FDA's nagging doubts about Sarepta's DMD gene therapy won't quit as advisory meeting looms

The key question that advisers will weigh when they meet Friday to discuss Sarepta Therapeutics’ Duchenne muscular dystrophy gene therapy is this: Is there enough evidence to support accelerated approval for use in ambulatory patients who have a confirmed mutation of the DMD gene?

The FDA, for its part, seems to be unconvinced. After reviewing all the evidence, “FDA notes that the clinical studies conducted to date do not provide unambiguous evidence that SRP-9001 is likely beneficial for ambulatory patients with DMD,” according to briefing documents (PDF) posted ahead of the Friday meeting.

“It is challenging to conclude with reasonable certainty from the data provided by the applicant either that SRP-9001 is likely effective for younger patients, or that it is likely ineffective for older patients or those with somewhat poorer functional status,” the FDA wrote. The agency also flagged safety concerns associated with SRP-9001 “related to the possibility of administering an ineffective gene therapy.”

The meeting of the Cellular, Tissue, and Gene Therapies Advisory Committee has been called to discuss four key topics. However, just one overarching question will be voted on: Does the available evidence, taking into consideration the uncertainties, support accelerated approval of SRP-9001 using a surrogate endpoint for the treatment of ambulatory patients with DMD who have a confirmed DMD gene mutation?


Question 1

The first question to be discussed is whether Sarepta’s proposed surrogate endpoint, which would qualify SRP-9001 for an accelerated approval, is reasonably likely to predict clinical benefit. The biotech has asked for advanced approval based on the gene therapy's ability to spur expression of micro-dystrophin protein at week 12. The FDA is skeptical that there’s any evidence to suggest that this biomarker proves a pharmacological effect from SRP-9001, but rather that it simply suggests the protein has been expressed in the target cells.

SRP-9001 is meant to change the disease trajectory of DMD to a milder form called Becker muscular dystrophy. To do that, the gene therapy is said to correct the expression of the dystrophin protein in the muscles, which Sarepta theorizes would be a correction of the underlying cause of DMD.

The FDA has had concerns about the clinical benefit of SRP-9001 since as far back as December 2018, when the agency questioned the surrogate endpoint and recommended that Sarepta pick one “that assesses clinically meaningful benefit, as manifested by how a patient feels, functions, or survives,” according to a regulatory timeline shown in the documents. The issue was again brought up in September 2020, July 2021 and April 2022 at various meetings with the company.

Sarepta, in the final interaction before filing the biologics license application, cited regulatory precedent for granting accelerated approval to drugs that promote expression of shortened forms of dystrophin. The FDA countered, saying that this is not necessarily equivalent to predicting clinical benefit to support accelerated approval.

The BLA was submitted on September 28, 2022.

Sarepta argues in its submission (PDF) that replacement of the protein has been shown through natural history studies to have clinical benefit and slow disease progression.

“Restoring functional dystrophin to patients with DMD has long been considered a critical and widely accepted therapeutic goal in the treatment of DMD,” Sarepta said, arguing that SRP-9001 has been shown to do just that. The company presented multiple different preclinical studies to support this.


Question 2

The second question addresses the only randomized, double-blind, placebo-controlled clinical study of SRP-9001 that has data available to date, part 1 of the phase 2 Study 102. The main goal was assessing the treatment’s effect on the North Star Ambulatory Assessment, or NSAA, which is a common rating scale used to assess functional motor abilities, progression and treatment effects in ambulant children who have DMD. The study did not demonstrate a statistically significant change on this measure at week 48 when compared to a placebo. The group that received the gene therapy showed no improvement from baseline.

Instead, Sarepta pointed to a subgroup analysis that suggested better outcomes in ambulatory patients between 4 to 5 years old. But patients aged 6 to 7 years showed no improvement. At the time, the company blamed an apparent imbalance between the two arms of the trial, arguing that the placebo patients had a higher functional score on the NSAA at baseline.

Sarepta has asserted that the mean change from baseline in NSAA score was “numerically greater at all time-points” for patients who received SRP-9001, according to the FDA.

“FDA’s assessment is that the difference between the SRP-9001 and placebo groups at all time-points is well within uncertainty bounds, which is also demonstrated by the lack of even a trend toward statistical significance,” the agency wrote.

