Introduction

Parkinson’s disease (PD) is the second most common neurodegenerative disease [1]. The traditional cardinal signs of PD are movement disorders, including resting tremor, cogwheel rigidity, bradykinesia, and postural instability [2]. Currently, the diagnosis of PD in Taiwan is mostly based on the National Institute for Health and Care Excellence (NICE) guidelines updated in 2017 [3]. These guidelines cover diagnosing and managing Parkinson's disease in people aged 18 and over. The guidelines state that Parkinson's disease should be suspected if a patient has tremors, stiffness, slowness of movement, balance problems and/or gait disturbances. Untreated patients should be referred quickly to a specialist with expertise in differential diagnosis. The progressive degeneration of dopamine neurons in the substantia nigra of the brain causes Parkinson’s disease [4,5]. In Parkinson disease, the misfolding of α­synuclein monomers in neurons leads to the formation of oligomers and fibrils, which further aggregate into protein inclusions called Lewy bodies [6]. The accumulation of these abnormal insoluble α-synuclein proteins in dopamine neurons leads to the gradual death of cells that secrete the neurotransmitter dopamine [7]. Dopamine is responsible for coordinated motor functions in the body, and when dopamine levels are insufficient, symptoms of Parkinson’s disease can occur.

Assaying α-synuclein in cerebral spinal fluid (CSF) can be used to assess PD, especially in early-stage PD [8-10]. In addition to α-synuclein, pathological evidence of amyloid plaques and tau tangles in the brain has been shown in PD patients [11-14]. This finding implies that amyloid b and tau protein could also be useful biomarkers for assessing PD. Many groups have demonstrated significant differences in the levels of CSF α-synuclein, amyloid b1-42 (Aβ_1-42) and total tau protein (Tau) between PD patients and normal controls (NCs) [15-18]. Eighty percent of PD patients suffer from cognitive impairment or dementia in the lifetime of the disease, so-called Parkinson’s disease dementia (PDD) [19]. It is necessary to identify the mechanisms related to the progression of PDD in PD to reduce its occurrence. Risk factors associated with progressing to PDD in PD patients include older age, male sex, lack of tremor, postural instability, and subtle impairment on cognitive tests [20,21]. Fluid biomarkers may be promising factors that could be used to predict progression to PDD. Although many studies on CSF α-synuclein, Aβ_1-42 or Tau in PD and PDD patients have been performed, the reported predictive power of these CSF biomarkers for predicting progressive cognitive decline in PD patients in numerous independent studies has not been consistent [22-27]. Thus, the feasibility of using CSF biomarkers to discriminate PDD from PD or to predict dementia in PD patients is under investigation. Due to side effects [28,29], lumbar puncture is not easy to use in frequent practice. Progress in exploring the role of CSF biomarkers in PD is limited. In contrast to lumbar puncture, blood collection is easy for routine assessment. The investigation of α-synuclein, Aβ_1-42 or Tau in plasma has attracted the interest of researchers and neurologists. However, the concentrations of these biomarkers in blood are extremely low, at approximately pg/ml or lower. Therefore, ultrasensitive assays are needed to explore these biomarkers in blood. In recent decades, several techniques for detecting proteins at the pg/ml level, such as immunomagnetic reduction (IMR) [30], single molecule array (SIMOA) and immunoprecipitation mass spectrometry (IP-MASS) [31,32], have been developed. Assays of plasma α-synuclein, Aβ_1-42 or Tau levels have become feasible [33]. Independent studies have revealed significant differences in the levels of plasma α-synuclein between PD/PDD patients and NCs [34-39]. The results demonstrate the feasibility of assessing PD/PDD using plasma α-synuclein. In addition to PD/PDD, atypical parkinsonism syndromes (APSs), such as dementia with Lewy bodies (DLB), multiple system atrophy (MSA), progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD), are characterized by increased plasma α-synuclein levels [40]. Lin et al. further showed that additional biomarkers, such as phosphorylated Tau and Aβ_1-42, should be combined with α-synuclein to differentiate NCs from PD/PDD and APS patients [40]. Starhof et al. reported that PSP patients have higher CSF concentrations of C-reactive protein, tumor necrosis factor α, interleukin-1b (IL-1β), IL-4, and IL-6 than PD patients [41]. NCs, PD patients and PSP patients can be differentiated by combining α-synuclein and Tau with other biomarkers. Abnormal concentrations of plasma Aβ_1-42 or Tau in PD/PDD patients have been reported [38,42-44]. The effects of regional atrophy in the brain on the concentrations of plasma Aβ_1-42 and Tau and on cognitive function have been described [38,42]. Furthermore, the prediction of dementia in PD patients using plasma α-synuclein concentrations has been demonstrated [34,37,42]. Although many significant results on plasma biomarkers in PD/PDD patients have been published, comprehensive studies investigating the discriminating power of plasma α-synuclein, Aβ_1-42 and Tau between PD/PDD patients and NCs, and between PDD patients and PD patients are rare. In this study, one hundred eighty-seven NCs, one hundred eighteen PD-NC patients and seventy-nine PDD patients were enrolled at National Taiwan University Hospital, Show Chwan Memorial Hospital, Tri-Service General Hospital, Taipei Veterans General Hospital and Shuang Ho Hospital. Plasma Tau, Aβ_1-42 and α-synuclein levels in each enrolled subject were assayed using IMR. The discrimination between PD-NC/PDD patients and NCs and between PDD and PD-NC patients using plasma biomarkers was investigated. The correlation between the plasma biomarkers and the Mini-Mental State Examination (MMSE) score was examined. In addition, the associations among plasma biomarkers were explored. The age and sex distributions of the plasma biomarkers in the NCs are discussed.

