Associations of DTI-ALPS index and choroid plexus volume with clinical severity across the Parkinson’s disease spectrum

Abstract

Impaired clearance of pathogenic proteins may contribute to Parkinson’s disease (PD) progression, but the clinical relevance of brain clearance-associated processes in humans remains incompletely understood. Using cross-sectional data from 1,861 participants (704 PD, 997 prodromal, 160 healthy controls) in Parkinson Progression Marker Initiative, we investigated whether the magnetic resonance imaging (MRI)-based indirect markers linked to brain clearance-related processes, diffusion tensor imaging analysis along the perivascular space (DTI-ALPS) index and normalized choroid plexus volume (NCPV), are associated with disease stage and motor/non-motor manifestations across the PD spectrum. The ALPS index declined with age, was lower in males, and showed a stepwise reduction from controls to prodromal individuals and PD patients. Clinically, lower ALPS index correlated with greater motor severity in prodromal and PD groups, with stronger associations at advanced stages. Lower ALPS index also correlated with rapid eye movement sleep behavior disorder, cognitive impairment, and depressive symptoms. NCPV showed a complementary trend, demonstrating positive associations with clinical severity measures and a significant negative correlation with the ALPS index. Together, these findings suggest that MRI markers linked to brain clearance processes are associated with clinical progression across the PD continuum and may provide imaging biomarkers relevant to disease staging and pathophysiology.

Introduction

Parkinson’s disease (PD) is a progressive neurodegenerative disorder characterized pathologically by the accumulation of misfolded α-synuclein and clinically by motor and non-motor manifestations that evolve over decades¹. Although extensive research has focused on α-synuclein production, aggregation, and cell-to-cell propagation, comparatively less attention has been paid to the mechanisms responsible for its clearance from the human brain². Increasing evidence suggests that an imbalance between α-synuclein production and elimination, rather than overproduction alone, may be a critical determinant of PD onset and progression³.

The glymphatic system is a brain-wide perivascular network that facilitates cerebrospinal fluid (CSF)-interstitial fluid exchange and contributes to the clearance of metabolic waste and aggregation-prone proteins⁴. Experimental studies have demonstrated that glymphatic transport is influenced by aging, sleep, and vascular integrity and participates in the removal of amyloid-β and tau^5,6. Emerging evidence suggests that α-synuclein may also be cleared through glymphatic and meningeal lymphatic pathways^5,7, raising the possibility that impaired clearance contributes to α-synucleinopathy. However, in vivo evidence linking brain clearance-related processes to clinical severity and disease stage in large PD cohorts remains limited.

Diffusion tensor image analysis along the perivascular space (DTI-ALPS) provides a noninvasive and indirect imaging marker associated with perivascular fluid dynamics and has shown sensitivity to aging and neurodegenerative diseases^8,9. Several small studies have reported reduced ALPS indices in PD^9–12. However, whether ALPS index alterations are present during prodromal stages, how they relate to motor and non-motor symptoms, and whether they are associated with regional neurodegeneration remain unclear.

Additionally, the choroid plexus (ChP) is a crucial structure in the central nervous system (CNS) responsible for producing approximately 60-80% of CSF and forming the blood-CSF barrier^13,14, thereby contributing to CNS homeostasis and mediating central-peripheral immune interactions¹⁵. Recent studies have indicated that choroid plexus volume (CPV) increases with age and is associated with the severity of neurodegenerative diseases, including PD^16–18, suggesting that alterations in CPV may reflect underlying pathological processes. Structural and functional alterations in the ChP may also influence CSF-mediated brain clearance processes that are related to glymphatic-associated processes, although the mechanisms underlying these interactions remain to be further elucidated.

In this study, we investigated imaging markers associated with brain clearance processes across the PD spectrum using the ALPS index and normalized choroid plexus volume (NCPV) in healthy controls (HCs), prodromal individuals, and patients with established PD. We examined the effects of age, sex, and disease stage on both the ALPS index and NCPV and assessed their associations with cortical thickness, motor severity, non-motor symptoms, and clinical staging. We further tested whether these imaging markers mediate the relationship between aging and clinical impairment. By integrating neuroimaging and clinical phenotyping in a large cohort, this study aims to clarify the clinical relevance of imaging markers associated with brain clearance processes in PD.

Results

Cohort characteristics

After excluding 26 participants due to poor image quality, missing key metrics, or incomplete imaging data, 1861 participants from the Parkinson's Precision Medicine Initiative (PPMI) cohort were included in the final analysis (Fig. 1). The cohort comprised 704 PD patients, 997 prodromal individuals, and 160 HCs. Prodromal individuals were defined according to the PPMI study criteria and included individuals with established prodromal markers, most commonly polysomnography-confirmed rapid eye movement sleep behavior disorder (RBD) and hyposmia, without clinically manifest parkinsonism at enrollment. Demographic and clinical characteristics of the study participants are summarized in Table 1.

