Access in ALS: lowering the bar to raise the horizon

ALS patients, advocates are calling for greater regulatory flexibility from FDA

The ALS community is demanding greater regulatory flexibility from FDA, arguing for drugs to be approved even when efficacy is in doubt and there’s no clear biomarker. Fired up by the approval of Aduhelm this year for Alzheimer’s disease, ALS patients and advocates want the agency to apply the same logic — that some therapy is better than none, and even long odds are better than no odds when circumstances are as dire as theirs. 

What counts as the minimum clinically meaningful benefit, and how much additional benefit is needed to offset a given amount of risk, are thorny issues in many diseases. They’re especially difficult in severe diseases with little or no history of clear clinical wins that could help set standards.

Tuesday’s announcement that  Amylyx Pharmaceuticals Inc. has submitted an NDA to FDA for AMX0035 to treat amyotrophic lateral sclerosis after months of discussions with the agency is a win for the ALS community, but beyond that it has been largely a frustrating year of regulatory setbacks and clinical failures.

In March, FDA sought to dampen expectations for a product from BrainStorm Cell Therapeutics Inc. (NASDAQ:BCLI) by taking the highly unusual step of commenting on the results of a Phase III trial before taking a regulatory action. “It has become clear that data do not support the proposed clinical benefit of this therapy,” the agency said of NurOwn, which had become a rallying point for patients and advocates.

In June, ALS patients watched as FDA approved Aduhelm aducanumab from Biogen Inc. (NASDAQ:BIIB) for Alzheimer’s based on controversial data that ALS patient advocates argue reflects a much more open approach than the agency has shown in their disease. The community believes that such flexibility is called for in a 2019 FDA guidance on ALS.

And in the past three weeks, at least two more therapies have failed trials for ALS. On Oct. 17, Biogen reported that its antisense therapy tofersen missed the primary endpoint in a high-profile Phase III study, and on Nov. 2 Clene Inc. (NASDAQ:CLNN) announced that CNM-Au8, a suspension of gold nanocrystals that stimulates antioxidant activity and supports cellular ATP production, missed the primary MUNIX biomarker endpoint and secondary forced vital capacity endpoint in the Phase II RESCUE-ALS study.  

Still, the ALS community received a few other boosts this year in addition to Amylyx’s progress, including the sense of urgency and broad bipartisan support demonstrated by lawmakers at a July 29 hearing by the House Energy & Commerce’s health subcommittee on neurodegenerative diseases. 

Rep. Anna Eshoo (D-Calif.), who chairs the subcommittee, channeled the desperation of the community, as she set the stage for a forum seeking clarity on the steps necessary to gain full approval for efficacious therapies in diseases such as ALS. Quoting her constituent Jamie Berry, she read: “With ALS, a piece of you dies every single day. We are simply asking for a fighting chance to live the lives we were meant to live.”

Berry’s case shows the extraordinary speed with which the disease can affect a patient — she was diagnosed July 14, 2020; 15 months later she is in a wheelchair and has lost nearly all use of her limbs. Disease progression is typically characterized by progressive muscle weakness with death occurring an average of three to five years after disease onset. 

FDA expressed sympathy with patients, but held its ground that the agency’s hands have been tied by the lack of efficacy and the absence of biomarkers in therapies reviewed so far. 

Patrizia Cavazzoni, director of FDA’s Center for Drug Evaluation and Research (CDER), told the hearing that the agency acknowledged the lack of new treatments is “deeply frustrating” for people living with ALS and their caregivers but pointed to the need for better translational models and biomarkers.

“While we have seen advances in some neurodegenerative diseases, in many other instances, such as with ALS, there are no easily measured biomarkers that are reliable predictors (or surrogates) for the rate of progression in individual patients,” Cavazzoni said in prepared remarks. “Such tools could improve the precision with which drug response could be evaluated, providing researchers with much more robust, and earlier, insights to distinguish the more promising drugs from those that are less likely to succeed.”

Patients also want greater access to therapies before approval, through broadening of expanded access and compassionate use pathways. Most ALS patients don’t qualify for clinical trials because standard inclusion criteria prevent participation when the disease has progressed beyond the earliest stages.