Methods

Recruitment of subjects

Participants in this study were given a medical checklist of major systemic diseases, operations and hospitalizations. Participants reporting uncontrolled medical conditions, including heart failure, recent myocardial infarction (in the past 6 months), malignancy (in the past 2 years), or poorly controlled diabetes (HbA1C > 8.5); a significant history of depression; a history of repeated strokes with stepwise progression; repeated head injury; antipsychotic drug use; definite encephalitis and/or oculogyric crises on no drug treatment; a negative response to large doses of levodopa (if malabsorption was excluded); strictly unilateral features after 3 years; other neurological features (supranuclear gaze palsy, cerebellar signs, early severe autonomic involvement, Babinski sign, early severe dementia with disturbances of language, memory or praxis); exposure to a known neurotoxin; or the presence of cerebral tumors or communicating hydrocephalus on neuroimaging were excluded. Volunteers also received physical examinations. PD and PDD were determined in patients using the United Kingdom Parkinson’s Disease Society Brain Bank diagnostic criteria [45]. Clinical evaluations included a history of the present illness, a family history, a medical history, and a review of systems, with an emphasis on movement disorders, psychiatric illnesses, and cognitive function. The patient evaluations included the Unified Parkinson’s Disease Rating Scale (UPDRS), Hoehn and Yahr stage (H-Y stage), and the MMSE [46]. PD patients were defined as those exhibiting tremor, postural instability or gait disorders. PDD was defined as a diagnosis of idiopathic PD that developed prior to the onset of dementia, with MMSE scores < 26 for more than 1 year [47]. Enrolled subjects without neurodegenerative diseases and without any of the abovementioned exclusion criteria were categorized as normal controls (NCs). This study was approved by the ethics committees of the hospitals. All research was performed in accordance with the relevant guidelines/regulations. Each participant provided informed consent. One hundred eighty-seven NCs, one hundred twenty-eight PD with normal cognition (PD-NC) patients and seventy-nine PDD patients were enrolled at Kaohsiung Chang Gung Memorial Hospital, National Taiwan University Hospital, Sho Chwan Memorial Hospital, Taipei Medical University Shuang-Ho Hospital, Taipei Veterans Hospital, and Tri-Service General Hospital.

Compliance with ethics guidelines

All of the subjects and/or their primary caregivers provided written informed consent prior to their participation in this investigation. This study was approved by the ethics committees and institutional review boards of all the involved hospitals, Kaohsiung Chang Gung Memorial Hospital, National Taiwan University Hospital, Show Chwan Memorial Hospital, Taipei Medical University Shuang-Ho Hospital (201408011), Taipei Veterans Hospital, and Tri-Service General Hospital (TSGHIRB 1-107-05-111). This study was conducted in accordance with the 1964 Declaration of Helsinki and its later amendments.

Preparation of human plasma

Blood was collected in 6-ml or 10-ml K3-EDTA tubes (455036; Greiner Bio-One), and the tubes were immediately inverted gently 10 times. A swing-out (bucket) rotor was used to centrifuge the blood at 20-25 °C and 2,500 × g for 10-15 minutes. Every 1 ml of plasma (supernatant) was transferred to a fresh 1.5 ml Eppendorf tube. All the aliquoted plasma samples were stored at -80 °C within 4.5 hours after blood collection.

Plasma biomarker assays

The tubes with the frozen plasma samples were transferred to ice at -80 °C. After 10-15 minutes, the tubes were moved to room temperature and kept at room temperature for 10-15 minutes. For the Tau/Aβ_1-42/α-synuclein assays, 40/60/40 ml of plasma was mixed with 80/60/80 ml of reagent (MF-TAU-0060, MF-AB2-0060, MF-ASC-0060, MagQu) for IMR measurement. For a given biomarker, measurements were performed in duplicate. The reported concentration of each biomarker was the average value of the duplicated measurements. An IMR analyzer (XacPro-S, MagQu) was used for the concentration measurements. All assays were performed in a blinded manner.

Statistical analysis

Continuous variables for each measurement are presented as the mean ± standard deviation. Continuous variables were compared using t tests, and the p values were determined. Pearson’s correlation, r, was assessed using GraphPad Prism. Receiver operating characteristic (ROC) curve analysis was performed to determine the cutoff value, clinical sensitivity, specificity and area under the curve (AUC).

Results

The demographics of the enrolled subjects are listed in Table 1. The PD group included both the PD with normal cognition (PD-NC) patients and the PD dementia (PDD) patients. First, comparisons of the patient demographics between the NC and PD groups were made. Then, further comparisons between PD-NC and PDD patients were made. The percentage of females was 64.7% in the NC group and 44.2% in the PD group. The subjects in the PD group (65.2 ± 10.3 years) were slightly older than those in the NC group (63.4 ± 6.4 years; p < 0.05). The H-Y stage was zero in the NC group and 2.09 ± 1.10 in the PD group (p < 0.0001). The NCs had significantly greater MMSE scores (28.6 ± 1.5) than did the PD patients (22.4 ± 5.4; p < 0.0001).

The plasma Tau tau levels in the NC group were 18.28 ± 7.54 pg/ml and increased to 24.17 ± 8.81 pg/ml in the PD group (p < 0.0001), as shown in Fig. 1(a). The NC group also had significantly lower plasma Aβ_1-42 levels (15.92 ± 2.23 pg/ml) than did the PD group (16.84 ± 2.62 pg/ml; p < 0.0001), as shown in Fig. 1(b). As shown in Fig. 1(c), the plasma α-synuclein concentration in the PD group was 1.952 ± 6.33 pg/ml, which was much greater than that in the NC group (0.084 ± 0.136 pg/ml; p < 0.0001).

There were 128 PD-NC patients and 79 PDD patients. The PDD patients (69.0 ± 8.6 years) were slightly older than the PD-NC patients (63.0 ± 10.6 years; p < 0.0001). The disease duration in the PD-NC patients was 0.5-6.9 years, whereas it was 1.2-16.5 years for the PDD patients. The H-Y stage in the PD-NC patients was 1.62 ± 0.85, while that for the PDD patients was higher (2.41 ± 1.13; p < 0.0001). In addition, PDD patients had lower MMSE scores (19.5 ± 4.5) than PD-NC patients (26.8 ± 3.3; p < 0.0001). Thus, the severity of movement disorders and dementia was worse in the PDD patients compared with the PD-NC patients.