Fig. 1. Flowchart of participant selection from the PPMI cohort.

Participants with PD (n = 1920), prodromal individuals (n = 4479), and HC (n = 440) were identified from the PPMI database. After applying DTI availability and quality control, 704 PD, 997 prodromal, and 160 HC participants were retained. Subsequent inclusion of subjects with high-resolution 3D T1-weighted MRI (3D T1WI) yielded the final analytical cohort of 439 PD, 475 prodromal, and 114 HC participants.

Table 1.

Demographic and clinical characteristics of the participants

	All (n = 1861)	HC (n = 160)	PD (n = 704)	Prodromal (n = 997)
Sex (Male/Female)	1073/788	92/68	447/257	534/463
Age (years)	66.074 ± 8.168	63.075 ± 11.277	64.970 ± 9.225*	67.335 ± 6.351***
Age (range)	30 ～ 86	30 ～ 85	35 ～ 86	38 ～ 86
UPDRS-I	NA	NA	5.078 ± 2.759	5.283 ± 3.899
UPDRS-II	NA	NA	7.660 ± 4.856	2.275 ± 3.605***
UPDRS-III	NA	NA	22.613 ± 10.198	5.059 ± 6.126***
UPDRS-IV	NA	NA	1.174 ± 2.280	NA
UPDRS-total	NA	NA	36.203 ± 14.163	12.603 ± 10.509***
MoCA	NA	NA	27.053 ± 2.694	26.597 ± 2.462**
RBDSQ†	NA	1.600 ± 1.258	3.017 ± 1.670***	NA
GDS-15	NA	NA	5.707 ± 1.649	5.439 ± 1.212***
UPSIT	NA	NA	22.838 ± 7.670	22.391 ± 7.928
STAI-State	NA	NA	44.541 ± 5.139	48.446 ± 15.646***
STAI-Trait	NA	NA	43.981 ± 3.908	46.443 ± 3.536***
Left ALPS index	1.239 ± 0.150	1.496 ± 0.168	1.175 ± 0.126***	1.242 ± 0.113***
Right ALPS index	1.230 ± 0.137	1.466 ± 0.135	1.165 ± 0.110***	1.238 ± 0.108***
Mean ALPS index	1.234 ± 0.131	1.481 ± 0.138	1.170 ± 0.103***	1.240 ± 0.095***
NCPV#	1.034 ± 0.200	0.909 ± 0.139	1.123 ± 0.219***	0.975 ± 0.154***

^* represents the significant difference compared with HC; When HC is not available (NA), * represents the significant difference compared with PD. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

† Number of participants with RBDSQ: HC (n = 25); PD (n = 120).

^# Number of participants with NCPV: All (n = 1028); HC (n = 114); PD (n = 439); Prodromal (n = 475).

ALPS index declines and NCPV increases with age, differs by sex, and changes across disease stages

The ALPS index was derived from diffusion magnetic resonance imaging (MRI) (Fig. 2a, b), and NCPV was quantified from 3D T1-weighted images in a subset of participants (Fig. 2c). Across all 1,861 participants, the ALPS index showed a significant negative association with age (Fig. 2d), after adjustment for sex and imaging sequences. This association was strongest in PD patients (Fig. 2e), and was also present in prodromal individuals (Fig. 2f) and HCs (Fig. 2g). Furthermore, a significant interaction effect was observed between age and disease group (F = 3.917, p = 0.02), indicating that the rate of age-related decline differed across groups, with the PD group exhibiting the steepest decline. After adjustment for age and imaging sequences, males exhibited significantly lower ALPS indices than females across the entire cohort (Fig. 2h), as well as within the PD (Fig. 2i) and prodromal (Fig. 2j) groups. However, in the HC group (Fig. 2k), males showed significantly higher ALPS indices than females. When controlling for age, sex, and imaging sequences, disease-related differences were evident, with the ALPS index showing a stepwise reduction from HCs to prodromal individuals and PD patients (Fig. 3a).