Biomarkers and surrogates

The lack of a clear identifying cause in most cases of the disease, and a highly heterogeneous patient population compound the difficulty of developing therapies. Without biomarkers of function, the field may stay locked out of FDA accelerated approvals for a while, because of the requirement of a surrogate endpoint that is “reasonably likely” to predict clinical benefit.  

Despite this year’s bad news, the field is ripe with activity, with more than 40 compounds disclosed, addressing at least 12 mechanistic rationales. While many drug developers are embedding potential markers in their trials — notably neurofilament light chain (NEFL; NF-L), a biomarker of axon degeneration — it’s not clear if any are working on validating a biomarker to be used as a surrogate endpoint.

In its guidance, FDA encourages sponsors to “incorporate exploratory biomarkers in all phases of development,” but does not identify any biomarker that may be appropriate as a surrogate endpoint.  

Tofersen had one of the strongest cases for hope because it targets SOD1, a gene that is mutated in about  2% of ALS cases, including 10-20% of familial cases, giving the compound a built-in biomarker that ties directly to disease biology in the patients in the study, which enrolled SOD1 mutation carriers. The antisense oligo, which was developed by Ionis Pharmaceuticals Inc. (NASDAQ:IONS), disrupts expression of SOD1.

Though Biogen wasn’t necessarily planning to seek accelerated approval, considering it took the drug into a Phase III trial with hard efficacy endpoints, VALOR’s design combined with tofersen’s mechanism raised the question whether the therapy could end up as Biogen’s second surprise accelerated approval.

Because Biogen decided to test tofersen in SOD1 mutation carriers in VALOR, the company made the study population similar to that of a genetic disease such as Duchenne muscular dystrophy (DMD). The drug reduced SOD1 levels in cerebrospinal fluid 26% and 38% in the trial’s slow-progressing and fast-progressing patient subsets, respectively.

Biogen has not announced its next step for the compound but noted it is “actively engaging with regulators, the medical community, patient advocacy groups and other key stakeholders.” 

The patient community is drawing parallels between tofersen in ALS, and Aduhelm in Alzheimer’s and Exondys 51  from Sarepta Therapeutics Inc. (NASDAQ:SRPT) in DMD, the latter two having been granted accelerated approval by FDA. 

Though tofersen did not produce a functional improvement, the amount of SOD1 knockdown required for functional benefit isn’t known. The same scenario was seen with Exondys 51 in DMD, and though that product and two of its successors for other DMD subpopulations received accelerated approval, it’s not clear that FDA is extrapolating that standard broadly in neurodegeneration. 

Aduhelm also failed to show convincing functional improvement, but FDA was convinced by the genetic link between β-amyloid and Alzheimer’s disease, and the reduction of β-amyloid produced by the drug, to grant accelerated approval. In that case, the patient population is enormous. The decision created one of the biggest controversies the agency has ever faced, and the link between amyloid and the disease is still questioned by many in the field.

Amylyx advances

Amylyx’s regulatory submission for AMX0035 is a signal that the agency is willing to change its thinking after working with a company during the regulatory process.

A year ago, the company believed it would have to complete an additonal study beyond its CENTAUR study before submitting its NDA to the agency. But a persistent dialogue between Amylyx executives and agency staffers, supplemented by additional data, changed FDA’s stance. The discussions came amid petitions by influential patient advocacy groups I AM ALS and the ALS Association that called on the agency to approve AMX0035 without a longer, larger study.

The company initially described the 137-patient CENTAUR study as a Phase II trial, but now refers to it as a pivotal Phase II/III study.

In the end, the agency encouraged Amylyx to seek approval, co-CEO Josh Cohen told BioCentury before the submission. 

Cohen told BioCentury Tuesday: “We do not know which review pathway will be utilized for AMX0035, and this will be determined by FDA.”

In a modified intent-to-treat population in CENTAUR, patients receiving Amylyx’s AMX0035 had a slowing of disease progression measured as 0.42 points per month less decline as measured by the Revised Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS-R) than those receiving placebo (p=0.03). ALSFRS-R is the most commonly used scale to measure function in the indication. Results were published in The New England Journal of Medicine.

