Personalized progression modelling and prediction in Parkinson’s disease with a novel multi-modal graph approach

Lian, J., Luo, X., et al. (2024).

Npj Parkinson S Disease, 10(1).

doi.org/10.1038/s41531-024-00832-w

Abstract

Parkinson’s disease (PD) is a complex neurological disorder characterized by dopaminergic neuron degeneration, leading to diverse motor and non-motor impairments. This variability complicates accurate progression modelling and early-stage prediction. Traditional classification methods based on clinical symptoms are often limited by disease heterogeneity. This study introduces an graph-based interpretable personalized progression method, utilizing data from the Parkinson’s Progression Markers Initiative (PPMI) and Stroke Parkinson’s Disease Biomarker Program (PDBP). Our approach integrates multimodal inter-individual and intra-individual data, including clinical assessments, MRI, and genetic information to make multi-dimension predictions. Validated using the PDBP dataset from 12 to 36 months, our AdaMedGraph method demonstrated strong performance, achieving AUC values of 0.748 and 0.714 for the 12-month Hoehn and Yahr Scale and Movement Disorder Society-Sponsored Revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) III on the PPMI test set. Ablation analysis reveals the importance of baseline clinical assessment predictors. This novel framework improves personalized care and offers insights into unique disease trajectories in PD patients.

The research is here.

Here are some thoughts:

The research introduces AdaMedGraph, an innovative tool designed to model and predict the progression of Parkinson’s Disease (PD) using personalized, multimodal graph-based methods. This approach integrates clinical assessments, MRI imaging, and genetic data to create individualized disease trajectories. AdaMedGraph demonstrates superior predictive performance compared to traditional machine learning methods, achieving high accuracy for progression markers like the Hoehn and Yahr Scale (e.g., AUC 0.748) and effectively addressing the challenges of disease heterogeneity. Its ability to incorporate diverse data sources allows for detailed predictions across motor (e.g., rigidity, tremors) and non-motor symptoms (e.g., cognition, sleep patterns), which are critical for comprehensive PD management.

For psychologists, this research has significant implications. The integration of cognitive and behavioral assessments into the predictive framework underscores the importance of psychological evaluations in understanding PD progression. The model’s ability to identify personalized trajectories enables psychologists to tailor interventions for both motor and non-motor symptoms, enhancing patient-centered care. Furthermore, the study’s exploration of medication effects on symptoms provides valuable insights into the interaction between treatments and behavioral outcomes, informing therapeutic adjustments. By identifying critical predictive features and offering interpretable patient similarity analyses, AdaMedGraph aligns well with the emphasis on individualized care in psychological practice. This tool has the potential to advance multidisciplinary collaboration, enabling psychologists to combine behavioral insights with advanced predictive analytics to improve outcomes for individuals with Parkinson’s Disease.

Prediction of Future Parkinson Disease Using Plasma Proteins Combined With Clinical-Demographic Measures

You, J., et al. (2024).

Neurology, 103(3).

https://doi.org/10.1212/wnl.0000000000209531

Abstract

Background and Objectives

Identification of individuals at high risk of developing Parkinson disease (PD) several years before diagnosis is crucial for developing treatments to prevent or delay neurodegeneration. This study aimed to develop predictive models for PD risk that combine plasma proteins and easily accessible clinical-demographic variables.

Results

A total of 52,503 participants without PD (median age 58, 54% female) were included. Over a median follow-up duration of 14.0 years, 751 individuals were diagnosed with PD (median age 65, 37% female). Using a forward selection approach, we selected a panel of 22 plasma proteins for optimal prediction. Using an ensemble tree-based Light Gradient Boosting Machine (LightGBM) algorithm, the model achieved an area under the receiver operating characteristic curve (AUC) of 0.800 (95% CI 0.785–0.815). The LightGBM prediction model integrating both plasma proteins and clinical-demographic variables demonstrated enhanced predictive accuracy, with an AUC of 0.832 (95% CI 0.815–0.849). Key predictors identified included age, years of education, history of traumatic brain injury, and serum creatinine. The incorporation of 11 plasma proteins (neurofilament light, integrin subunit alpha V, hematopoietic PGD synthase, histamine N-methyltransferase, tubulin polymerization promoting protein family member 3, ectodysplasin A2 receptor, Latexin, interleukin-13 receptor subunit alpha-1, BAG family molecular chaperone regulator 3, tryptophanyl-TRNA synthetase, and secretogranin-2) augmented the model's predictive accuracy. External validation in the PPMI cohort confirmed the model's reliability, producing an AUC of 0.810 (95% CI 0.740–0.873). Notably, alterations in these predictors were detectable several years before the diagnosis of PD.

Discussion

Our findings support the potential utility of a machine learning-based model integrating clinical-demographic variables with plasma proteins to identify individuals at high risk for PD within the general population. Although these predictors have been validated by PPMI, additional validation in a more diverse population reflective of the general community is essential.

The article is cited above, but paywalled.

Here are some thoughts:

A recent study published in Neurology demonstrates the potential for early detection of Parkinson's disease (PD) using machine learning techniques. Researchers developed a predictive model that analyzes blood proteins in conjunction with clinical data, allowing for the identification of individuals at high risk of developing PD up to 15 years before symptoms appear. The study involved over 50,000 participants from the UK Biobank, focusing on 1,463 different blood plasma proteins. By employing machine learning to identify patterns in protein levels alongside clinical information—such as age, history of brain injuries, and blood creatinine levels—the researchers were able to achieve significant accuracy in predicting Parkinson's risk.

The findings revealed 22 specific proteins that are significantly associated with the risk of developing PD, including neurofilament light (NfL), which is linked to brain cell damage, as well as various proteins involved in inflammation and muscle function. This model not only offers a non-invasive and cost-effective screening method but also presents opportunities for early intervention and improved disease management, potentially enabling the development and assessment of neuroprotective treatments.

However, the study does have limitations that warrant consideration. The participant population was predominantly of European descent, which may limit the generalizability of the findings to more diverse groups. Additionally, the reliance on medical records for PD diagnosis raises concerns about potential misdiagnoses. Future research will need to validate the model in diverse populations and utilize more precise measurement techniques for protein levels. Longitudinal studies that incorporate repeated measurements could further enhance the predictive power of the model.

Overall, this groundbreaking research offers new hope for the early detection and intervention of Parkinson's disease, potentially revolutionizing the approach to managing this neurodegenerative disorder.