Machine learning reveals potential Parkinson’s disease treatments in database

Screening a large database with machine learning tools helped scientists identify drugs approved for other diseases that reduced the risk of developing Parkinson’s disease, a study found.

According to the researchers, these potential treatments for Parkinson’s disease “deserve confirmation” in larger studies.

“Drug repurposing or repositioning is the application of a known drug to new indications and can lead to shorter, inexpensive drug development cycles with an increased likelihood of success,” the team wrote.

Among the drugs identified for possible further study were simple sulfonamide diuretics, particularly furosemide, which are used to reduce excess body fluid caused by conditions such as heart failure. Exposure to these drugs was linked to a reduced risk of Parkinson’s, the scientists noted.

The Machine Learning Study,”Identifying protective drugs for Parkinson’s disease in healthcare databases using machine learningwas published in the journal Movement disorders.

recommended reading

Using machine learning in research

The treatments available for Parkinson’s disease to date are only partially effective and fail to significantly delay the progression of the disease. Thus, there is growing interest in repurposing existing drugs as an accelerated method of therapeutic development.

These approved treatments, having already been rigorously tested in clinical trials, generally have established safety profiles.

Studies have suggested that people treated with certain medications, including immunosuppressants or those that widen the airways, called bronchodilators, have a lower risk of developing Parkinson’s disease.

These findings prompted researchers based at the University of Paris-Saclay, France, to use machine learning tools to automatically screen a large database of marketed therapies to detect those linked to a lower risk of heart disease. Parkinsons.

“This study is part of a research effort to identify previously developed compounds associated with reduced [Parkinson’s disease] risk,” the researchers note.

Data were collected from the French national health information system. A total of 40,760 incident patients with Parkinson’s were identified based on details of at least one claim for an anti-Parkinson’s drug from 2016 to 2018. A control group of 176,395 people of age, sex and area of Similar residence was included for comparison. .

As the data was accessible from 2006, a follow-up of at least 10 years was available before a diagnosis of Parkinson’s. An eight-year lag before the date of diagnosis was considered to be due to the long prodromal or pre-diagnostic phase of Parkinson’s disease. Notably, many patients face a late diagnosis of Parkinson’s disease because its symptoms are common to other disorders.

Given this, the team assessed therapeutic exposure and associated factors during the two years before the latency period (period of exposure) to find associations with a reduced risk of developing Parkinson’s disease.

The machine learning analysis identified eight potential therapies under six subgroups.

recommended reading
inflammation of the brain

Identify potential therapies

The strongest signal corresponded to simple sulfonamide diuretics, drugs that promote the release of water from the body. Among them, furosemide – used to reduce excess body fluid caused by conditions such as heart failure, liver disease and kidney disease – was the most represented. Both frequent and constant exposure – any exposure is in contact with a drug at any time – to furosemide stood out.

A suggestive signal came from frequent exposure to drugs linked to nicotine addiction. Nicotine is the chemical in tobacco that keeps smokers from quitting. This exposure was evenly split between users of nicotine and varenicline, a therapy to help people quit smoking. Of these two drugs, only regular exposure to varenicline was associated with a lower risk of Parkinson’s.

Insulin aspart, a short-acting lab-made version of human insulin, was the most represented substance among insulin and related therapies, with only sporadic use as a signal.

Among adrenergic, adrenaline-related drugs, frequent exposure to the lung disease bronchodilator formoterol, together with the anti-inflammatory corticosteroid budesonide, generated a signal. This signal matched those already exposed to the immunosuppressive mycophenolic acid.

Subgroups of soft paraffins and fatty products and direct-acting muscle relaxants were also identified via machine learning, but no signals were detected for specific drugs within these classes.

Following these tests, the team refined the analysis using a more precise definition of Parkinson’s disease. Here, 29,873 patients with Parkinson’s disease and 176,395 controls were evaluated. Overall, these analyzes yielded results consistent with those of the group as a whole.

All signals identified in the primary analysis were also generated using the refined group, with the exception of insulin and related therapies. The signal for single sulfonamide diuretics, especially furosemide, was stronger compared to the primary analysis.

Two signals were regenerated for the two main treatments for nicotine dependence (nicotine and varenicline), alongside any exposure to the bromodilator tiotropium bromide and direct-acting muscle relaxants.

New signals not found in the primary analysis were identified, including medical moisturizers called emollients used to treat skin conditions, as well as anti-inflammatory drugs and non-steroids for topical use. Anticholinergics, or substances that block the action of the nerve cell signaling molecule called acetylcholine, and mucolytics – used to clear mucus from the airways – also generate a signal.

“The search for new [Parkinson’s disease] drug repositioning therapies have attracted attention given the current lack of fully satisfactory treatment options,” the researchers wrote. “We have reviewed… a large number of drugs and identified single sulfonamide diuretics as a chemical subgroup of drugs potentially inversely associated with [Parkinson’s disease] risk.”

“Our results lead to new hypotheses that deserve to be replicated and could lead to the development of new therapeutic or preventive strategies in [Parkinson’s disease]“, added the researchers.

Comments are closed.