubiquitination-mediated degradation of α-syn and the release of aggregated forms of α-syn (Rott et al., 2017).Maybe this approach could serve not only for treatment but also for discovering biomarkers for PD.

| Lipid metabolism, following GCase
GCase, mentioned earlier as a lysosomal hydrolase, can metabolise the glucosylsphingosine and glucosylceramide (GlcCer) to ceramide and glucose, whose inherited deficiency is reported in Gaucher disease (GD) patients (Rolfs et al., 2013).Glucosylsphingosine can also be formed through the de-acylation of accumulated GlcCer.To improve the diagnosis of GD, researchers have also investigated elevated chitotriosidase levels and GCase substrates released by Gaucher cells (Rolfs et al., 2013).Interestingly, glucosylsphingosine was also noticed in PD patients and was able to participate in the α-syn aggregation (Taguchi et al., 2017).Glucosylsphingosine were significantly higher in GBA-PD plasma (den Heijer et al., 2023;Surface et al., 2022).In GBA-PD, abnormalities in ceramide metabolism in plasma have also been detected.Elevations in ceramide species such as C16:0, C18:0, C20:0, C22:0 and C24:1, as well as monohexosylceramide species C16:0, C20:0 and C24:0, were found to be associated with the severity of Parkinson's disease dementia (PDD) as measured by the Geriatric Depression Scale (GDS) (Mielke et al., 2013).Another study has reported the consistent results (Xing et al., 2016).Analysis of over 40 lipid classes revealed elevated levels of some key lipids such as monohexosylceramide, ceramide and sphingomyelin, while phosphatidic acid (PA), phosphatidylethanolamine (PE), plasmalogen phosphatidylethanolamine (PEp) and acyl phosphatidylglycerol (AcylPG) were decreased in the serum of GBA-PD patients.Chitotriosidase was similar across these groups (Guedes et al., 2017).These studies provide direction for future research, and larger cohorts are needed to confirm these findings.

| METABONOMICS, COMPREHENSIVE PERSPECTIVE
In Hatano study, PD patients exhibited significantly lower levels of tryptophan, caffeine and its metabolites, bilirubin and ergothioneine, as well as significantly higher levels of levodopa metabolites and biliverdin than those of normal controls (Taku et al., 2016).These findings provide valuable insights for further in-depth research.In the plasma of PD-related depressive patients, the predominant metabolites among the 85 are related to lipid and sugar metabolism.Subsequent proteomic analysis revealed that these products are mainly involved in oxidative stress.This study also reported a powerful biomarker, neurogenic locus notch homologue protein 2 (NOTCH2) to diagnose PD-related depressive, with a cut-off point of 0.91 ng/mL, a sensitivity value of 95.65% and a specificity value of 81.58% (Dong et al., 2018).The use of vanillic acid, 3-hydroxykynurenine, isoleucylalanine, 5-acetylamino-6-amino-3-methyluracil and theophylline was able to discriminated PD and HC (AUC = 0.955 with 87.5% sensitivity and 93.0%specificity).The use of His-Asn-Asp-Ser, 3,4-dihydroxyphenylacetone, desaminotyrosine, hydroxy-isoleucine, alanylalanine, putrescine [-2H], purine [+O] and its riboside (AUC = 0.862 with 80.0% sensitivity and 77.0%specificity) was able to discriminate PD with no dementia with PD with incipient dementia (Han et al., 2017).Fifteen plasma constituents, including three with xanthine structures and four medium-or long-chain fatty acids closely related to the progression of PD, were identified using UPDRS (0.87, p = 2.2e-16) (LeWitt et al., 2017).The spermidine/spermine ratio was significantly decreased in PD patients and had the ability to distinguish PD and HC (AUC = 0.868), and they could enhance longevity via autophagy induction (Saiki et al., 2019).Bilirubin, biliverdin and glycodeoxycholic acid were able to differentiate essential tremor patients and OFF-PD (12 h after the last dose of antiparkinsonian drugs) that indicated these metabolites were associated with oxidative stress (Albillos et al., 2021).Serum levels of caffeine and nine of its downstream metabolites were significantly decreased even in patients with early PD.Using caffeine alone, the AUC reached 0.78, and when combined with its metabolites, the AUC increased to 0.87 (Fujimaki et al., 2018).It had been demonstrated that caffeine had neuroprotective effects in MPTP mice models.Decreased 14 lipid metabolites in Parkinson's disease-related anxiety disorder (PDA) patients' plasma serve as diagnostic biomarkers separately with AUC ranged from 0.681 to 0.798 (Dong et al., 2021).
The sensitivity and comprehensiveness of omics research provide significant advantages, but extracting meaningful clinical insights from the vast amount of data generated remains a challenge.This is particularly true in the context of PD, where there is clinical heterogeneity among patients, variations in drug treatments and different stages of disease progression, all of which can impact the identification of reliable biomarkers and therapeutic targets.Moreover, interpreting the data and unravelling the underlying disease mechanisms require further research and investigation.
LncRNAs were beyond 200 nt in length with functions of direct DNA or mRNA binding, chromatin modifier regulation, posttranscriptional modification and chromatin 3D structure formation (Kuo et al., 2021).In the circulating leukocytes of PD patients, four lncRNAs (AC131056.3-001,HOTAIRM1, lnc-MOK-6:1 and RF01976.1-201)upregulated, and HOTAIRM1 and AC131056.3-001may contribute to PD pathogenesis by promoting the apoptosis of dopaminergic neurons (Fan et al., 2019).Six dysregulated lncRNAs were detected both in PD plasma and brain, and four of them, including SNCA-AS1, MAPT-AS1, AK127687 and AX747125, were detected in exosomes (Elkouris et al., 2019).In Kong study, 129 upregulated and 282 downregulated circRNAs are reported in PD plasma, and circRNAs (Kong et al., 2021) may play an important role in oxidative stress in PD (Dias et al., 2013).Explaining the changes in RNA from a pathogenic perspective is a challenge we face.What is more, the heterogeneity among PD patients poses a difficulty in data analysis.Standardised sampling and detection procedures can yield more reliable results.Furthermore, rich and upto-date databases can provide clues for further research.