Fig. 2. ALPS index- and NCPV-based assessment of glymphatic function and their associations with age and sex.

a Schematic illustration of ROI placement for DTI-ALPS. Bilateral spherical ROIs (radius = 3 mm) were placed at the level of the lateral ventricles, targeting projection fibers and association fibers. b Calculation of the ALPS index using diffusivities along orthogonal axes, where the X-axis diffusivity of projection and association fibers ($D_{\mathit{xx},\mathit{proj}}$, $D_{\mathit{xx},\mathit{assoc}}$) and the Y- and Z-axis diffusivities of projection and association fibers ($D_{\mathit{yy},\mathit{proj}}$, $D_{\mathit{zz},\mathit{assoc}}$) were used to estimate glymphatic function. c CPV were automatically segmented from 3D T1-weighted images using an active-learning segmentation pipeline and subsequently normalized for analysis of NCPV. Red highlights indicate the segmented ChP. Scatter plots showing the relationship between age and the ALPS index across all participants (d), PD patients (e), prodromal individuals (f), and HCs (g). Comparisons of the ALPS index between sexes across the entire cohort (h) and within the PD (i), prodromal (j), and HC (k) groups. Scatter plots showing the association between age and NCPV across all participants (l), PD patients (m), prodromal individuals (n), and HCs (o). Sex differences in NCPV across the entire cohort (p) and within the PD (q), prodromal (r), and HC (s) groups. Values are means ± S.E.M. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Fig. 3. Correlation of the ALPS index with NCPV and regional cortical thickness, and differences across disease groups.

Group-wise comparison of the ALPS index (a) and NCPV (b) among PD, prodromal, and HC groups after controlling for age, sex, and imaging sequences. Scatter plots illustrating the associations between NCPV and the ALPS index across all participants (c), as well as within PD patients (d), prodromal individuals (e), and HCs (f). g Scatter plot showing a positive correlation between the ALPS index and cortical thickness of the right parahippocampal gyrus in PD patients, with lower ALPS index values associated with reduced cortical thickness. h Scatter plot illustrating the association between the left ALPS index and cortical thickness of the left entorhinal cortex in PD patients, with lower left ALPS index values correlated with reduced entorhinal thickness. i Relationship between the left ALPS index and cortical thickness of the left pars orbitalis, demonstrating a significant positive correlation in PD. Values are means ± S.E.M. ***, p < 0.001. Brain visualizations in g-i were generated using MRIcroGL (https://github.com/rordenlab/MRIcroGL).

Meanwhile, NCPV showed a significant positive association with age across the 1028 participants with available NCPV data (Fig. 2l), particularly within the PD group (Fig. 2m). This correlation was not observed in the prodromal (Fig. 2n) or HC (Fig. 2o) groups. After controlling for age and imaging sequences, males exhibited higher NCPV than females in the overall cohort (Fig. 2p) as well as within PD (Fig. 2q), prodromal (Fig. 2r), and HC (Fig. 2s) groups. When controlling for age, sex, and imaging sequences, disease-related differences were significant in NCPV, with a stepwise increase from HCs to prodromal individuals and PD patients (Fig. 3b).

Additionally, the ALPS index and NCPV were significantly negatively correlated in all participants (Fig. 3c), as well as within PD (Fig. 3d), prodromal (Fig. 3e), and HC (Fig. 3f) groups, suggesting complementary alterations in imaging markers related to brain clearance processes.

Reduced ALPS index is associated with regional cortical thinning in PD

In PD patients with available cortical thickness data, lower ALPS indices were associated with reduced cortical thickness in specific regions. Specifically, a lower mean ALPS index correlated with decreased thickness in the right parahippocampal cortex (Fig. 3g), while a lower left ALPS index was significantly associated with reduced thickness in the left entorhinal (Fig. 3h) and left pars orbitalis (Fig. 3i). These associations were regionally selective and persisted after adjustment for age, sex, and imaging sequences, suggesting a potential link between ALPS-derived measures and cortical regions implicated in PD-related cognitive decline.

ALPS index is negatively associated with motor severity and disease progression

Lower ALPS indices were associated with greater motor symptom severity in both prodromal and PD groups. In the prodromal cohort, lower ALPS indices were associated with higher Unified Parkinson’s Disease Rating Scale (UPDRS) I (Fig. 4a), UPDRS II (Fig. 4b), and total UPDRS scores (Fig. 4d), but not with UPDRS III (Fig.4c). In the PD group, similar but stronger correlations were observed, with lower ALPS indices associated with higher UPDRS I (Fig. 4e), UPDRS II (Fig. 4f), UPDRS III (Fig. 4g), and total scores (Fig. 4h), whereas no correlation was observed with UPDRS IV (Supplementary Table 1). Mediation analyses indicated that the ALPS index partially mediated the relationship between aging and UPDRS scores (Fig. 4i–k), suggesting that age-related decline in brain clearance may contribute to worsening motor symptoms. Consistently, ALPS index differed across Hoehn and Yahr (H-Y) stages (Fig. 4l), with the most prominent decline occurring during the transition from early to mid-stage disease. In contrast, no correlation was observed between NCPV and motor symptom severity assessed by UPDRS in PD or prodromal groups (Supplementary Table 2 and Fig. S1a–h), nor were differences observed across H-Y stages (Fig. S1i).

Fig. 4. ALPS index is negatively associated with motor symptom severity in PD and prodromal individuals.

