3 distinct Parkinson’s subtypes seen based on disease progression
AI used to analyze records over time of 406 newly diagnosed patients
Machine learning, a form of artificial intelligence (AI), was able to define three distinct subtypes of Parkinson’s disease based on how fast symptoms in newly diagnosed patients progressed, potentially helping doctors to more accurately detect the disease and predict its course.
In the study led by Cornell University researchers, each subtype appears to have its own genetic signature, which could offer targets for more personalized treatment with new or existing medications.
“We may need to consider customized treatment strategies based on a patient’s disease subtype,” Fei Wang, PhD, the founding director of the Institute of AI for Digital Health (AIDH) at Weill Cornell Medicine in New York, said in a university news story.
The study, “Identification of Parkinson’s disease PACE subtypes and repurposing treatments through integrative analyses of multimodal data,” was published in npj Digital Medicine.
Parkinson’s progresses in different ways and at differing rates in patients
Not every person with Parkinson’s experiences the disease’s motor and nonmotor symptoms in the same way or at the same pace. This can make it difficult to diagnose Parkinson’s in its early stages or decide which disease treatment is likely to be most effective.
“Parkinson’s disease is highly heterogeneous, which means that people with the same disease can have very different symptoms,” said Wang, a professor of health informatics at Weill Cornell. “This indicates there is not likely to be a one-size-fits-all approach to treating it.”
Wang and his team turned to machine learning to recognize patterns in how symptoms manifest over five or more years, by feeding computers disease progression data covering 406 adults with newly diagnosed Parkinson’s. The group’s average age at onset was 59.6, and a majority (65.5%) were men.
Machine learning is a branch of artificial intelligence that involves the use of algorithms that analyze and learn from existing data, enabling computers or machines to “learn” and make predictions.
“We used five-year longitudinal records of individuals in over 140 items of diverse motor and non-motor assessments … [to build] a deep learning model,” the researchers wrote.
Three distinct subtypes were defined based on data from the Parkinson’s Progression Markers Initiative (PPMI) and validated with data from the National Institute of Neurological Disorders and Stroke (NINDS) Parkinson’s Disease Biomarkers Program (PDBP).
The first subtype, called Inching Pace (PD-I), involved 145 patients. All had mild motor and nonmotor symptoms at diagnosis, including milder sleep and cognitive impairments, and experienced slow disease progression. Fewer than half (46.2%) were men.
The 207 people classified under the second subtype, Moderate Pace (PD-M), started with mild symptoms but progressed at a moderate speed, with symptoms continuing to worsen from the second year of follow-up onward compared with the PD-I group. A majority of PD-M patients, 155 or 74.9%, were men.
Almost 82% of patients in the fastest progressing group were male
The Rapid Pace (PD-R) subtype, which included 54 patients, was marked by fast disease progression, with the most rapid rate of symptom worsening. These patients were assessed as having the most severe motor symptoms at Parkinson’s onset, and greater nonmotor symptoms, particularly affecting cognition, the researchers noted. Patients with PD-R, on average, were older at disease onset than the other patients — 64.4 years — and 81.5% of this group were men.
To identify a genetic signature for each of the three subtypes, the team focused on roughly 90 known genetic variants linked to Parkinson’s, including APOE, which has been tied to faster cognitive decline and a higher risk of death in patients.
For example, PD-R — the fastest progressing subtype — could be identified by a set of 14 genes, most of which were linked to inflammation and oxidative stress, two mechanisms known to affect how Parkinson’s develops and progresses.
Researchers also looked for differentially expressed genes, those that become more active (upregulated) or less active (downregulated), with each subtype. Gene expression refers to the process whereby information encoded in a gene is turned into a product, like proteins.
In total, the PD-I subtype had 2,176 differentially expressed genes, PD-M had 2,376, and PD-R subtype had 2,305 genes that were expressed differentially compared with healthy individuals. Different modules of around 210 to 240 genes were specified to each subtype.
Potential seen for treating Parkinson’s using metformin, a diabetes drug
Beyond defining the different subtypes, the researchers explored medications that could target the specific genetic signatures in each subtype. In addition to disease-approved treatments like levodopa, they looked at medications for other diseases that could be repurposed for these Parkinson’s subtypes.
Drawing on two real-world databases, the INSIGHT Clinical Research Network and the OneFlorida+ Clinical Research Consortium, they identified metformin, approved for type 2 diabetes, as particularly promising.
“We found that people taking the diabetes drug metformin appeared to have improved disease symptoms — especially symptoms related to cognition and falls — compared with those who did not take metformin,” said Chang Su, PhD, the study’s first author.
This was especially evident in those with the PD-R subtype, who are more likely to experience cognitive deficits during early disease stages.
Metformin’s safety in treating Parkinson’s patients, however, still must be “rigorously evaluated through dedicated studies,” the scientists noted.
“This work helps better understand [the] clinical and pathophysiological complexity of [Parkinson’s] progression and accelerate precision medicine,” the researchers concluded. Pathophysiology refers to changes that cause or occur as a result of disease.
While more work is needed to validate this study, its findings support “the existence of different pathophysiologic mechanisms driving different [Parkinson’s] subtypes,” leading to distinct “progression trajectories” of clinical relevance, they added.
About the Author