| OTHER PROTEINS IN EVS
In addition to the mentioned α-syn in EVs, several other proteins have also been investigated.Different types of EVs that have been detected, including NDEs, ADEs and ODEs, were all significantly increased in PD and had corrections with the course of PD and MSA with predominant parkinsonism (MSA-P) (Ohmichi et al., 2019).Three exosomal proteins, namely, clusterin, complement C1r subcomponent and apolipoprotein A1 (Apo A 1) and fibrinogen gamma chain decreased in PD patients as potential biomarkers and Apo A1 could reflect the progression of PD (Jiang et al., 2020;Kitamura et al., 2018).These findings were consistent with another study that clusterin, gelsolin, complement C1r and apolipoprotein D in exosomes as being related to different stages of PD (Jiang et al., 2019).DJ-1 and α-syn in exosomes derived from neurons were significantly higher in PD patients compared with controls, and a positive correlation between DJ-1 and α-syn was observed.Nonetheless, no relationship between levels of two proteins and disease progression was discovered (Zhao et al., 2018).The release of neurofilament light chain (NfL) was followed by axonal injury (Zetterberg, 2016).However, in PD, axonal damage in neurons is not as severe, especially in the early stages.Therefore, research has not yielded significant elevations of this protein in PD plasma, exosomes (Chung et al., 2020) or CSF (Parnetti et al., 2019).Nevertheless, Nfl may have value in distinguishing PD patients from other diseases with remarkable damage of myelinated axons such as MSA (AUC = 0.95), PSP (AUC = 0.97) (Hansson et al., 2017) and amyotrophic lateral sclerosis (ALS) (AUC = 0.90) (Haji et al., 2022).Further analysis of proteins in exosomes is needed.Compared with plasma, exosomes have a cleaner background and can directly pass through BBB, providing us with a better subject for research.

| COMBINED WITH AΒ42, TAU
Aβ42 and tau, which are biomarkers associated with AD, have also been considered as potential biomarkers for PD in CSF and plasma.Aβ42 and tau could facilitate the process of α-syn aggregation.In CSF of PD patients, levels of Aβ42 and tau were not consistent (Kang et al., 2013;Kang et al., 2016).Nevertheless, Aβ42 in CSF had a good discriminative power of PDD (Nutu et al., 2013).Nai-Ching also reported that Aβ40 (AUC = 0.791) and tau (AUC = 0.726) in plasma are potential biomarkers to detect cognitive impairment in PD patients (Chen et al., 2020).When combined with α-syn, Aβ42 and tau could enhance the differential diagnosis of PD from atypical Parkinsonism syndromes (APS) (Lin et al., 2018).Moreover, exosomal tau was significantly higher in PD patients and related to CSF tau (Shi et al., 2016).These proteins in exosomes may hold greater value of diagnosis.In conclusion, Aβ42 and tau used alone may not arrive reliable diagnostic power because plasma tau is driven from extra-CNS events, but when used in combination with other biomarkers or that in NDEs, they can improve diagnostic accuracy.

| IRON AND IRON-RELATED PROTEINS
Iron plays a role in various biological processes within cells, including mitochondrial respiration and oxygen transport (Ward et al., 2014).Its oxidation state changes participate in electron transfer processes, suggesting that iron may be a potential pathogenic mechanism.The release of reactive oxygen species (ROS) is thought to be closely associated with neurodegenerative diseases, but the exact role of iron in this process is still unclear (Ndayisaba et al., 2019).However, postmortem results from PD patients have shown a decrease in iron content in the temporal cortex compared with normal individuals (Yu et al., 2013).Several studies have explored the levels of iron and iron-binding proteins in blood and CSF, investigating their potential as biomarkers.In a metaanalysis, CSF and serum/plasma ferritin and transferrin concentrations and serum/plasma lactoferrin and haptoglobin concentrations did not differ significantly between PD patients and controls, and serum/plasma iron levels were marginally significantly lower (Jiménez-Jiménez et al., 2021).Different meta-analyses have exhibited conflicting results, similar (Wei et al., 2018), increased (Jiao et al., 2016) or decreased (Genoud et al., 2020) iron levels in PD serum/plasma.Iron levels in serum/plasma can be affected by various factors especially RBC and the qualified samples are obligatory.