Partial correlation analyses in the prodromal cohort showing associations between the ALPS index and UPDRS I (a), UPDRS II (b), UPDRS III (c), and total UPDRS scores (d). Partial correlation analyses in PD patients demonstrating stronger negative correlations between the ALPS index and UPDRS I (e), UPDRS II (f), UPDRS III (g), and total UPDRS scores (h). Mediation analyses testing whether the ALPS index mediates the relationship between age and clinical impairment in PD. The ALPS index partially mediated the effect of age on UPDRS II scores (i), UPDRS III scores (j), and total UPDRS scores (k). l) Comparison of the ALPS index across H-Y stages in PD patients. Values are means ± S.E.M. ns, not significant; *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Associations of ALPS index and NCPV with non-motor symptoms

We next examined associations between ALPS index and NCPV, and non-motor symptoms, including RBD, hyposmia, anxiety, cognitive impairment, and depression. Participants with RBD exhibited lower ALPS indices independent of disease stage (Fig. 5a, b). PD patients with probable RBD (PD-pRBD) had lower ALPS indices than prodromal individuals with RBD (Pro-RBD) (Fig. 5c), and PD patients without RBD (PD-Non) also had lower indices than prodromal individuals without RBD (Pro-Non) (Fig. 5d). NCPV was significantly higher in Pro-RBD individuals compared with Pro-Non individuals (Fig. 5f), whereas no significant difference was observed between PD-pRBD and PD-Non patients (Fig. 5e). PD-pRBD patients exhibited higher NCPV than Pro-RBD individuals (Fig. 5g), whereas no significant difference was observed between PD-Non and Pro-Non individuals (Fig. 5h). Receiver operating characteristic (ROC) analyses demonstrated the diagnostic potential of the ALPS index, NCPV and their combination in differentiating participant groups (Fig. 5i-s). The ALPS index achieved high area under the curve (AUC) value when distinguishing prodromal and HC groups.

Fig. 5. ALPS index is lower and NCPV is higher in PD and prodromal individuals with RBD compared with those without RBD.

Comparisons of the ALPS index between participants with and without RBD in PD (a) and prodromal groups (b). Comparisons of the ALPS index across disease progression with (c) or without (d) RBD. Comparison of NCPV between participants with and without RBD in PD (e) and prodromal group (f). NCPV comparisons across disease progression stages with (g) or without (h) RBD. ROC analyses evaluating the performance of the ALPS index, NCPV, and their combination in distinguishing PD (i), prodromal (j), and HC (k) groups. ROC analyses assessing the ability of the ALPS index, NCPV, and their combination to discriminate PD-pRBD with PD-Non (l), PD-pRBD with HC (m), and PD-Non with HC (n). ROC analyses assessing the ability of the ALPS index, NCPV, and their combination to discriminate Pro-pRBD with Pro-Non (o), Pro-pRBD with HC (p), and Pro-Non with HC (q). The ability of the ALPS index, NCPV, and their combination to discriminate the disease status within the RBD (r) and Non-RBD (s) cohort. Values are means ± S.E.M. ns, not significant; *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Olfactory function was assessed using the University of Pennsylvania Smell Identification Test (UPSIT)¹⁹. No significant associations were observed between UPSIT scores and either the ALPS index or NCPV in the PD and prodromal groups, and neither metric differed significantly across olfactory subgroups (Fig. S2a–h). Furthermore, no significant correlations were observed between the ALPS index or NCPV and UPSIT scores within the PD-pRBD and Pro-RBD groups, and comparisons across olfactory subgroups were also non-significant (Fig. S2i–p). Similarly, neither the ALPS index nor NCPV showed significant correlations with state anxiety (Fig. S3a–h) or trait anxiety (Fig. S3i–p) scores in the PD and prodromal groups.

Cognitive status was evaluated using the Montreal Cognitive Assessment (MoCA)²⁰. In PD patients, lower ALPS indices were associated with worse cognitive performance (Fig. 6a). PD patients with normal cognition had higher ALPS indices than those with possible mild cognitive impairment (PD-pMCI) or possible dementia (PD-pD), whereas the difference between PD-pMCI and PD-pD groups was not significant (Fig. 6b). MoCA scores were also positively correlated with left entorhinal cortical thickness. Mediation analysis indicated that left ALPS index partially mediated the relationship between left entorhinal thickness and MoCA scores (Fig. 6c). In contrast, NCPV was not significantly correlated with MoCA scores in PD patients (Fig. 6d), although PD-pD group exhibited significantly higher NCPV than the PD-pMCI group (Fig. 6e). In the prodromal individuals, neither the ALPS index nor NCPV showed significant correlations with MoCA scores (Fig. S4a–d).