The effect sizes of the plasma biomarkers among the NC, PD-NC and PDD groups are listed in Table 2. The effect size was obtained by calculating the ratio of the means of the concentrations between two groups. For plasma Tau, the effect size for the PD-NC to NC comparison, denoted PD-NC/NC, was 1.32, and that for the PDD/NC comparison was 1.41. This implies that the concentration of plasma Tau in the PD-NC group was 32% greater than that in the NC group and 42% greater than that in the NC group. Thus, the plasma Tau level increased by approximately 10% in PD-NC patients compared with PDD patients, as listed in Table 2 (PDD/PD-NC). The effect size for the plasma Aβ_1-42 levels for the PD-NC/NC comparison was 1.06 and that for the PDD/NC comparison was 1.05. This finding implies that the concentrations of plasma Aβ_1-42 in PD-NC and PDD patients were increased by approximately 5% compared to that in the NCs. The PD-NC and PDD groups had equivalent levels of plasma Aβ_1-42. Regarding the α-synuclein levels, the levels in PD-NC patients were 7.6-fold higher than those in NCs, and PDD patients had much higher levels, almost 50-fold higher than those in the NCs. Thus, PDD patients had plasma α-synuclein levels 6.35-fold greater than those of PD-NC patients. The results shown in Table 2 suggest that Tau, Aβ_1-42 and α-synuclein are associated with the incidence of PD. Furthermore, α-synuclein and Tau, but not Aβ_1-42, are related to the progression to dementia in PD patients according to the effect size analysis.

The clinical sensitivity and specificity of plasma biomarkers for differentiating NC from PD and PD-NC from PDD were investigated. Through the receiver operating characteristic (ROC) curve analysis using the individual biomarkers shown in Fig. 1(a)-(c), the cutoff value, clinical sensitivity, specificity and area under the curve (AUC) for differentiating PD patients from NCs were determined. The criterion for determining the optimal cutoff value of a biomarker is to find the maximum value of the Youden index, which is calculated as follows: Sensitivity% + Specificity% - 100%. The cutoff plasma Tau concentration for discriminating PD patients from NCs was 20.41 pg/ml, with a corresponding clinical sensitivity of 0.623 (95% CI: 0.553–0.690), a specificity of 0.626 (95% CI: 0.552–0.695) and an AUC of 0.705. The ROC curve is plotted with the black dashed line in Fig. 2(a). The cutoff concentration of plasma Aβ_1-42 for differentiating PD patients from NCs was 16.17 pg/ml, with a sensitivity of 0.585 (95% CI: 0.514–0.652), a specificity of 0.578 (95% CI: 0.503–0.649) and an AUC of 0.614. The ROC curve is plotted with the gray dashed line in Fig. 2(a). Using plasma α-synuclein to discriminate PD patients from NCs, the cutoff concentration was 0.0916 pg/ml. The clinical sensitivity, specificity and AUC were 0.802 (95% CI: 0.741–0.854), 0.733 (95% CI: 0.663–0.795) and 0.872, respectively. The ROC curve is plotted in black in Fig. 2(a). Compared to Tau and Aβ_1-42, plasma α-synuclein showed greater potential for differentiating PD patients from NCs. This result is consistent with the pathogenesis of Parkinson’s disease, which is an α-synucleinopathy.

The cutoff concentrations, clinical sensitivity, specificity and AUC for discriminating PDD patients from PD-NC patients using plasma biomarkers were investigated. Since there was no significant difference in plasma Aβ_1-42 levels between PDD patients and PD-NC patients, Aβ_1-42 was excluded. Plasma Tau showed a cutoff concentration of 21.22 pg/ml for differentiating PDD patients from PD-NC patients, with a corresponding sensitivity of 0.684 (95% CI: 0.569–0.784), a specificity of 0.508 (95% CI: 0.418–0.597) and an AUC of 0.591. The ROC curve is plotted with the dashed line in Fig. 2(b). The solid line in Fig. 2(b) denotes the ROC curve for discriminating PDD patients from PD-NC patients using plasma α-synuclein levels. The cutoff concentration was 0.577 pg/ml. The sensitivity was 0.747 (95% CI: 0.636–0.838), the specificity was 0.750 (95% CI: 0.666–0.822), and the AUC was 0.735. Thus, α-synuclein is a dominant factor for determining progression to dementia in PD patients.

The results in Fig. 2 reveal that the AUCs of plasma Aβ_1-42, Tau tau and α-synuclein for discriminating PD patients from NCs were 0.614, 0.704 and 0.872, respectively. The AUCs of plasma Tau and α-synuclein for discriminating PDD patients from PD-NC patients were 0.591 and 0.735, respectively. This seems to suggest that Tau and α-synuclein are strongly associated with the incidence of PD. Moreover, α-synuclein was the dominant factor contributing to the progression of dementia in PD patients. It is worth investigating the correlation between plasma α-synuclein levels and cognition using scores on tests such as the MMSE. Such correlations for plasma Aβ_1-42 and Tau were also examined.

As listed in Table 3, all of these biomarkers were negatively and moderately correlated with the MMSE score. Notably, plasma Tau showed a relatively strong correlation (r = -0.259, p = 0.0009) with the MMSE scores compared to α-synuclein (r = -0.204, p = 0.0092) and Aβ_1-42 (r = -0.252, p = 0.0012). However, α-synuclein showed a relatively weak correlation with the MMSE scores. Although α-synuclein is the dominant factor related to dementia in PD patients, Tau but not α-synuclein is related to the severity of cognitive impairment in PD patients. This seems to imply that certain proteins act as triggers for dementia, while other proteins may reflect functional deterioration in PD.

Figure 1: Dot plot of plasma (a) Tau (b) Aβ_1-42 and (c) α-synuclein concentrations in the NC (´), PD-NC (·) and PDD (·) groups.

Figure 2: ROC curves for discriminating (a) PD patients from NCs using plasma Tau, Aβ_1-42 and α-synuclein levels and (b) PDD patients from PD-NC patients using plasma Tau and α-synuclein levels.

Discussion

The understanding of brain-derived Tau, Aβ_1-42 and α-synuclein levels in the peripheral blood of PD patients is limited. Due to the development of ultrasensitive assay technologies [31-33], studies on plasma biomarkers of brain-derived Tau, Aβ_1-42 and α-synuclein in PD are currently attracting the interest of researchers. Examinations of plasma biomarkers in PD are continuously reported [34-39,42-44]. One interesting topic is the correlation of biomarker concentrations between blood and CSF. Many studies exploring CSF biomarkers in PD patients have been published [24,26,48,49]. Although the CSF results reported by independent research groups might not be consistent with each other, some of the previously reported findings support the observations in this study.

