Identifying network state-based Parkinson's disease subtypes using clustering and support vector machine models.

Developing reliable biomarkers capable of differentiating Parkinson's disease from other neurological conditions is crucial for both patient care and research. In this study, we leveraged recent advances in high-throughput proteomic technology and machine learning to develop candidate biomarkers for Parkinson's disease. Using the Olink Explore 3072 assay, we obtained plasma proteomic profiles from 698 study participants, comprising Parkinson's disease cases (n = 149), neurologically healthy controls (n = 230), and participants with other neurological conditions (n = 319). The study cohort was split into Training Set (n = 560) and Test Set (n = 138). We conducted differential protein abundance analysis and pathway enrichment analysis, and subsequently applied the Boruta algorithm to identify differentially abundant proteins that are predictive of Parkinson's disease. To create a diagnostic biomarker panel, we trained a stacking ensemble machine learning (ML) model on the Training Set (n = 118 Parkinson's patients, n = 184 healthy controls, and n = 258 individuals with other neurological disorders) using eleven proteins (APOH, ARG1, CCN1, CXCL1, CXCL8, DDC, GRAP2, IL1RAP, OSM, PRL, and SPRY2) as model features. We used the Shapley Additive Explanations (SHAP) framework and network analysis to evaluate predictive importance and biological relevance of each protein in the ML model. The model demonstrated high accuracy in the held-out Test Set (n = 138) and three external cohorts-the UK Biobank (n = 43,969), the Parkinson's Disease Biomarkers Program (n = 138), and the Parkinson's Progression Markers Initiative (n = 385), with areas under the receiver operating characteristic curve of 0.939, 0.789, 0.909, 0.816, respectively. Additionally, network and pathway analyses helped interpret the model, revealing activity related to inflammatory mediators, ErbB signaling, T-cell receptor signaling, and lipid metabolism. Our findings highlight the potential of plasma protein biomarkers to improve Parkinson's disease diagnosis and deepen biological understanding of this complex neurological disorder. Our model demonstrates high specificity and reliability across multiple independent cohorts, indicating the significant potential of proteomics-based biomarkers and the clinical utility of ML-supported diagnosis in Parkinson's disease care. The model also helps to elucidate potential novel risk factors and pathways associated with Parkinson's disease.

The Parkinson's Progression Marker Initiative is an international multicenter study whose main goal is investigating markers for Parkinson's disease (PD) progression as part of a path to a treatment for the disease. This manuscript describes the baseline genetic architecture of this study, providing not only a catalog of disease-linked variants and mutations, but also quantitative measures with which to adjust for population structure. Three hundred eighty-three newly diagnosed typical PD cases, 65 atypical PD and 178 healthy controls, from the Parkinson's Progression Marker Initiative study have been genotyped on the NeuroX or Immunochip arrays. These data are freely available to all researchers interested in pursuing PD research within the Parkinson's Progression Marker Initiative. The Parkinson's Progression Marker Initiative represents a study population with low genetic heterogeneity. We recapitulate known PD associations from large-scale genome-wide association studies and refine genetic risk score models for PD predictability (area under the curve, ∼0.74). We show the presence of six LRRK2 p.G2019S and nine GBA p.N370S mutation carriers. The Parkinson's Progression Marker Initiative study and its genetic data are useful in studies of PD biomarkers. The genetic architecture described here will be useful in the analysis of myriad biological and clinical traits within this study.