Fig. 6. ALPS index and NCPV are associated with cognitive impairment and depressive symptom severity in PD.

a Partial correlation analyses controlling for age, sex, and imaging sequences showing positive associations between MoCA scores and the ALPS index in PD patients. b Comparison of the ALPS index among PD patients with normal cognition, PD-pMCI, and PD-pD. c Mediation analysis demonstrating that the left ALPS index partially mediates the relationship between left entorhinal cortical thickness and MoCA scores. d Partial correlation analysis showing no significant association between MoCA scores and NCPV. e Comparison of NCPV among PD patients with normal cognition, PD-pMCI, and PD-pD. f Partial correlation analysis revealing a significant negative association between the ALPS index and depression severity, assessed by the GDS-15, in PD patients. g Comparison of the ALPS index among PD-ND, PD-MD, and PD-MTSD groups, showing that the ALPS index progressively decreased with increasing depressive severity. h Mediation analysis showing that age-related reductions in the ALPS index fully mediate the relationship between age and depression severity in PD. i Partial correlation analysis revealing a significant positive association between NCPV and depression severity, assessed by the GDS-15, in PD patients. j Comparison of NCPV among PD-ND, PD-MD, and PD-MTSD groups, showing that NCPV progressively increased with worsening depressive symptoms. k Mediation analysis showing that age-related increases in NCPV fully mediate the relationship between age and depression severity in PD. Values are means ± S.E.M. ns, not significant; *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Depressive symptoms were assessed using the Geriatric Depression Scale-15 (GDS-15). The ALPS index was significantly negatively associated with depression severity in PD patients (Fig. 6f). ALPS indices decreased stepwise with increasing depression severity from PD patients having no depression (PD-ND) to mild depression (PD-MD) and reaching the lowest values in moderate-to-severe depression (PD-MTSD) (Fig. 6g). Mediation analysis further indicated that the ALPS index significantly mediated the relationship between age and depression severity. Specifically, the indirect effect of the ALPS index on depression severity was significant, whereas the direct effect of age became non-significant after controlling for the ALPS index (Fig. 6h). Bootstrap analysis (95% confidence intervals (CI) [0.0213, 0.0362]) confirmed a full mediation effect, indicating that the ALPS index accounted for the association between age and depressive symptom severity in PD. Meanwhile, in PD patients, NCPV was significantly positively correlated with depression severity (Fig. 6i) and increased stepwise from the PD-ND to PD-MD and then to PD-MTSD groups (Fig. 6j). Mediation analysis revealed that NCPV also fully mediated the relationship between age and depression severity (Fig. 6k). However, in the prodromal group, neither the ALPS index nor NCPV was significantly associated with depression severity (Fig. S4e–h).

Discussion

In this large, multi-stage cohort, we demonstrate that imaging markers associated with brain clearance-related processes, reflected by both the ALPS index and NCPV, show consistent associations with clinical features across the PD spectrum. Additionally, our findings suggest that the brain clearance pathway may represent a mechanistic link connecting aging, cortical vulnerability, and both motor and non-motor manifestations of PD.

The strong age dependence of both ALPS index and NCPV in PD suggests that aging-related changes in brain clearance-related processes may lower the threshold for pathogenic protein accumulation^21,22. Consistent sex differences were also observed, with males showing lower ALPS indices and higher NCPV, paralleling established epidemiologic patterns of PD risk and progression, although the biological mechanisms underlying these differences remain unclear^23,24. Potential contributors may include sex-specific differences in vascular function, astroglial physiology, or sleep architecture^25–28.

The stepwise decrease in ALPS index, together with a corresponding stepwise increase in NCPV, from HC to prodromal individuals and then to established PD patients suggests that changes in brain clearance-related processes emerge early and progressively worsen with disease progression. These findings are consistent with the hypothesis that impaired clearance-related processes may precede overt neurodegeneration rather than arise solely as a downstream consequence^14,29. Supporting this possibility, recent studies in isolated RBD (iRBD) have reported reduced DTI-ALPS indices and glymphatic alterations associated with clinical risk markers and structural brain changes linked to disease progression^30–32, suggesting that such alterations may already be present in prodromal stages of neurodegeneration. The observed associations between lower ALPS index and thinning of limbic cortical regions further support a potential link between brain clearance-related imaging markers and regional vulnerability, particularly in networks involved in cognition and mood.

Clinically, lower ALPS index was associated with greater motor severity and partially mediated the relationship between aging and motor impairment. These findings raise the possibility that alterations captured by ALPS-derived measures may represent one pathway through which aging contributes to worsening motor manifestations in PD^12,33. This supports a model in which reduced efficiency of clearance-related processes could exacerbate nigrostriatal and extranigral pathology by limiting the removal of toxic metabolites and misfolded proteins. The pronounced decline in ALPS index across H-Y stages further underscores its relevance to disease progression rather than symptom burden alone.

