MDA 2023: Issues with ALS scale scores noted in NurOwn trial failure
Analysis recommends new progression scores be used in ALS trials
Difficulties in measuring the progression of amyotrophic lateral sclerosis (ALS) among patients with advanced disease can cause complications in clinical trials, according to a new analysis that highlights how such problems may have affected the failed NurOwn study.
The work was led by scientists at BrainStorm Cell Therapeutics, which is developing the cell therapy NurOwn as a potential treatment for ALS.
Late last year, the U.S. Food and Drug Administration (FDA) declined to consider an application from BrainStorm seeking NurOwn’s approval. The company now is working to identify the next steps for the experimental therapy.
The findings of the new analysis were shared at the recent Muscular Dystrophy Association’s MDA Clinical & Scientific Conference, in a poster titled “Measuring the rate of impairment in ALS patients using the Revised-ALS Functional Rating Scale: Key Insights into the Floor Effect of the Scale.”
“While widely used, the [current] scale is hampered by its ability to measure function in those with low functional status,” such as individuals with advanced disease, the researchers wrote.
More than 20% of patients in NurOwn trial had low ALSFRS-R scores
NurOwn is a cell-based therapy in which mesenchymal stem cells — cells with the capacity to transform into a variety of cell types — are isolated from the bone marrow and matured in the lab to produce large amounts of signaling molecules that promote nerve cell survival and growth.
As part of the treatment’s development, BrainStorm had conducted a Phase 3 clinical trial (NCT03280056) to test NurOwn against a placebo in people with fast-progressing ALS. Completed in 2020, that trial involved 189 patients, ages 18-60. Its main goal was to test whether the treatment could slow the progression of ALS symptoms.
The trial, however, failed to meet this goal: the rate of progression was not significantly different between patients given NurOwn or a placebo.
Like many ALS trials, that Phase 3 study of NurOwn used a measure of disease progression called the ALS Functional Rating Scale-Revised (ALSFRS-R).
This scale assigns numeric scores to a patient’s ability to function across several categories, from swallowing to turning in bed, which are then added to calculate the total score. Higher scores reflect a better ability to function, and scores tend to decrease as ALS progresses and patients lose functional abilities.
In the MDA poster, researchers highlighted that the ALSFRS-R is subject to a so-called floor effect: once a particular measure has reached zero, it can’t go any lower, even if the patient is continuing to lose functionality. As a result, ALSFRS-R scores can plateau or stabilize in patients with advanced disease, failing to reflect the progression that the patient is actually experiencing.
Through statistical analyses, the researchers showed that more than 1 in 5 patients (22.3%) given a placebo in the Phase 3 trial exhibited a floor effect — ALSFRS-R scores specifically tended to plateau around 27 points.
For future trials, these data suggest requiring ALSFRS-R items be at least 2 or seeking a more suitable measure for patients with advanced ALS.
Analyses of data for nearly 4,000 ALS patients from a large database called PRO-ACT showed that nearly 5% of patients exhibited a similar floor effect, usually around 25 points on the ALSFRS-R, suggesting that this issue is not unique to the NurOwn trial.
Floor effects were most common among patients who started off with lower ALSFRS-R scores. The researchers noted that the average starting ALSFRS-R score in the NurOwn trial was less than 32 points, which is markedly lower than for other modern ALS trials, where average scores have been 36 or higher.
The researchers then re-analyzed data from the NurOwn trial, but assessing only data for the 106 patients who did not show any signs of a floor effect at the start of the study. Results showed that, over the course of the six-month Phase 3 study, the average decline in ALSFRS-R scores was 2.68 for patients given NurOwn and 4.99 for patients given a placebo.
“Analyses conducted on those not impacted by the floor … reveal statistically significant, clinically meaningful treatment effects with NurOwn,” the researchers wrote.
To prevent issues with the floor effect in future trials that use the ALSFRS-R, the researchers suggested that it may be beneficial to limit studies to only patients who have a score of at least 2 on all parts of the ALSFRS-R.
“For future trials, these data suggest requiring ALSFRS-R items be at least 2 or seeking a more suitable measure for patients with advanced ALS,” the team wrote.
About the Author