| DIAGNOSIS OF PD BY BLOOD BIOMARKERS, IS IT POSSIBLE?
This review explores potential biomarkers in the plasma and serum of PD patients.There are two approaches for identifying reliable biomarkers.One approach focuses on biomarkers directly associated with pathogenic processes, such as the different α-syn species, as well as the transport mechanisms involved.Pathological mechanisms like lysosomal damage and exosome-mediated transport are also considered, as they can serve as both therapeutic targets and diagnostic biomarkers.These biomarkers are responsible for CNS and potential to arrive in blood.The second approach involves screening potential biomarkers through omics studies, with a focus on molecules that exhibit significant changes in PD patients, including those related to amino acid metabolism, lipid metabolism and RNAs.
In the process of searching for blood biomarkers, we still encounter some challenges.First, we discussed the role of α-syn, an important molecule, in disease development.Although current research has not answered the causal relationship between syn and PD, it still holds value as a potential biomarker.A meta-analysis has shown that α-syn can differentiate PD patients from healthy individuals (Zubelzu et al., 2022), but several studies that included the analysis suggested the insufficient diagnostic power.Unlike the CNS, many peripheral cells can produce α-syn (Barbour et al., 2008), making its impact less significant in a population with a large concentration base.Additionally, the larger peripheral blood volume compared with the CNS results in lower detection accuracy due to dilution effects (Manne et al., 2019).Other forms of α-syn, such as phosphorylated α-syn and oligomers, are more closely related to the CNS and show better diagnostic power.The RT-QuIC Assay shows good diagnostic capability in CSF, but its use in blood is unproven.The role of plasmin in shearing α-syn aggregates should be considered (Kim et al., 2012).We can reasonably believe that pathological α-syn can reach the blood directly or by EVs, which has been confirmed (Shi et al., 2014), but our understanding of the protein structure of pathological α-syn may still be unclear (Daniele et al., 2018), which may limit the development of detection methods.
To improve the accuracy of background detection, we focused on NDEs, vesicles carrying CNS-derived proteins, some of which have shown diagnostic value.In addition to EVs, α-syn can directly reach the bloodstream.This suggests that some large molecular proteins have the capability to cross BBB, although we still do not fully understand this process.It could be through weak points in the BBB such as the subfornical organ (SFO) and organum vasculosum of the lamina terminalis (OVLT), where angiotensin II (Ag II) can pass through, or the pathological proteins themselves may possess special abilities to penetrate the BBB.Studying the mechanisms and affecting factors of this communication can enhance the diagnostic accuracy of blood biomarkers because the relationship among levels of blood biomarkers, their levels in CNS and disease development should be ascertained.If the levels in the blood do not change with those in the CNS within a wide concentration range, the diagnostic efficacy will be greatly reduced, especially in the early stages of diseases.Furthermore, studying the interaction of large molecular proteins between the CNS and the blood can lead to new surprises, including new biomarkers and ways for drugs to reach the CNS.
Omics-based screening is another method to consider.However, it is important to ensure that the screening is closely linked to the disease mechanism in order to accurately reflect the disease development.Meanwhile, careful attention must be paid to standardising sampling and detection procedures, considering the disease heterogeneity among PD patients, and addressing data analysis challenges.For example, lipidomics screening is related to GCase.However, its application, similar to GCase activity, may have higher diagnostic efficacy for GBA-PD.
The high heterogeneity and vague preclinical stages of PD patients make it difficult to conduct high-quality cohort studies.This requires us to potentially include subjects early on, continuously collect samples, and not limit ourselves to CSF and blood samples.Different studies have also reported potential biomarkers in various tissues, such as α-syn in saliva, urine and the gastrointestinal tract (Jiménez-Jiménez et al., 2023), which is less invasive compared with blood collection.It is needed more convinced evidence for their diagnostic power, and these biomarkers are resident or from CNS are needed further study.It is important to discover and validate proposed biomarkers through prospective cohort studies.The review also summarises recent prospective cohort studies based on the aforementioned discussion (Table 1).Is it possible that we group based on test results in cohort studies?
Effective biomarkers should be capable of not only diagnosing PD but also distinguishing it from APS.Furthermore, they should have predictive power for prognosis and treatment response in PD.The ability to rapidly, minimally invasively and consistently reflect the condition of PD through peripheral blood tests would be highly desirable.
Combining blood biomarkers for PD diagnosis and prognosis in prospective cohort studies.
T A B L E 1