Among non-motor symptoms, the prominent associations between ALPS index and NCPV with RBD are notable, given the established role of sleep in glymphatic activity. These findings provide in vivo human evidence supporting a link between sleep-related glymphatic dysfunction and PD pathophysiology^{10,31,34–36}. In contrast, the lack of associations with olfaction and anxiety suggests that alterations in brain clearance-related processes may preferentially affect specific neural systems^37,38 rather than contributing uniformly to all non-motor features. A recent study in iRBD reported a positive association between the ALPS index and olfactory performance (UPSIT)³¹. However, in our cohort, neither ALPS index nor NCPV correlated significantly with UPSIT scores in the overall prodromal group or the prodromal-RBD subgroup. This discrepancy may reflect differences in cohort composition, as our prodromal group included non-manifesting carriers of PD-related genetic variants, individuals with hyposmia, and patients with RBD, whereas the prior study examined a relatively homogeneous iRBD population. Variations in imaging acquisition and analytical approaches may also contribute. Longitudinal studies are needed to clarify the relationships between ALPS index- or NCPV-related alterations and olfactory function across the synucleinopathy continuum.

Notably, depressive severity showed a unique relationship with both ALPS index and NCPV. The full mediation of these measures between aging and depressive symptoms raises the possibility that alterations in brain clearance-related processes may contribute to mood disturbances in PD. These findings expand the conceptual scope of brain clearance failure beyond protein aggregation, suggesting potential links with neuromodulatory and affective circuit dysfunction.

Several limitations in this study should be acknowledged. First, the cross-sectional design precludes causal inference, and longitudinal studies are needed to determine whether alterations in ALPS index or NCPV predict phenoconversion or disease progression in PD, and whether interventions targeting brain clearance-related processes can influence clinical outcomes²⁹. Second, both the ALPS index and NCPV should be interpreted as indirect MRI markers rather than direct measures of glymphatic function. Emerging evidence indicates that the ALPS index may reflect a composite of factors, including perivascular fluid movement, white matter microstructural integrity, and vascular-related tissue changes^39,40. In addition, technical factors—such as regions of interest (ROI) definition, fiber-crossing effects, scanner-related variability across imaging platforms, and limited validation against other measures of glymphatic activity—may influence the interpretation of ALPS measurements. Although imaging protocols were included as covariates in the statistical analyses, residual inter-scanner effects inherent to the multi-center PPMI dataset cannot be completely excluded. Third, RBD status in the PD cohort was derived from the REM Sleep Behaviour Disorder Screening Questionnaire (RBDSQ) available in the PPMI dataset, and the threshold used to define probable RBD was based on the distribution in HCs, as adopted in previous studies⁴¹. This methodological constraint may affect the precision of RBD classification, and the observed association between ALPS index and RBD therefore warrants further investigation in future studies using polysomnography-confirmed cohorts. Fourth, cognitive status was classified using MoCA scores. Although MoCA is widely used as a cognitive screening tool in PD cohorts, screening measures alone cannot establish a clinical diagnosis of dementia. Future work integrating longitudinal designs, harmonized multi-center imaging cohorts, multimodal glymphatic imaging approaches, and comprehensive neuropsychological and functional assessments will be important to refine the interpretation and clinical relevance of ALPS- and NCPV-related measures.

In summary, the ALPS index and NCPV, as indirect MRI markers associated with brain clearance-related processes, are linked to clinical severity across the PD spectrum. These findings support further investigation of multimodal imaging approaches to better understand the biological underpinnings of these markers and their potential utility as biomarkers of disease progression in PD.

Methods

Participants

This study analyzed diffusion tensor imaging data from the PPMI database (https://www.ppmi-info.org/access-data-specimens/download-data, RRID:SCR_006431) as of November 20, 2025^42,43. The PPMI study was approved by the institutional review boards of all participating centers, and written informed consent was obtained from all participants prior to study enrollment. The study was conducted in accordance with the Declaration of Helsinki. All participants underwent comprehensive clinical evaluations at baseline and throughout longitudinal follow-up.

Clinical assessments

Clinical characteristics were assessed using standardized instruments available in the PPMI dataset. Clinical assessments included the UPDRS, which assesses cognitive and neuropsychiatric symptoms (UPDRS I), activities of daily living (UPDRS II), motor function (UPDRS III), and treatment-related motor complications (UPDRS IV)⁴⁴.

Disease severity in PD patients was assessed according to the H-Y staging scale⁴⁵. Demographic details of participants stratified by H-Y staging are summarized in Supplementary Table 3.

RBD status in prodromal individuals was confirmed by polysomnography at enrollment, and participants were categorized as Pro-RBD or Pro-Non. In PD patients, RBD status was assessed using RBDSQ⁴⁶. PD patients with RBDSQ scores exceeding one standard deviation above the HC mean were classified as PD patients with probable RBD (PD-pRBD), whereas those with scores within or below one standard deviation of the HC mean were classified as PD-Non⁴¹. Demographic details of participants stratified by RBD status are summarized in Supplementary Table 4.