Buddhala et al. reported significant differences in CSF Aβ_1-42 and α-synuclein levels between NCs and PD-NC patients [48]. This reveals that not only α-synuclein but also Aβ_1-42 play roles in PD. Similar results were also observed with plasma α-synuclein and Ab_1-42­ levels, as shown in Fig. 2(a). Stav et al. reported that various CSF biomarkers, such as Aβ_1-42, Tau and α-synuclein, have different abilities to predict regional atrophy of the brain in PD patients [50]. This finding demonstrates the different roles of biomarkers in aspects of cognitive or motor dysfunctions in PD patients. Hence, the effects of multiple nonindividual biomarkers on brain atrophy and clinical symptoms in PD patients should be considered.

Tau has been shown to promote microtubule polymerization and drive the assembly of tubulin into microtubules, which form the cytoskeleton of neurons and define neuronal morphology [51]. In neurodegenerative diseases, hyperphosphorylated tau inhibits microtubule assembly [52] and the free tau molecules aggregate into paired helical filaments that cause nerve death [53]. Tau is a key plasma biomarker associated with neural damage in the brain. When neurons become damaged or fibrillary, abundant tau protein is released in the brain, leading to a change in the Tau protein concentration in cerebrospinal fluid (CSF) and blood. Moreover, brain atrophy occurs regionally or globally due to neuronal damage. Thus, Tau levels in fluid are significantly correlated with brain atrophy. Lin et al. utilized IMR to assay Tau levels in plasma and used magnetic resonance imaging (MRI) to measure brain atrophy in PD patients [54]. A positive correlation between plasma Tau levels and brain atrophy in PD patients has been reported. Additionally, using IMR and MRI, Chiu et al. and Liu et al. reported a negative correlation between plasma Tau levels and brain volume in patients with amnesic mild cognitive impairment and Alzheimer’s disease [55,56]. On the other hand, Chen et al.et al. revealed a negative correlation between plasma α-synuclein levels and brain atrophy, especially in the region of the limbic cortex [57]. These results imply that other plasma biomarkers in addition to Tau play roles in brain atrophy in PD patients. Further explorations are needed to clarify the roles of biomarkers in brain atrophy in PD patients.

Skogseth et al. revealed an association between reduced CSF α-synuclein concentrations and cognition in PD patients and suggested that pathological α-synuclein contributes to early cognitive impairment in PD patients [24]. The results support the discrimination between PDD and PD-NC using α-synuclein, as shown in Fig. 2(b). Consistent results for biomarkers between CSF and blood were not only found in this work. Schirinzi et al. reported that brain-derived Tau, Aβ_1-42 and α-synuclein may change correspondingly in both the peripheral blood and CSF of PD patients [49].

In fact, there are inconsistent results for biomarkers between blood and CSF. Yousaf et al. showed that both CSF Aβ_1-42 and Tau levels significantly differ in PD-NC and PDD patients [26]. However, the results in Figs. 1(a) and (b) show that only plasma Tau not the levels of Ab­_1-42, significantly differed between the PDD and PD-NC groups. More explorations to correlate blood biomarkers with CSF biomarkers are needed.

Even among studies on CSF biomarkers, inconsistent changes in CSF α-synuclein levels have been found in PD patients among cohorts. Several studies have shown that CSF α-synuclein levels are lower in PD patients than in NCs [58-60]. Other studies have reported that there is no significant difference in CSF α-synuclein levels between PD patients and NCs [61,62]. These inconsistencies might be attributed to several factors, such as the CSF preparation method, assay technologies, the antibodies used in the assays, and the subjects enrolled. A standard protocol from sample preparation to data analysis should be developed to ensure consistent results among studies.

IMR assay kits and the analyzer for α-synuclein, Aβ_1-42 and Tau in human plasma have been registered with CE IVDD. Standard processes, including blood draw, plasma preparation, Good-Manufacture-Production grade reagents/calibrators/controls, assay steps and data analysis, were established. Consistent results in assaying plasma α-synuclein, Aβ_1-42 and Tau using the IMR assay among studies on PD have been reported. A definite increase in plasma α-synuclein and Tau levels in PD patients has been shown independently by many groups [34,36-38,42,63]. This reveals the importance of standardization for assays.

As listed in Table 3, plasma Tau levels were relatively strongly correlated (r = -0.259, p = 0.0009) with the MMSE scores compared to α-synuclein (r = -0.204, p = 0.0092) and Aβ_1-42 (r = -0.252, p = 0.0012) levels. In a previous study, correlations between high plasma Tau levels and cognitive evaluation scores (the MMSE and clinical dementia rating) and the severity of PD (UPDRS) were detected, further supporting the role of tau in neuronal degeneration. Similar to the case of Alzheimer’s disease (AD), there is an obvious change in the Aβ_1-42 level in prodromal AD patients compared to that in NCs, followed by a nearly stable level of plasma Aβ_1-42 as the disease progresses to mild or severe AD [64]. However, Tau levels continue to increase as cognition declines in individuals with AD [52,65].

According to previous studies [38,42,43,57], in addition to the MMSE scores, plasma Tau levels are associated with executive dysfunction and regional atrophy in the brain of PD patients. Several groups have reported a negative correlation between plasma Tau levels and MMSE scores [65], between plasma Tau levels and hippocampal volume [55], and between plasma Tau levels and cortical thickness in AD patients [66]. Therefore, Tau levels are a promising index reflecting the severity of neuronal damage and cognitive impairment in neurodegenerative diseases.

With the measured concentrations in NCs, the age dependency and sex-dependent dependency of biomarker concentrations were explored. In the NC group, there were 121 female subjects and 66 male subjects. All NC subjects were either middle-aged or elderly (³ 50 years old). The correlations between age and plasma Tau, Aβ_1-42 and α-synuclein levels were analyzed for females and males, as listed in Table 4. The Pearson’s correlation coefficients were within the -0.110 to 0.114 range, and all the p values were greater than 0.05. All plasma Tau, Aβ_1-42 and α-synuclein levels were independent of age in middle-aged and elderly NCs, regardless of sex. The age independence of plasma Tau, Aβ_1-42 and α-synuclein levels in the NC group is consistent with previously reported results [67]. Hence, the measured concentrations of plasma Tau, Aβ_1-42 and α-synuclein do not need to be adjusted for age for discussion.