Data published in Muscle & Nerve showed that after up to 35 months of follow-up, there was a 44% lower risk of death among ALS patients who received AMX0035 compared with patients who originally received placebo in CENTAUR, regardless of whether they enrolled in the study’s open-label extension (HR=0.56; 95% CI: 0.34, 0.92; p=0.023). Median overall survival was 25.0 months versus 18.5 months in the two groups, for a difference of 6.5 months.

NurOwn differences

Cavazzoni and acting FDA Commissioner Janet Woodcock have outlined what the agency needs to see to approve a therapy for ALS. The patient community has its own perspective, and NurOwn has been a flash point of the difference between the two.  

“While we strongly desire a curative therapy, a treatment that provides a meaningful incremental benefit would still be desirable,” Cavazzoni said at the July congressional hearing. “A drug with a dramatic effect size might be adequately evaluated without a comparator, but to detect a more modest, yet very relevant benefit, only a controlled trial — comparing the candidate to placebo — can serve.” 

FDA has yet not seen enough in BrainStorm’s NurOwn to change its approach, but members of the ALS community are clamoring for the product to be approved.

BrainStorm has now released biomarker data for its therapy in a poster for the Northeast Amyotrophic Lateral Sclerosis (NEALS) Consortium meeting in early October. NurOwn, which comprises autologous bone marrow derived mesenchymal stem cells induced to secrete neurotrophic factors, produced changes in biomarkers, such as MCP-1 and VEGF-A. 

Stacy Lindborg, BrainStorm EVP and head of global clinical research, believes further evaluating biomarkers will enable the company to build the case for NurOwn.

“Not only do we see important changes in biomarkers, but they’re relevant and do give insight into the trial,” Lindborg, who is a veteran of Biogen, told BioCentury. She added that the company is learning that “biomarkers that are predictive of clinical response, such as these, span multiple pathways that are important to ALS and align with NurOwn’s mechanism of action.”

Patient advocates point to the responses in small numbers of patients as sufficient for approval, regardless of whether there’s a test that can identify responders.

Berry and hundreds of other patients called on Woodcock early this year to approve the therapy.

Woodcock responded to Berry to explain the agency’s stance.

“I understand that ALS is horrible and unfair. The way we have found hope for a lot of serious diseases that killed people like testicular cancer, AIDS, other cancers and so forth was to do clinical trials and find treatments that worked at least a bit and then build on that till we got to life-saving therapies. I surely hope we get there with ALS and as soon as possible,” Woodcock wrote in an email that Berry shared with BioCentury.

Michelle Lorenz, head of Voices for ALS, points to a June 2019 meeting at which she says Woodcock gave a “simple efficacy test” summed up as: “If it helps you function, and you can prove it, we’ll approve it!”

Lorenz, Berry and other advocates cite multiple accounts of ALS patients in wheelchairs and on respirators who, after receiving NurOwn, have experienced dramatic improvement.

However, 27.7% of patients in the placebo arm of BrainStorm’s Phase III trial also qualified as “responders” on the trial’s primary endpoint of improvement of 1.25 points per month as measured by ALSFRS-R, versus 34.7% of NurOwn-treated patients. The difference was not statistically significant (p=0.453) and suggested the number of true responders to the therapy may be about 7%.

Though BrainStorm did some predictive biomarker modeling in its Phase III study, it’s not yet possible to identify who those responders are before treatment. 

Data from NurOwn’s secondary efficacy endpoint did not help its case. Change in ALSFRS-R total score from baseline to week 28 was -5.52 for the NurOwn arm versus -5.88 for placebo (p= 0.693).

ALSFRS-R is the most commonly used scale to measure function in the indication; however, patient advocates point to the endpoint’s limitations, including a lack of sensitivity of some of the scale’s 48 points.

BrainStorm is planning to publish detailed data from the trial in a peer-reviewed journal, Lindborg said, but did not give a timeline. The company is also planning to seek input from ALS KOLs, before deciding on its next steps, including “formal engagement with FDA.”