Cognitive status was evaluated using MoCA²⁰. PD patients were categorized as PD with possible dementia (PD-pD, MoCA ≤ 21), PD with possible mild cognitive impairment (PD-pMCI, MoCA = 22–25), or PD-Normal (MoCA ≥ 26)⁴⁷. Demographic details of participants stratified by cognitive status are summarized in Supplementary Table 5.

Depressive symptoms were assessed using GDS-15. PD patients were categorized as PD-ND (GDS-15 < 5), PD-MD (GDS-15 = 5–8), or PD-MTSD (GDS-15 > 8). Demographic details of participants stratified by depression are summarized in Supplementary Table 6.

Olfactory function was assessed using UPSIT¹⁹. PD patients were categorized as having anosmia (PD-Anos, UPSIT ≤ 18), severe hyposmia (PD-SH, UPSIT 19-24), moderate hyposmia (PD-ModH, UPSIT 25-29), mild hyposmia (PD-MH, UPSIT 30-33), or normal olfaction (PD-Norm, UPSIT ≥ 34). Demographic details of participants stratified by olfactory function are summarized in Supplementary Table 7.

Anxiety severity was assessed using the State-Trait Anxiety Inventory (STAI), including STAI-State (STAI-S) and STAI-Trait (STAI-T)⁴⁸. PD patients were categorized as having no state anxiety/no trait anxiety (PD-NSA/PD-NTA, STAI-S/STAI-T < 39), mild state anxiety/mild trait anxiety (PD-MSA/PD-MTA, STAI-S/STAI-T, 40-49), or moderate state anxiety/moderate trait anxiety (PD-MoSA/PD-MoTA, STAI-S/STAI-T, 50-59). Demographic details of participants stratified by anxiety symptoms are summarized in Supplementary Tables 8 and 9.

DTI acquisition

DTI scans for the 1,861 participants included in this study were acquired across multiple research centers using Siemens Prisma_fit, Philips Achieva dStream, and GE DISCOVERY_MR750 scanners. Detailed DTI acquisition parameters are available on the PPMI website (http://www.ppmi-info.org/study-design/research-documents-and-sops/). The following DTI acquisition protocols were used: (1) DTI_B0_PA, DTI_revB0_AP, DTI_B700_64dir_PA, DTI_B1000_64dir_PA, DTI_B2000_64dir_PA: slice thickness = 2 mm; number of slices = 72; field of view = 230 × 230 mm; repetition time = 3800 ms; echo time = 90–100 ms; flip angle = 90°; number of directions = 64 per shell; b-values = 0, 700, 1000, and 2000 s/mm²; (2) DTI_RL, DTI_LR: slice thickness = 2 mm; number of slices = 80; field of view = 256 × 256 mm; repetition time = 10,000 ms; echo time = 80 ms; flip angle = 90°; number of directions = 32; b-values = 0 and 1000 s/mm². Demographic details of participants stratified by DTI protocols are summarized in Supplementary Table 10. No significant differences in ALPS indices were detected between two protocols.

ALPS index calculation

DTI images were preprocessed and ALPS indices were calculated using DSI Studio (Hou build, July 15, 2025; http://dsi-studio.labsolver.org). Raw DICOM files were converted to NIFTI format using MRIcron (https://www.nitrc.org/projects/mricron) and imported into DSI Studio, where they were subsequently converted to SRC format. Image correction included compensation for phase distortions and eddy currents using TOPUP/EDDY, followed by motion correction to align all time points to the first b0 image, thereby correcting for involuntary head movements during acquisition. Images were subsequently reconstructed in MNI space using Q-Space Differential Reconstruction with a diffusion sampling length ratio of 1.25 and aligned to the Human ICBM152_adult template. This process generated color-coded fractional anisotropy maps and diffusion tensor maps along the X, Y, and Z axes.

Spherical ROIs (radius = 3 mm) were placed bilaterally at the level of the lateral ventricles (Fig. 2a). Projection fibers were represented by the superior corona radiata, while association fibers were represented by the superior longitudinal fasciculus. ROI center coordinates were as follows: left projection fibers (53, 48, 49), left association fibers (59, 48, 49), right projection fibers (26, 48, 49), and right association fibers (20, 48, 49). Minor manual adjustments were made if necessary to optimize placement. The ALPS indices for the left and right hemispheres were calculated using the following formula⁸:

$$\mathrm{ALPS}\mathrm{index}=\frac{D_{\mathit{xx},\mathit{proj}}+D_{\mathit{xx},\mathit{assoc}}}{D_{\mathit{yy},\mathit{proj}}+D_{\mathit{zz},\mathit{assoc}}}$$