Clinical and dopamine transporter imaging characteristics of non-manifest LRRK2 and GBA mutation carriers in the Parkinson's Progression Markers Initiative (PPMI): a cross-sectional study

Parkinson's disease (PD) heterogeneity complicates disease-modification clinical trial design and lead to ambiguous results, necessitating robust subtyping frameworks to identify rapid progressors. This study compared progression milestones across previously defined clinical (tremor-dominant [TD], postural instability/gait difficulty [PIGD], indeterminate), pathological (brain-first, body-first), and data-driven (diffuse malignant [DM], intermediate [IM], mild motor-predominant [MMP]) PD subtypes. Using data from the Parkinson's Progression Markers Initiative (PPMI), we analyzed milestone attainment across six functional domains and performed competing risks regression. Randomized controlled trial sample size simulations evaluated the impact of subtyping on trial efficiency. Data-driven subtypes exhibited the highest progression rates, with DM patients attaining 50% of milestones, surpassing PIGD (43.0%) and body-first (42.0%) subtypes. Competing risks regression confirmed that DM patients had more than double the hazard of progression compared to MMP (SHR 2.02, 95% CI 1.49 to 2.75; P < 0.001). Trial power simulations demonstrated that enrolling DM patients could reduce sample size requirements by approximately 50% using standard trial durations compared to unstratified PD cohorts. This analysis showed that data-driven subtyping, particularly the identification of the DM subtype, offers a promising strategy to optimize disease-modification trials in PD by capturing patients who meet progression milestones earlier.

The Parkinson's Progression Marker Initiative (PPMI), sponsored by The Michael J. Fox Foundation for Parkinson's Research, is designed to develop disease biomarkers to accelerate Parkinson's disease (PD) therapeutics. During the past 15 years PPMI has acquired comprehensive clinical, imaging, biomarker and genetic data on thousands of participants that has guided clinical therapeutic studies for PD and related synucleinopathies. The study has established standardized strategies for acquisition and analysis of data and set a standard for data sharing and open-source publication. PPMI has by design consistently evolved to incorporate new knowledge and science. A key example is that PPMI data has enabled the transition from clinical to biologic definition of synucleinopathy. The study continues to expand our understanding of disease biology particularly in precise PD subsets in both in-clinic and a rapidly expanding remote participant cohort, using myPPMI, an online portal. This special issue reviews PPMI's continuing contribution to early detection, understanding disease biologic and staging, and innovative therapeutic strategies in PD and related synucleinopathies. ANN NEUROL 2026.

This study aimed to identify CSF proteomic signatures characteristic of Parkinson disease (PD) and evaluate their clinical utility. This observational study used data from the Parkinson's Progression Markers Initiative (PPMI), which enrolled patients with PD, healthy controls (HCs), and non-PD participants carrying GBA1, LRRK2, and/or SNCA pathogenic variants (genetic prodromals) at international sites. Study participants were chosen from PPMI enrollees based on the availability of aptamer-based CSF proteomic data, quantifying 4,071 proteins, and classified as patients with PD without GBA1, LRRK2, and/or SNCA pathogenic variants (nongenetic PD), HCs, patients with PD carrying the aforementioned pathogenic variants (genetic PD), or genetic prodromals. Differentially expressed protein (DEP) analysis and the least absolute shrinkage and selection operator (LASSO) were applied to the data from nongenetic PD and HCs. Signatures characteristics of nongenetic PD were quantified as a PD proteomic score (PD-ProS), validated internally and then externally using data of 1,556 CSF proteins from the LRRK2 Cohort Consortium (LCC). We further tested the PD-ProS in genetic PD and genetic prodromals and examined associations with clinical progression. Data from 279 patients with nongenetic PD (mean ± SD, age 62.0 ± 9.6 years; male 67.7%) and 141 HCs (age 60.5 ± 11.9 years; male 64.5%) were used for PD-ProS derivation. From 23 DEPs, LASSO determined weights of 14 DEPs for the PD-ProS (area under the curve [AUC] 0.83, 95% CI 0.78-0.87), validated in an independent internal validation cohort of 71 patients with nongenetic PD and 35 HCs (AUC 0.81, 95% CI 0.73-0.90). In the LCC, only 5 of the 14 DEPs were also measured. Notably, these 5 DEPs still distinguished 34 patients with nongenetic PD from 31 HCs with the same weights (AUC 0.75, 95% CI 0.63-0.87). Furthermore, the PD-ProS distinguished 258 patients with genetic PD from 365 genetic prodromals. Finally, regardless of genetic status, the PD-ProS independently predicted both cognitive and motor decline in PD (dementia, adjusted hazard ratio in the highest quintile [aHR-Q5] 2.8 [95% CI 1.6-5.0]; Hoehn and Yahr stage IV, aHR-Q5 2.1 [95% CI 1.1-4.0]). By integrating high-throughput proteomics with machine learning, we identified PD-associated CSF proteomic signatures crucial for PD development and progression. ClinicalTrials.gov (NCT01176565). A link to the trial registry page is clinicaltrials.gov/ct2/show/NCT01141023. This study provides Class II evidence that the CSF proteome contains clinically important information regarding the development and progression of Parkinson disease that can be deciphered by a combination of high-throughput proteomics and machine learning.