The concentrations of plasma Tau, Aβ_1-42 and α-synuclein in female and male NCs are listed separately in Table 5. There was no significant difference in age between the females (62.9 ± 6.1 years) and males (64.3 ± 6.5 years; p > 0.05). Among the three plasma biomarkers, only α-synuclein was significantly different between females (0.0064 ± 0.054 pg/ml) and males (0.120 ± 0.213 pg/ml; p < 0.05). The reason for the increase in plasma α-synuclein levels in males compared to females is not clear. Further studies should be performed to clarify this difference. The fact that plasma Tau tau and Aβ_1-42 levels are not sex-dependent in the NC group is consistent with that reported in a previous paper [68]. Hence, the measured concentrations of plasma Tau and Aβ_1-42 do not need to be adjusted according to sex for discussion.

The correlation between the H-Y stage and plasma biomarker levels was investigated in PD patients. Because of the small sample size, PD patients with H-Y stage 5 disease were not included. The Pearson correlation coefficients (r) and p values between the plasma biomarkers and H-Y stage (0.5-4) in PD patients are listed in Table 6. For α-synuclein, Aβ_1-42 and Tau, there was no significant correlation with the H-Y stage. In addition to these three biomarkers, Lin et al. reported that the plasma level of Ser129-phosphorated α-synuclein was significantly and positively correlated with the severity of motor disorders [69].

The correlation between plasma biomarkers was analyzed. The results are listed in Table 7. In both NCs and PD patients, plasma Tau levels were significantly and moderately positively correlated with plasma Aβ_1-42 levels. Plasma Tau concentrations also showed a significant and moderately positive correlation with plasma α-synuclein concentrations in the NC group but not in the PD group. The plasma Aβ_1-42 concentration did not significantly correlate with the plasma α-synuclein concentration in the NCs or PD patients.

The results shown in Fig. 2(a) reveal that plasma Aβ_1-42 is less crucial to the incidence of PD than plasma Tau and α-synuclein are. In addition, PDD patients and PD-NC patients have equivalent levels of plasma Aβ_1-42. Thus, Aβ_1-42 does not play a key role in PD. Similar to the results of Chojdak-Lukasiewicz et al., plasma Aβ_1-42 levels cannot be used to differentiate NCs from PD patients with a disease duration of less than 5 years, as they can only discriminate NCs from PD patients with a disease duration of more than 5 years [70]. Regarding plasma Tau and α-synuclein concentrations, significant differences between PD patients and NCs and between PDD patients and PD-NCs were found, as shown in Table 1 and Figs. 2(a) and (b). This finding implies that Tau and α-synuclein are more important for the incidence of PD and dementia in PD. However, plasma Tau is not correlated with plasma α-synuclein in PD patients. This finding suggests that α-synuclein should be regarded as an independent biomarker in PD. It is possible that there are two mechanisms associated with these two biomarkers leading to PD and dementia. Therefore, not only α-synuclein but also Tau should be considered when treating PD.

In published papers, plasma biomarkers are promising for differentiating PDD patients from PN-NC patients. However, the reported results among studies are not consistent. For example, Lin et al. used IMR for assaying plasma α-synuclein and phosphorylated α-synuclein levels [34,69]. α-synuclein was found to be able to discriminate PDD from PD-NC patients, while phosphorylated α-synuclein enabled differentiation of motor disorder in PD patients. The ability to differentiate PDD from PD-NC patients using plasma α-synuclein levels was also validated by Chang et al. using IMR [63]. Syafrita et al. used enzyme-linked immunoassay (ELISA) for assaying plasma Aβ_1-42, α-synuclein and Tau in PD-NC and PDD patients [71]. It was found that plasma Aβ_1-42 levels significantly discriminate PDD from PD-NC patients, whereas plasma α-synuclein and Tau levels cannot. The results reported by Syafrita et al. are inconsistent with those of this study and Lin et al. Mizutani et al. used single molecule array (SIMOA) to quantitatively detect plasma biomarkers for predicting cognitive impairment in PD patients [72]. Plasma glial fibrillary acidic protein (GFAP) and neurofilament light chain (NfL), but not Aβ_1-42 or phosphorylated Tau (pTau181), were able to discriminate PDD patients from PD-NC patients. The feasibility of differentiating PDD patients from PD-NC patients using plasma NfL levels assayed with IMR was independently demonstrated by Chen et al. and Chen et al. [38,73]. In addition to soluble plasma biomarkers, Blommer et al. utilized electrochemiluminescence assays to quantify α-synuclein in extracellular vesicles (EVs) in human blood [74]. A significant difference in EV α-synuclein levels was detected between PD-NC and PDD patients. By using IMR, Chung et al. reported significant increases in EV Aβ_1-42 and Tau levels in PDD patients but not in PD-NC patients [44]. According to published papers, some groups have revealed the role of plasma α-synuclein in differentiating PDD from PD-NC patients, while other groups have reported the prediction of cognitive impairment in PD patients using plasma Aβ_1-42; however, other groups have proposed utilizing plasma for discriminating PDD from PD-NC patients. Among biomarkers, plasma α-synuclein is the most dominant biomarker for predicting cognitive impairment in PD patients. However, other biomarkers contribute to cognitive impairment in PD patients.

Table 1: Demographics and plasma biomarker concentrations of the enrolled subjects.

Table 2: Effect sizes of plasma biomarkers among the NC, PD-NC and PDD groups.

Table 3: Pearson’s correlation coefficient (r) and p values between the plasma biomarkers and MMSE scores.

Table 4: Pearson’s correlation coefficient (r) and p values between the plasma biomarkers and age in the NC group.

Table 5: Age and concentrations of plasma , Aβ_1-42, and α-synuclein in female and male NCs.

Table 6: Pearson’s correlation coefficient (r) and p values between the plasma biomarkers and H-Y stage (0.5-4) in PD patients.

Table 7: Correlations between biomarkers.

Limitations

In this study, PD patients were diagnosed mainly using the United Kingdom Parkinson’s Disease Society Brain Bank diagnostic criteria. Enrolled subjects were not examined with CSF biomarker assays. PD patients did not undergo neuroimaging examinations, namely, MRI or 18-FDG-PET, to exclude other differential diagnoses, i.e., vascular parkinsonism, or to achieve a more accurate differential diagnosis between PDD and dementia with Lewy bodies.