Later on, the FDA suggests that, while age is an important factor in disease progression for DMD, the post-hoc analysis was not prespecified.

“Post hoc subgroup tests following an overall nonsignificant test in the population as a whole can only be considered hypothesis-generating, and this subgroup analysis therefore must be interpreted with caution,” the FDA said.

Sarepta presented evidence from the entire clinical program for SRP-9001 to back up clinical efficacy, including studies 103, 102 and 101 with an integrated analysis across the trials. Study 103 is an ongoing open-label phase 1, meaning patients knew they were receiving the gene therapy, and the NSAA score evaluation was an exploratory objective. Sarepta said that cohort 1 showed a mean increase in NSAA scores at week 52 and a statistically significant score increase at one year.

Study 101—an open-label, single-arm phase 1/2 test—similarly showed a score increase that was maintained at four years.

As for Study 102, Sarepta again notes the baseline NSAA imbalance and said the results are “challenging to interpret.”

“The treatment effect and overall pattern of NSAA total score from the three trials is consistent, durable and attributable to SRP-9001,” Sarepta explained.


Question 3

The FDA has asked the committee to consider the risks of SRP-9001 gene therapy treatment in ambulatory patients for the third question. It’s a lofty topic; the FDA is flagging the overall safety of the AAV gene therapy class, which has been associated with treatment emergent serious adverse events including liver toxicity, organ damage, anemia and even death. Some animal research has also shown the potential for cancer, although the formation of tumors in humans has not yet been demonstrated. The FDA says this class risk must be considered in the risk-benefit analysis for SRP-9001.

By the FDA’s count, “85 subjects had a total of 1230 treatment-emergent adverse events,” but there were no deaths observed in the studies. Eleven patients had serious adverse events, with the most commonly reported issues being vomiting, nausea, acute liver injury, fever and low platelets.

Sarepta noted the class effect and said liver injury has been observed with SRP-9001 treatment between weeks four and eight but these issues have resolved without complications. The company said these events are predictable.

Noting the rarity of DMD, Sarepta said that the three studies represented “182.75 patient-years of exposure” and prove that “SRP-9001 was generally safe and well tolerated.”

SRP-9001 has been placed on two clinical holds over the course of development, which began in October 2017. The first time, the FDA paused the program in June 2018 over the risk that human patients “were or could have been exposed to an unreasonable and significant risk of illness or injury,” and the application for human testing did not have enough risk assessment information. The hold was removed three months later.

The second hold occurred in August 2021 when the FDA again flagged the lack of risk assessment information. At the same time, a serious adverse event was reported in a 9-year-old patient who required hospitalization and respiratory support after being dosed with SRP-9001.


Question 4

Last week, Sarepta executives previewed the questions they’d likely face at the meeting during a first-quarter earnings call. CEO Doug Ingram predicted the clinical benefit discussion but also suggested that the FDA might have questions about whether Sarepta can pull off a confirmatory study.

The FDA is indeed asking the committee to review the confirmatory study, which Sarepta is doing through Part 1 of the phase 3 Study 301, which is called EMBARK. But the agency asked more specifically for the committee to consider the impact that early market access might have on the completion of that trial.

Topline results are expected in the fourth quarter of this year. The study is fully enrolled with 126 patients aged 4 to 7 years and is testing SRP-9001 against a placebo in a 52-week, randomized, double-blind format. There’s also a crossover portion called Part 2 that is already underway.

Sarepta says that the risk of patients dropping out once SRP-9001 reaches the market is “extremely low” because there are 45 patients in the trial outside the U.S., which will be the only place the gene therapy is available.

Mizuho Securities analyst Uy Ear said the briefing documents "may make it difficult for a positive" vote at the advisory committee meeting. The FDA does not have to follow the advisory committee's vote but uses the discussion to inform its decision. 

"Overall, we believe the FDA briefing document reads very unfavorably, with the FDA essentially pointing out that the currently available data does not support accelerated approval," Ear wrote in a Wednesday note to clients. 

But investors apparently see the single voting question as a positive sign, since the FDA did not ask the committee to consider whether approval should wait until EMBARK is finished. Investors still think the FDA may approve the medicine by the May 29 date.