Diagnosis and forecasting disease progression is critical for effective treatment of Parkinson's disease (PD). The motivation of this study is to use machine learning methods such as neural networks, logistic regression, and random forests to diagnose PD and to determine whether the severity of the disease increases based on clinical information from early onset. We used data-driven models including traditional machine learning models and deep neural networks to determine and predict the PD condition. The proposed methods were validated with the Parkinson's Progression Markers Initiative (PPMI) dataset, which is the most widely known and validated source of PD data. The Hoehn & Yahr (H&Y) scale was used to determine if there was a change in disease severity. The results show that the accuracy of applying neural network to diagnose PD is 94.26%, and the accuracy of applying the random forest model to predict changes in disease severity achieved 82.24%.

Development of robust plasma-based biomarkers in Parkinson's disease (PD) could lead to new approaches for identifying those at risk for PD and developing novel therapies. Here, we validate plasma apolipoprotein A1 (ApoA1) as a correlate of age at onset and motor severity in PD. Plasma ApoA1 and high-density lipoprotein at baseline, 6 months, and 12 months were measured in 254 research volunteers (154 patients with PD and 100 normal controls) enrolled in the Parkinson's Progression Markers Initiative (PPMI) study. Lower baseline plasma ApoA1 levels associate with an earlier age at PD onset in early-stage, drug-naïve PPMI PD patients (P = 0.023). Moreover, lower baseline ApoA1 levels trend toward association with worse motor severity in PPMI PD patients (p = 0.080). Over 12 months of follow-up, plasma ApoA1 levels do not predict motor decline in the PPMI PD cohort. Finally, a meta-analysis of five PD cohorts encompassing >1,000 patients confirms significant association of lower plasma ApoA1 with earlier age at PD onset (P < 0.001) and greater motor severity (P < 0.001). Our results confirm the previously reported association of lower plasma ApoA1 levels with two clinical features suggesting poorer dopaminergic system integrity-earlier age at PD onset and greater motor severity-in early-stage, drug-naïve PD patients. This is the first report of a plasma-based biomarker evaluated in the PPMI study. Future investigations are warranted evaluating plasma ApoA1 as a longitudinal correlate of disease progression as well as investigating the potential of ApoA1 as a therapeutic target in PD.