Conclusion

Plasma Aβ_1-42, Tau and α-synuclein levels were significantly greater in PD patients than in NCs. Further increases in plasma Tau and α-synuclein levels were found in PDD patients compared to PD-NC patients. Among these three biomarkers, α-synuclein showed relatively strong discriminating power between PD patients and NCs, as well as between PDD patients and PD-NC patients. Tau is the second most abundant protein. Tau showed a relatively strong correlation with MMSE scores. All three protein levels were found to be independent of age in the NCs aged 50 years and older. These results suggest that both Tau and α-synuclein play roles in the occurrence of PD and dementia in PD. However, Tau levels are not associated with α-synuclein levels in PD patients, which implies that Tau and α-synuclein should be regarded as independent biomarkers of PD.

Data availability

The dataset generated and analyzed in the current study is available from the corresponding author upon reasonable request.

Acknowledgments

We thank MagQu Co., Ltd., for their assistance with the plasma biomarker assays.

Author contributions

This study was designed by H.C. Liu, S.Y.Y. and W.C.L. and was executed by P.Y.C., C.H.L., M.J.C., C.J.H., P.N.W., T.F.C., F.C.Y., L.K.H. and C.H.L. S.Y.Y. prepared the manuscript. All authors approved the final manuscript.

Funding

This study was supported by the Ministry of Science and Technology, Taiwan (CMRPG8K0221, CORPG8L0191, 107-2321-B-039-004, 106-2314-B-038-001, 107-2314-B-038-050, 108-2314-B-016-023, and 108-2314-B-016-020).

Competing interests

H.C. Liu is an employee of MagQu Co., Ltd. S.Y. Yang is a shareholder and an employee of MagQu Co., Ltd. The other authors report no disclosures.

