FDA AdComm Votes Down BrainStorm Cell Therapeutics’ ALS Cell Therapy NurOwn
Committee members were overwhelmingly against the available data supporting NurOwn as an effective treatment for mild to moderate ALS, with 17 no votes, 1 yes vote, and 1 abstention.
An FDA Cellular, Tissue, and Gene Therapies Advisory Committee (AdComm) has voted against BrainStorm Cell Therapeutics’ debamestrocel (NurOwn), an investigational autologous mesenchymal stem cell (MSC) neurotrophic factor–secreting cell therapy product intended to treat amyotrophic lateral sclerosis (ALS), ahead of its Prescription Drug User Fee Act action date, which is slated for December 8, 2023.1 Notably, in the days leading up to the AdComm meeting, BrainStorm requested to change the indication targeted with NurOwn from “ALS” to “mild to moderate ALS”.
The committee overwhelmingly voted negatively, with 17 members voting no and 1 member voting yes to the question: “Do the data presented demonstrate substantial evidence of effectiveness for treatment of mild to moderate ALS?” One member abstained. In the public hearing portion of the AdComm, a number of patients, caregivers, and family members spoke in favor of approval based on their personal experiences with NurOwn.
“It's possible that this therapy has some benefit for some patients,” Lynn A. Raymond, MD, PhD, the director of the Djavad Mowafaghian Centre for Brain Health, said after voting no.1 “But we look at the total when we have to decide whether this goes forward to public market and there wasn't evidence for the whole group that this was effective and there was evidence to potentially suggest it was actually deleterious at least for those who are maybe more advanced with ALS, causing more deaths and causing more bulbar dysfunction. So for that reason, I vote no...”
Do you agree with the FDA Advisory Committee's recommendation against approval of BrainStorm Cell Therapeutics' NurOwn as a treatment for ALS?
BrainStorm’s argument in favor of approval of the biologics license application (BLA) largely focused on the potential impact of the floor effect on the results of the pivotal phase 3 clinical trial (NCT03280056) that evaluated the treatment against a placebo in patients with ALS, which did not meet any of its primary or key secondary end points. Regarding the floor effect, Brainstorm noted that in a prespecified subgroup of patients with less advanced disease, who had a score of at least 35 on the Revised Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS-R), a statistically significant difference in the change from baseline in ALSFRS-R was observed favoring the group of patients who received NurOwn compared with the group of patients who received a placebo.2 A posthoc sensitivity analysis of patients who had an ALSFFRS-R score of 26 or higher also found that participants treated with NurOwn progressed 2 points less on the ALSFRS-R scale on average compared with patients who received the placebo.
BrainStorm also included data from the phase 3 study showing changes in biomarkers including neurofilament light (NfL) in its argument. After accounting for baseline disease covariates, treated participants had reduced NfL values from baseline to week 20 compared to placebo (P < .05).3
The FDA’s response focused mainly on the fact that the prespecified primary and key secondary end points were not met and that posthoc analyses, such as the analyses related to the floor effect that BrainStorm presented, are not sufficient to support approval on their own. The FDA emphasized that prespecification is the cornerstone of reliable evidence in clinical trials and that the results seen in the posthoc analysis are not reliable. An additional concern brought up was that more deaths occurred in the group of patients treated with NurOwn in the trial than in the placebo group. With regard to the NfL and other biomarker data from the trial, the FDA stated that the analysis did not support efficacy, as a greater NfL reduction was associated with worse outcomes in the modified intention to treat population and there was no clear association between changes in the levels of other biomarkers, including galectin-1, LAP (TGFβ-1), MCP-1, and VEGF, and measurements of clinical benefit. Further concerns noted by the FDA were the absent or incomplete critical manufacturing controls and lack of established product quality.
Overall, the FDA concluded that the available data do not support approval of NurOwn for either ALS or mild to moderate ALS indications. The agency suggested that the exploratory data regarding the floor effect may inform the basis of a new clinical trial aimed at demonstrating efficacy, if BrainStorm chooses to continue to pursue the product’s approval.
“I think it is very clear that I include data that no one else considers as real data...” Kathleen O’Sullivan-Fortin, Esq., the founder of ALD Connect, Inc., said while making the sole ‘yes’ vote.1 “I think there's no bigger risk than imminent certain deaths from ALS and these are unique and desperate circumstances that would require us to exercise flexibility. With everyone else here, I wish that the sponsor’s data was definitive and broad and fixed and helped and saved everyone, and it wasn't going to be that, but I still think there was a value.”
The meeting concluded with a discussion of potential design decisions for future clinical trials evaluating the therapy that could more effectively demonstrate efficacy. Some suggestions included accounting for the floor effect in prespecified analyses, using a different measure of efficacy than the ALSFRS-R in order to avoid the floor effect altogether, and developing an allogeneic version of the therapy in order to ensure more standardization of the product ultimately administered to patients.
“My suggestion to the sponsor would be to add a key secondary end point or quality of life measure that would capture the anecdotal improvements in daily activities and capture the testimonials of benefit [mentioned during the public hearing],” Wendy B. London, PhD, an associate professor of pediatrics at Boston Children’s Hospital/Dana-Farber Cancer Institute, who voted no, added during this portion.1