BackgroundCurrently, prognosis of Parkinson's Disease (PD) is limited. Emerging literature highlights potential of multi-modal biomarkers and neuroimaging to provide critical insight into clinical progression, potentially improving prediction of long-term outcomes.MethodsData were extracted from the Parkinson's Progression Markers Initiative (PPMI). Hierarchical clustering was applied to Movement Disorder Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS) scores at year-five follow-up, identifying two clusters. Differences in progression, as well as retrospective assessment of baseline differences, between clusters were explored for pathological biomarkers, neuroimaging, and prodromal measures. Additionally, logistic regression, receiver operating characteristic curve analyses and machine learning were employed to determine utility of variables at baseline as predictors of cluster membership.ResultsThe more impaired cluster demonstrated worse motor and non-motor outcomes, including higher rates of dementia and cognitive complaints at year-five, as well as more profound rigidity than cluster one. Further, retrospective comparisons showed cluster two performing worse in all prodromal measures and demonstrated lower striatal dopamine transporter and cognitive ability. Logistic regression determined that membership in this cluster was predicted by higher autonomic dysfunction and p-tau, along with reduced smell and alpha-syn, predicting 49.1% of variance (AUC = 0.92). This was significantly higher (p < 0.001) than the model including MDS-UPDRS scores alone, only accounting for 27.4% of variance (AUC = 0.74). Findings were corroborated by machine learning, whereby multi-modal assessment corresponded to 74% classification accuracy, compared to 60% with MDS-UPDRS alone.ConclusionPrediction of more marked impairment at year-five was substantially improved via multi-modal assessment, specifically, pathological biomarkers, suggesting that incorporating biomarkers into clinical criteria could enhance long-term prognosis.

Pathology in the Parkinson's Progression Markers Initiative; a Finale but also a start

A history of traumatic brain injury (TBI) is associated with an increased risk of developing neurodegenerative disorders, including Parkinson's Disease (PD). However, TBI's influences on disease progression remain underassessed. This study explored whether a history of TBI influences the progression of pathological and clinical outcomes up to 5 years of follow-up in individuals with early PD. Longitudinal data were extracted from the Parkinson's Progression Markers Initiative (PPMI) and the PostCEPT observational study. Participants in PostCEPT had complete head injury data, while PPMI participants were eligible if they completed the head injury section of the PD Risk Factor Questionnaire (n = 208). Principal component analysis was used to derive composite scores of cognitive ability and mood dysfunction, with motor outcomes calculated using the Movement Disorders Society Unified Parkinson's Disease Rating Scale. Progression of clinical and pathological outcomes up to 5 years and 4 years following study entry were compared, including subset analyses in PPMI examining injury severity. Individuals with a history of TBI in the PPMI dataset exhibited a younger age of onset; however, a history of TBI did not affect progression rates of any assessed variables across both cohorts. Exploratory analysis determined that injury severity significantly predicted striatal dopamine transporter binding but accounted for only a small portion of outcome variance. While the history of TBI was associated with earlier PD onset, it did not correspond to a differential disease course. However, given differences in TBI characterisation between cohorts, additional research must be conducted to validate these findings.

Objective:To analyze longitudinal levels of CSF biomarkers in drug-naive patients with Parkinson disease (PD) and healthy controls (HC), examine the extent to which these biomarker changes relate to clinical measures of PD, and identify what may influence them.Methods:CSF α-synuclein (α-syn), total and phosphorylated tau (t- and p-tau), and β-amyloid 1–42 (Aβ42) were measured at baseline and 6 and 12 months in 173 patients with PD and 112 matched HC in the international multicenter Parkinson's Progression Marker Initiative. Baseline clinical and demographic variables, PD medications, neuroimaging, and genetic variables were evaluated as potential predictors of CSF biomarker changes.Results:CSF biomarkers were stable over 6 and 12 months, and there was a small but significant increase in CSF Aβ42 in both patients with patients with PD and HC from baseline to 12 months. The t-tau remained stable. The p-tau increased marginally more in patients with PD than in HC. α-syn remained relatively stable in patients with PD and HC. Ratios of p-tau/t-tau increased, while t-tau/Aβ42 decreased over 12 months in patients with PD. CSF biomarker changes did not correlate with changes in Movement Disorder Society–sponsored revision of the Unified Parkinson’s Disease Rating Scale motor scores or dopamine imaging. CSF α-syn levels at 12 months were lower in patients with PD treated with dopamine replacement therapy, especially dopamine agonists.Conclusions:These core CSF biomarkers remained stable over 6 and 12 months in patients with early PD and HC. PD medication use may influence CSF α-syn. Novel biomarkers are needed to better profile progressive neurodegeneration in PD.