References

1. Pringsheim T, Jette N, Frolkis A, et al. The Prevalence of Parkinson’s Disease: A Systematic Review and Meta-analysis. Move Disorders 29 (2014): 1583-1590.
2. Kirollos C, O'Neill CJ, Dobbs RJ, et al. Quantification of the Cardinal Signs of Parkinsonism and of Associated Disability in Spouses of Sufferers. Age Ageing 22 (1993): 20-26.
3. NICE Guideline. Parkinson’s disease in adults: diagnosis and management. London: National Institute for Health and Care Excellence (NICE) (2017).
4. Spillantini MG, Schmidt ML, Lee VM, et al. Alphα-synuclein in Lewy Bodies. Nature 388 (1997): 839-940.
5. Lang AE, Lozano AM. Parkinson's Disease. First of Two Parts. N Engl J Med 1339 (1998): 1044-1053.
6. Wakabayashi K, Tanji K, Odagiri S, et al. The Lewy Body in Parkinson's Disease and Related Neurodegenerative Disorders. Mol Neurobiol 47 (2013): 495-508.
7. Lotharius J, Brundin P. Impaired Dopamine Storage Resulting from Alphα-synuclein Mutations May Contribute to the Pathogenesis of Parkinson's Disease. Hum Mol Genet 11 (2002): 2395-2407.
8. Tokuda T, Qureshi MM, Ardah MT, et al. Detection of Elevated Levels of α-synuclein Oligomers in CSF from Patients with Parkinson Disease. Neurology 75 (2010): 1766-1770.
9. Mollenhauer B, Trautmann E, Taylor P, et al. Total CSF α-synuclein is Lower in de novo Parkinson Patients than in Healthy Subjects. Neurosci Lett 532 (2013): 44-48.
10. Lim EW, Aarsland D, Ffytche D, et al. Amyloid-β and Parkinson's Disease. J Neurol 266 (2019): 2605-2619.
11. Braak H, Braak, E. Cognitive Impairment in Parkinson's Disease: Amyloid Plaques, Neurofibrillary Tangles, and Neuropil Threads in the Cerebral Cortex. J Neural Transm Park Dis Dement Sect 2 (1990): 45-47.
12. Lei P, Ayton S, Finkelstein DI, et al. Tau Protein: Relevance to Parkinson's Disease. Internl J Biochem Cell Biol 42 (2010): 1775-1778.
13. Zhang X, Gao F, Wang D, et al. Tau Pathology in Parkinson's Disease. Front Neurol 9 (2018): 809.
14. Chalatsa I, Melachroinou K, Emmanouilidou E, et al. Assessment of Cerebrospinal Fluid α-synuclein as a Potential Biomarker in Parkinson’s Disease and Synucleinopathies. Neuroimmunol Neuroinflammation 7 (2020): 132-140.
15. Mollenhauer B, Cepek L, Bibl M, et al. Tau Protein, Aβ42 and S-100B Protein in Cerebrospinal Fluid of Patients with Dementia with Lewy Bodies. Dement Geriatr Cogn Disord 19 (2005): 164-170.
16. Bibl M, Mollenhauer B, Esselmann H, et al. CSF Amyloid-beta-peptides in Alzheimer's Disease, Dementia with Lewy Bodies and Parkinson's Disease Dementia. Brain 129 (2006): 1177-1187.
17. Alves G, Brønnick K, Aarsland D, et al. CSF Amyloid-β and Tau Proteins, and Cognitive Performance, in Early and Untreated Parkinson's Disease: the Norwegian Park West Study. J Neurol Neurosur Psychi 81 (2010): 1080-1086.
18. Montine T, Shi M, Quinn JF, et al. CSF Aβ(42) and Tau in Parkinson's Disease with Cognitive Impairment. Mov Disord 25 (2010): 2682-2685.
19. Aarsland D, Andersen K, Larsen JP, et al. Prevalence and Characteristics of Dementia in Parkinson Disease: an 8-year Prospective Study. Arch Neurol 60 (2003): 387-392.
20. Stern Y, Marder K, Tang MX, et al. Antecedent Clinical Features Associated with Dementia in Parkinson’s Disease. Neurology 43 (1993): 1690-1692.
21. Uc EY, McDermott MP, Marder KS, et al. Incidence of and Risk Factors for Cognitive Impairment in an Early Parkinson Disease Clinical Trial Cohort. Neurology 73 (2009): 1469-1477.
22. Siderowf A, Xie SX, Hurtig H, et al. CSF Amyloid b 1-42 Predicts Cognitive Decline in Parkinson Disease. Neurology 75 (2010): 1055-1061.
23. Stewart T, Liu C, Ginghina C, et al. Cerebrospinal Fluid α-synuclein Predicts Cognitive Decline in Parkinson Disease Progression in the DATATOP Cohort. Am J Pathol 184 (2014): 996-975.
24. Skogseth RE, Bronnick K, Pereira JB, et al. Associations between Cerebrospinal Fluid Biomarkers and Cognition in Early Untreated Parkinson's Disease. J Parkinsons Dis 5 (2015): 783-792.
25. Schrag A, Siddiqui UF, Anastasiou Z, et al. Clinical Variables and Biomarkers in Prediction of Cognitive Impairment in Patients with Newly Diagnosed Parkinson's Disease: a Cohort Study. Lancet Neurol 16 (2017): 66-75.
26. Yousaf T, Pagano G, Niccolini F, et al. Predicting Cognitive Decline with Non-Clinical Markers in Parkinson's Disease (PRECODE-2). J Neurol 266 (2019): 1203-1210.
27. Baek MS, Lee MJ, Kim HK, et al. Temporal Trajectory of Biofluid Markers in Parkinson's Disease. Sci Rep 11 (2021): 14820.
28. Evans RW. Complications of Lumbar Puncture. Neurol Clinics 16 (1998): 83-105.
29. Kruesi MJ, Swedo SE, Coffey ML, et al. Objective and Subjective Side Effects of Research Lumbar Punctures in Children and Adolescents. Psychi Res 25 (1998): 59-63.
30. Chiu MJ, Horng HE, Csieh JJ, et al. Multi-channel SQUID-based Ultrahigh-Sensitivity in-vitro Detections for Bio-markers of Alzheimer’s Disease via Immunomagnetic Reduction. IEEE Trans Appl Supercond 21 (0211): 477-480.
31. Rissin DM, Kan CW, Campell TG, et al. Single-molecule Enzyme-linked Immunosorbent Assay Detects Serum Proteins at Subfemtomolar Concentrations. Nat Biotechnol 28 (2010): 595-599.
32. Nakamura A, Kaneko N, Villemagne VL, et al. High Performance Plasma Amyloid-β Biomarkers for Alzheimer’s Disease. Nature 554 (2018): 249-254.
33. Chiu PY, Yang FC, Chiu MJ, et al. Relevance of Plasma Biomarkers to Pathologies in Alzheimer’s Disease, Parkinson’s Disease and Frontotemporal Dementia. Sci Reports 12 (2022): 17919-1-9.
34. Lin CH, Yang SY, Horng HE, et al. Plasma α-synuclein Predicts Cognitive Decline in Parkinson's Disease. J Neurol Neurosurg Psychi 88 (2017): 818-824.
35. Wang HL, Lu CS, Yeh TH, et al. Combined Assessment of Serum Alphα-synuclein and Rab35 is a Better Biomarker for Parkinson's Disease. J Clin Neurol 15 (2019): 488-495.
36. Lin WC, Lu CH, Chiu PY, et al. Plasma Total α-synuclein and Neurofilament Light Chain: Clinical Validation for Discriminating Parkinson’s Disease from Normal Control. Dement Geriatr Cogn Disord 49 (2020): 401-409.
37. Chang CW, Yang SY, Yang CC, et al. Plasma and Serum Alphα-synuclein as a Biomarker of Diagnosis in Patients with Parkinson's Disease. Front Neurol 10 (2020): 1388.
38. Chen PC, Yu CC, Chen YS, et al. The Potential Effects of Oxidative Stress-related Plasma Abnormal Protein Aggregate Levels on Brain Volume and its Neuropsychiatric Consequences in Parkinson’s Disease. Oxidative Med Cell Longevity 2021 (2021): 366327.
39. Chung CC, Chan L, Chen JH, et al. Plasma Extracellular Vesicle α-synuclein Level in Patients with Parkinson's Disease. Biomol 11 (2021): 744.
