summary: Using machine learning, researchers have identified three subtypes of Parkinson’s disease based on the rate of progression. These subtypes, characterized by distinct genetic factors, may enhance diagnostic and treatment strategies.
The study also found that the diabetes drug metformin may improve symptoms of Parkinson’s disease, especially in more advanced cases, a finding that paves the way for a personalized treatment approach for Parkinson’s patients.
Key Facts:
- There are three subtypes: Subtypes of Parkinson’s disease defined by the rate of progression: gradual, moderate, or rapid.
- Distinct drivers: Each subtype has unique genetic and molecular markers.
- Potential Treatments: Metformin is expected to be particularly effective in improving symptoms of the rapid pace subtype.
sauce: Weill Cornell University
Researchers at Weill Cornell Medicine used machine learning to define three subtypes of Parkinson’s disease based on how quickly the disease progresses.
These subtypes are characterized by distinct driver genes, which, in addition to potentially being important diagnostic and prognostic tools, could also suggest how subtypes can be targeted with new and existing drugs if these markers are validated.
The study was published July 10. npj digital medicine.
“Parkinson’s disease is highly heterogeneous, and people with the same disease can experience very different symptoms,” said lead author Fei Wang, PhD, professor of population health science in the Department of Population Health Sciences at Weill Cornell Medicine and founding director of the AI Institute for Digital Health (AIDH).
“This indicates that there is probably no one-size-fits-all approach to treating this disease. We may need to consider customized treatment strategies based on a patient’s disease subtype.”
The researchers defined subtypes based on distinct patterns of disease progression: They named those with mild baseline disease severity and a slow rate of progression the “slowly progressive” subtype (PD-I, approximately 36% of patients), those with mild baseline disease severity but a moderate rate of progression the “moderate rate of progression” subtype (PD-M, approximately 51% of patients), and those with the fastest rate of symptom progression the “rapid rate of progression” subtype (PD-R).
The researchers were able to identify the subtypes by using a deep learning-based approach to analyze anonymized clinical records from two large databases, and they also investigated the molecular mechanisms associated with each subtype by analyzing patients’ genetic and transcriptomic profiles using network-based methods.
For example, PD-R subtypes activated specific pathways related to neuroinflammation, oxidative stress, and metabolism. The team also found distinct brain imaging and cerebrospinal fluid biomarkers for each of the three subtypes.
Dr. Wang’s lab has been studying Parkinson’s disease since 2016, when they participated in the Parkinson’s Progression Markers Initiative (PPMI) Data Challenge, sponsored by the Michael J. Fox Foundation. The team won the challenge focused on subtype derivation and has since received funding from the Foundation to continue this research.
They employed data collected from the PPMI cohort as the study’s main subtype development cohort and validated it in the National Institute of Neurological Disorders and Stroke (NINDS) Parkinson’s Disease Biomarker Program (PDBP) cohort.
The researchers used their findings to identify potential drug candidates that could be repurposed to target specific molecular changes found in the various subtypes. They then used two large, real-world databases of patient health records to confirm that these drugs could help mitigate the progression of Parkinson’s disease. These databases, INSIGHT
The New York-based Clinical Research Network and the OneFlorida+ Clinical Research Consortium are both part of the National Patient-Centered Clinical Research Network (PCORnet). INSIGHT is led by Rainu Kaushal, PhD, senior vice dean for clinical research at Weill Cornell Medicine and chair of the Department of Population Health Sciences at Weill Cornell Medicine and NewYork-Presbyterian/Weill Cornell Medical Center.
“By examining these databases, we found that people taking the diabetes drug metformin appear to have improved diabetes symptoms compared to people who are not taking the drug, particularly in terms of cognitive function and symptoms related to falls,” said lead author Chang Su, PhD, assistant professor of population health sciences at Weill Cornell Medical College and member of the AIDH.
This was especially true for patients with the PD-R subtype, who are most likely to present with cognitive impairment in the early stages of Parkinson’s disease.
“We hope that our study will inspire other researchers to consider using diverse data sources when conducting studies like ours,” Dr Wang said.
“We also believe that translational bioinformatics researchers can further validate our findings both computationally and experimentally.”
Many collaborators contributed to this research, including scientists from the Cleveland Clinic, Temple University, University of Florida, University of California, Irvine and University of Texas at Arlington, as well as doctoral students from Cornell Tech’s Computer Science Program and Cornell University’s Ithaca Campus Computational Biology Program.
News about AI and Parkinson’s disease research
author: Barbara Prempe
sauce: Weill Cornell University
contact: Barbara Prempe – Weill Cornell University
image: Image courtesy of Neuroscience News
Original Research: Open access.
“Identifying Parkinson’s disease PACE subtypes and repurposing treatments through integrated analysis of multimodal data” By Fei Wang et al. npj Digital Medicine
Abstract
Identifying Parkinson’s disease PACE subtypes and repurposing treatments through integrated analysis of multimodal data
Parkinson’s disease (PD) is a severe neurodegenerative disorder characterized by significant clinical and progression heterogeneity. This study aims to address the heterogeneity of PD through an integrated analysis of different data modalities.
We used machine learning and deep learning to analyze clinical progression data (over 5 years) of new PD patients and characterize their phenotypic progression trajectories across PD subtype classifications.
We found three pace subtypes of PD that show different progression patterns: the Inching Pace subtype (PD-I), which has mild baseline severity and progression rate, the Moderate Pace subtype (PD-M), which has mild baseline severity but a moderate progression rate, and the Rapid Pace subtype (PD-R), which has the fastest rate of symptom progression.
The P-tau/α-synuclein ratio in cerebrospinal fluid and atrophy of specific brain regions were found to be potential markers of these subtypes. Analysis of gene and transcriptomic profiles by a network-based approach identified molecular modules associated with each subtype.
For example, the PD-R specific module is STAT3, Fina, BECN1, Apoa 1, NEDD4and Gath 2 As potential driver genes of PD-R, neuroinflammation, oxidative stress, metabolism, PI3K/AKT, and angiogenesis pathways were suggested as potential drivers of rapid PD progression (i.e., PD-R).
Furthermore, we identified reusable drug candidates by targeting these subtype-specific molecular modules using a network-based approach and drug-gene signature data from cell lines. Furthermore, we estimated the therapeutic effects using two large-scale real-patient databases. The obtained real-world evidence highlighted the potential of metformin to alleviate PD progression.
In conclusion, this study will help to better understand the clinical and pathophysiological complexities of PD progression and accelerate precision medicine.
