Model That Can Predict Placebo Effect May Help Clinical Researchers
Researchers at Tools4Patient, which is developing Placebell, and other institutions presented data on the approach at the International Parkinson and Movement Disorder Society Virtual Congress 2021, held online Sept. 17–22, in a poster, titled “Modeling the Placebo Response in Parkinson’s Disease.”
“The Parkinson’s disease study extends Placebell’s success as a platform solution for multiple indications,” Dominique Demolle, PhD, CEO of Tools4Patient, said in a press release.
The placebo effect is a powerful but poorly understood phenomenon where a person who is ill may get better after taking a “treatment” that is known to be ineffective. The classic example is a pill made of sugar instead of medicine.
In the context of clinical trials, which are the gold standard for determining whether or not an experimental treatment works, the placebo effect can complicate results, making it hard to tell whether patients are actually responding to a treatment or getting better because they believe that they will.
“The placebo response is a significant barrier to demonstrating efficacy of experimental therapies for Parkinson’s disease,” said Olivier Rascol, MD, PhD, a professor at Toulouse University Hospital in France and co-author of the poster.
The basic idea behind Placebell is to estimate the amount of benefit that Parkinson’s patients in a clinical trial are likely to get that is purely attributable to the placebo effect. Then, researchers can mathematically remove the placebo-derived benefit from trial results to determine how much of the benefit is attributable to a treatment itself.
To estimate the placebo effect in Parkinson’s patients, researchers conducted a small trial in which 94 people with Parkinson’s were given an oral “treatment,” which they were unaware was a placebo, for three months. About two-thirds of the patients were male, and the mean age was 64 years.
Before starting on placebo, and after one and three months of the placebo, the participants underwent a battery of disease-relevant assessments, such as the Movement Disorder Society Unified Parkinson’s Disease Rating Scale (MDS-UPDRS).
Results indicated a small but substantial placebo effect. Participants experienced statistically significant improvements in motor symptoms one month after starting on placebo, and also 30 minutes after taking the pill. The change in motor symptoms at three months was numerically similar, but not statistically significant.
The researchers then constructed mathematical models of the data, using factors like disease severity at the study’s start and patients’ psychological traits. They then calculated how much variability might be removed from patient responses by applying these mathematical models. For example, in part 3 of the MDS-UPDRS, the part of the test that assesses motor symptoms, the models were estimated to remove 33.2% of the variability.
This revealed that factors that associated with the placebo response were related to baseline intensity of Parkinson’s, patients’ psychological traits, as well as other factors.
“This study is a significant step towards the prediction of the placebo response in Parkinson’s disease [randomized clinical trials],” the researchers concluded.
Rascol added: “This predictive model of the placebo response in Parkinson’s disease is an important step forward in managing this issue and accelerating the delivery of medicines to this patient population.”