40. Lin CH, Yang SY, Horng HE, et al. Plasma Biomarkers Differentiate Parkinson’s Disease from Atypical Parkinsonism Syndromes. Front Aging Neurosci 10 (2018): 123-1-9.
41. Starhof C, Winge K, Heegaard NHH, et al. Cerebrospinal Fluid Pro-inflammatory Cytokines Differentiate Parkinsonian Syndromes. J Neuroinflammation 15 (2018): 305-1-7.
42. Yu CC, Lu CY, Chen MH, et al. Brain Atrophy Mediates the Relationship between Misfolded Proteins Deposition and Cognitive Impairment in Parkinson's Disease. J Pers Med 11 (2021): 702.
43. Lin WT, Shaw JS, Cheng FY, et al. Plasma Total Tau Predicts Executive Dysfunction in Parkinson's Disease. Acta Neurol Scand 2021 (2021): 1-8.
44. Chung CC, Chan L, Chen JH, et al. Plasma Extracellular Vesicles Tau and β-amyloid as Biomarkers of Cognitive Dysfunction of Parkinson's Disease. FASEB J 35 (2021): e21895.
45. Hughes AJ, Daniel SE, Kilford L, et al. Accuracy of Clinical Diagnosis of Idiopathic Parkinson's Disease: a Clinic-pathological Study of 100 Cases. J Neurol Neurosurg Psychi 55 (1992): 181-184.
46. Folstein MF, Folstein SE, McHugh PR. Mini-Mental State: A Practical Method for Grading the Cognitive State of Patients for the Clinician. J Psychi Res 12 (1975): 189-198.
47. Emre M, Aarsland D, Brown R, et al. Clinical Diagnostic Criteria for Dementia Associated with Parkinson's Disease. Mov Disord 22 (2007): 689-707.
48. Buddhala C, Campbell MC, Perlmutter JS, et al. Correlation between Decreased CSF α-synuclein and Aβ_1-42in Parkinson Disease. Neurobiol Aging 36 (2015): 476-484.
49. Schirinzi T, Zenuni H, Grillo P, et al. Tau and amyloid-β Peptides in Serum of Patients with Parkinson's Disease: Correlations with CSF Levels and Clinical Parameters. Front Neurol 13 (2022): 748599-1-5.
50. Stav AL, Johanse KK, Auning E, et al. Hippocampal Subfield Atrophy in Relation to Cerebrospinal Fluid Biomarkers and Cognition in Early Parkinson's Disease: a Cross-sectional Study. NPJ Parkinsons Dis 14 (2016): 15030-1-7.
51. Gendron TF, Petrucelli L. The Role of Tau in Neurodegeneration. Mol Neurodegen 4 (2009): 13-1-19.
52. Busciglio J, Lorenzo A, Yeh J, et al. Beta-amyloid Fibrils Induce Tau Phosphorylation and Loss of Microtubule Binding. Neuron 14 (1995): 879-888.
53. Alonso AC, Grundke-Iqbal I, Iqbal K. Alzheimer's Disease Hyperphosphorylated Tau Sequesters Normal Tau into Tangles of Filaments and Disassembles Microtubules. Nat Med 2 (1996): 783-787.
54. Lin WC, Lee PL, Lu CH, et al. Linking Stage-specific Plasma Biomarkers to Gray Matter Atrophy in Parkinson Disease. AJNR Am J Neuroradiol 42 (2021): 1444-1451.
55. Chiu MJ, Chen YF, Chen TF, et al. Plasma Tau as a Window to the Brain-negative Associations with Brain Volume and Memory Function in Mild Cognitive Impairment and Early Alzheimer's Disease. Hum Brain Mapp 35 (2014): 3132-3142.
56. Liu HC, Chen HH, Ho CS, et al. Investigation of the Number of Tests Required for Assaying Plasma Biomarkers Associated with Alzheimer’s Disease Using Immunomagnetic Reduction. Neurol Ther 10 (2021): 1015-1028.
57. Chen CH, Lee BC, Lin CH. Integrated Plasma and Neuroimaging Biomarkers Associated with Motor and Cognition Severity in Parkinson’s Disease. J Parkinsons Dis 10 (2020): 77-88.
58. Wennström M, Surova Y, Hall S, et al. Low CSF Levels of Both α-synuclein and the α-synuclein Cleaving Enzyme Neurosin in Patients with Synucleinopathy. PLoS One 8 (2013): e53250-1-8.
59. van Dijk KD, Bidinosti M, Weiss A, et al. Reduced α-synuclein Levels in Cerebrospinal Fluid in Parkinson's Disease are Unrelated to Clinical and Imaging Measures of Disease Severity. Euro J Neurol 21 (2014): 388-394.
60. Majbour N, Aasly J, Abdi I, et al. Disease-associated α-synuclein Aggregates as Biomarkers of Parkinson Disease Clinical Stage. Neurol 99 (2022): e2417-e2427.
61. Brockmann K, Quadalti C, Lerche S, et al. Association between CSF Alphα-synuclein Seeding Activity and Genetic Status in Parkinson’s Disease and Dementia with Lewy Bodies. ACTA Neuropathol Commun 9 (2021): 175-1-11.
62. Agnello L, Lo Sasso B, Vidali M, et al. Evaluation of Alphα-synuclein Cerebrospinal Fluid Levels in Several Neurological Disorders. J Clin Med 31 (2022): 3139-1-7.
63. Chang KH, Liu CH, Wang YR, et al. Upregulation of APAF1 and CSF1R in Peripheral Blood Mononuclear Cells of Parkinson’s Disease. Int J Mol Sci 24 (2023): 7095-1-12.
64. Chiu MJ, Yang SY, Horng HE, et al. Combined Plasma Biomarkers for Diagnosing Mild Cognition Impairment and Alzheimer's Disease. ACS Chem Neurosci 4 (2013): 1530-1536.
65. Tsai CL, Liang CS, Lee JT, et al. Associations between Plasma Biomarkers and Cognition in Patients with Alzheimer’s Disease and Amnestic Mild Cognitive Impairment: A Cross-sectional and Longitudinal Study. J Clin Med 8 (2019): 1893.
66. Fan LY, Tzen KY, Chen YF, et al. The Relation between Brain Amyloid Deposition, Cortical Atrophy, and Plasma Biomarkers in Amnesic Mild Cognitive Impairment and Alzheimer’s Disease. Front Aging Neurosci 10 (2018): 175.
67. Lue LF, Pai MC, Chen TF, et al. Age-dependent Relationship between Plasma Ab40 and Ab42 and Total Tau Levels in Cognitively Normal Subjects. Front Aging Neurosci 11 (2019): 222.
68. Hu CJ, Chiu MJ, Pai MC, et al. Assessment of High Risk for Alzheimer’s Disease Using Plasma Biomarkers in Subjects with Normal Cognition in Taiwan: A Preliminary Study. J Alzheimers Dis Reports 5 (2021): 761.
69. Lin CH, Liu HC, Yang SY, et al. Plasma pS129-α-synuclein is a Surrogate Biofluid Marker of Motor Severity and Progression in Parkinson’s Disease. J Clin Med 8 (2019): 1601-1-11.
70. Chojdak-Lukasiewicz J, Malodobra-Mazur M, Zimny A, et al. Plasma Tau Protein and Aβ42 Level as Markers of Cognitive Impairment in Patients with Parkinson's Disease. Adv Clin Exp Med 29 (2020): 115-121.
71. Syafrita Y, Istarini A, Busra M, et al. Relationship between Plasma Level of Beta-amyloid, Alphα-synuclein, and Tau Protein with Cognitive Impairment in Parkinson Disease. J Med Sci 10 (2022): 663-667.
72. Mizutani Y, Ohdake R, Tatebe H, et al. Associations of Alzheimer's-related Plasma Biomarkers with Cognitive Decline in Parkinson's Disease. J Neurol 270 (2023): 5461-5474.
73. Chen NC, Chen HL, Li SH, et al. Plasma Levels of α-synuclein, Aβ-40 and T-tau as Biomarkers to Predict Cognitive Impairment in Parkinson’s Disease. Front Aging Neurosci 12 (2020): 112-1-8.
74. Blommer J, Pitcher T, Mustapic M, et al. Extracellular Vesicle Biomarkers for Cognitive Impairment in Parkinson’s Disease. Brain 146 (2023): 195-208.