where $D_{\mathit{xx},\mathit{proj}}$ is the X-axis diffusivity of projection fibers, $D_{\mathit{xx},\mathit{assoc}}$ is the X-axis diffusivity of association fibers, $D_{\mathit{yy},\mathit{proj}}$ is the Y-axis diffusivity of projection fibers, and $D_{\mathit{zz},\mathit{assoc}}$ is the Z-axis diffusivity of association fibers (Fig. 2b). The mean of the left and right ALPS indices (mean ALPS index) was used as the final measure. The ALPS index is an indirect MRI-based marker associated with perivascular fluid dynamics and has been proposed to reflect brain clearance-related processes in humans, with higher and lower values indicating relatively greater or reduced function.

CPV segmentation and normalization

CPV were obtained from 3D T1-weighted MRI using the ChP-Seg pipeline (https://github.com/princeleeee/ChP-Seg) via Docker. This pipeline employs a two-stage 3D U-Net to segment the lateral ventricles followed by the ChP, with preprocessing steps including reorientation, resampling, and skull stripping. A human-in-the-loop active-learning strategy⁴⁹ was applied, with manual corrections of challenging cases to improve segmentation accuracy. Segmented ChP masks were restored to the original image space, and absolute volumes were calculated from voxel counts (Fig. 2C)^49,50. Normalized CPV (NCPV) was calculated as the ratio of CPV to total intracranial volume (ratio×10³)^22,51. Detailed information regarding the acquisition parameters of T1-weighted, 3D volumetric sequence is available on the PPMI website (http://www.ppmi-info.org/study-design/research-documents-and-sops/). Key parameters were as follows: plane = sagittal; slice thickness = 1.0 mm; number of slices = 192; voxel size = 1.0 × 1.0 mm in plane resolution; phase encode direction = anterior-posterior; matrix = 256 × 256; field of view = 256 mm; and scan time ≈ 7 min.

Cortical thickness analysis

Cortical thickness data were obtained from 428 PD patients with appropriate 3D T1-weighted MRI data in the PPMI database (270 males, 158 females; mean age = 65.30 ± 9.29 years; age range = 36–86). Cortical thickness was calculated using FreeSurfer (version 7.2.3) with the automated recon-all processing pipeline applied to T1-weighted MRI scans⁵². This pipeline performs standardized preprocessing, cortical surface reconstruction, and tissue segmentation, with Nipoppy feature extraction tools (https://nipoppy.readthedocs.io/en/latest/user_guide/extraction.html) tracking all runtime parameters to ensure analytical consistency and reproducibility. Cortical regions were parcellated according to the Desikan-Killiany atlas, and thickness measurements for each cortical subregion were obtained for subsequent analyses. To ensure data quality, all cortical reconstructions were visually inspected. Scans with major segmentation errors, evident misclassification, or inclusion of non-brain tissue were excluded.

Statistical analysis

All statistical analyses were performed using IBM SPSS Statistics v26.0. The normality of continuous variables was assessed through visual inspection of histograms and Q-Q plots, along with skewness and kurtosis measures. Normally distributed variables were compared using one-way ANOVA followed by Bonferroni-corrected post hoc pairwise comparisons. Non-normally distributed variables were assessed using the Kruskal-Wallis test, and categorical variables were compared using the chi-square test. Partial correlation analyses were conducted to evaluate associations between the ALPS index or NCPV and clinical scales, and cortical thickness, with age, sex, and imaging sequences included as covariates. False discovery rate correction was applied to account for multiple comparisons. Analysis of covariance (ANCOVA) was used to perform cross-sectional group comparisons of ALPS indices or NCPV and clinical scale scores while controlling for age, sex, and imaging sequences. ROC curves were employed to assess the diagnostic performance of the ALPS index, NCPV, and their combination in distinguishing functional groups and assessing their potential utility as early biomarkers. Structural equation modeling was used to evaluate potential mediation effects. Mediation analyses were conducted using Model 4 of the SPSS PROCESS macro with bootstrap resampling (5,000 iterations) to assess indirect effects and their 95% CI. Mediation effects were considered significant if the 95% CI did not include zero.

Author contributions

Y.W., Y.Y., and J.Z. conceptualized this work. Y.W. developed the analytical methods, analyzed the data, and drafted the manuscript. Y.W. and Y.Y. prepared the figures. Y.W., Z.L., D.W., Y.Y., and J.Z. interpreted the data and edited the manuscript. All the authors approved the submitted version.

Data availability

This study analyzed DTI data from the PPMI database (https://www.ppmi-info.org/access-data-specimens/download-data, RRID:SCR_006431) as of November 20, 2025. Other data generated or analyzed during this study are included in this article and its supplementary information files.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41531-026-01432-6.