Researchers at Weill Cornell Medicine have made a significant breakthrough by using machine learning to identify three distinct subtypes of Parkinson’s disease. These subtypes are categorized based on how quickly the disease progresses and are linked to specific genes.
This discovery, published on July 9 in npj Digital Medicine, could become a crucial tool for diagnosing and predicting the disease, and might also lead to new treatments.
Dr. Fei Wang, a professor at Weill Cornell Medicine, explains that Parkinson’s disease varies widely among patients, meaning a one-size-fits-all treatment approach is unlikely to be effective. Customized treatments based on the newly defined subtypes might be necessary.
The researchers named the subtypes based on the progression speed of the disease:
- Inching Pace (PD-I): This subtype, affecting about 36% of patients, is characterized by mild symptoms and slow progression.
- Moderate Pace (PD-M): This is the most common subtype, affecting about 51% of patients, with moderate symptom progression.
- Rapid Pace (PD-R): This subtype involves the fastest symptom progression and affects a smaller group of patients.
The team identified these subtypes by analyzing anonymous clinical records from two large databases using advanced deep learning techniques. They also examined the genetic and molecular profiles of patients to understand the underlying mechanisms of each subtype.
For instance, the PD-R subtype showed activation in pathways related to neuroinflammation, oxidative stress, and metabolism. Distinct biomarkers in brain imaging and cerebrospinal fluid were also found for each subtype.
Dr. Wang’s lab has been researching Parkinson’s since 2016, starting with their participation in the Parkinson’s Progression Markers Initiative (PPMI) data challenge. Their work, which won the challenge, has continued to build on these initial findings.
They used data from the PPMI cohort to develop the subtypes and validated their findings with data from the National Institute of Neurological Disorders and Stroke (NINDS) Parkinson’s Disease Biomarkers Program (PDBP) cohort.
One exciting aspect of their research is the potential for repurposing existing drugs to target the specific molecular changes in each subtype.
By analyzing patient health records from large databases like the INSIGHT Clinical Research Network and the OneFlorida+ Clinical Research Consortium, the researchers identified that the diabetes drug metformin might improve Parkinson’s symptoms, particularly for those with the PD-R subtype who are more likely to experience cognitive deficits early on.
Dr. Chang Su, the first author of the study, noted that patients taking metformin showed improvements in symptoms related to cognition and falls.
This finding suggests that metformin, commonly used to treat diabetes, could be a promising candidate for slowing the progression of Parkinson’s, especially in the PD-R subtype.
Dr. Wang hopes this research will inspire other scientists to use diverse data sources in similar studies. He also believes that further validation, both computationally and experimentally, will be crucial to confirming these findings.
This groundbreaking work involved contributions from many collaborators, including scientists from the Cleveland Clinic, Temple University, University of Florida, University of California at Irvine, and University of Texas at Arlington.
Doctoral candidates from Cornell Tech’s computer science program and Cornell University’s computational biology program also played key roles in the research.
By uncovering these subtypes and potential treatments, the study opens new avenues for personalized medicine in Parkinson’s disease, offering hope for better management and improved outcomes for patients.
If you care about Parkinson’s disease, please read studies that Vitamin B may slow down cognitive decline, and Mediterranean diet could help lower risk of Parkinson’s.
For more information about brain health, please see recent studies that blueberry supplements may prevent cognitive decline, and results showing Plant-based diets could protect cognitive health from air pollution.
The research findings can be found in npj Digital Medicine.
Copyright © 2024 Knowridge Science Report. All rights reserved.