The Parkinson's Progression Markers Initiative (PPMI) has remained at the vanguard of Parkinson's disease (PD) research through evolving recruitment strategies that mirror advances in PD biology. PPMI was initially focused on untreated, early-stage PD. Over time, it has expanded to include genetic cohorts, prodromal participants identified via olfactory testing and risk factor questionnaires targeting people with rapid eye movement (REM) sleep behavior disorder (RBD). Most recently, PPMI has focused on the biology underlying PD and sought to recruit individuals defined by neuronal alpha-synuclein disease (NSD) biomarkers. This evolution from clinical to biological enrollment demonstrates the feasibility of biomarker-driven cohort design and establishes a scalable model for early detection and prevention research in neurodegenerative disease. ANN NEUROL 2026.

Nonmotor symptoms (NMS) in Parkinson's disease (PD) can start up to a decade before motor manifestations and strongly correlate with the quality of life. Understanding patterns of NMS can provide clues to the incipient site of PD pathology. Our goal was to systematically characterize the progression of NMS in PD (n = 489), compared to healthy controls, HC (n = 241), based on the sex of the subjects and laterality of motor symptom onset. Additionally, NMS experienced at the onset of PD were also compared to subjects with scans without dopaminergic deficit, SWEDD (n = 81). The Parkinson's Progression Markers Initiative (PPMI) database was utilized to analyze several NMS scales. NMS experienced by PD and SWEDD cohorts were significantly higher than HC and both sex and laterality influenced several NMS scales at the onset of motor symptoms. Sex Differences. PD males experienced significant worsening of sexual, urinary, sleep, and cognitive functions compared to PD females. PD females reported significantly increased thermoregulatory dysfunction and anxious mood over 7 years and significantly more constipation during the first 4 years after PD onset. Laterality Differences. At onset, PD subjects with right-sided motor predominance reported significantly higher autonomic dysfunction. Subjects with left-sided motor predominance experienced significantly more anxious mood at onset which continued as Parkinson's progressed. In conclusion, males experienced increased NMS burden in Parkinson's disease. Laterality of motor symptoms did not significantly influence NMS progression, except anxious mood. We analyzed NMS in a large cohort of PD patients, and these data are valuable to improve PD patients' quality of life by therapeutically alleviating nonmotor symptoms.

To evaluate the course and predictors of neuropsychiatric symptoms (NPS) and cognition in patients with de novo Parkinson disease (PD). Cross-sectional study of the cohort of de novo, untreated (at enrollment) patients with PD and healthy controls (HCs) from the Parkinson's Progression Markers Initiative. Participants have serial assessments of global cognition and symptoms of depression, anxiety, psychosis, impulse control disorders (ICDs), sleep and wakefulness, apathy, and fatigue. Available data up to 24 months of follow-up were included. The available sample size was as follows: baseline (PD = 423, HCs = 196), 12 months (PD = 261, HCs = 145), and 24 months (PD = 96, HCs = 83). Patients with PD experienced more depression, fatigue, apathy, and anxiety than HCs at all time points, and apathy (p = 0.001) and psychosis (p = 0.003) increased over time in patients with PD. Approximately two-thirds of patients with PD who screened positive for depression at any given visit were not taking an antidepressant. The Montreal Cognitive Assessment score decreased significantly over time in patients with PD (p < 0.001), but the change was comparable to that in HCs. At the 24-month visit, 44% of patients had been on dopamine replacement therapy (DRT) for at least 1 year, and this group reported more incident ICDs (p = 0.009) and excessive daytime sleepiness (p = 0.03). Multiple NPS are more common in de novo, untreated patients with PD compared with the general population, but they also remain relatively stable in early disease, while global cognition slightly deteriorates. In contrast, initiation of DRT is associated with increasing frequency of several other NPS.