Sarepta Therapeutics Inc.’s miss on a key phase II ambulatory endpoint in its Duchenne muscular dystrophy (DMD) trial may have been caused by a dramatic disparity in functional ability at baseline among older vs. younger patients afflicted with the progressively worsening disorder.

In any case, Wall Street had knives out, carving 51% of the value away from shares (NASDAQ:SRPT) of the Cambridge, Mass.-based firm, which closed at $82.29, a loss of $86.66, or 51%.

Sarepta made public data from the first part of Study 102, testing the safety, efficacy and tolerability of a single dose of SRP-9001 (rAAVrh74.MHCK7.micro-dystrophin) in 41 patients with DMD. A gene transfer therapy intended to deliver its microdystrophin-encoding gene to muscle tissue for the targeted production of the protein, SRP-9001 met the primary biological endpoint of microdystrophin protein expression at 12 weeks post-treatment, as measured by western blot, in DMD participants vs. placebo. Patients on active drug showed an increase in North Star Ambulatory Assessment (NSAA) total score compared to placebo at 48 weeks, but results of the phase II experiment fell short of statistical significance on the primary functional endpoint of improvement in the NSAA total score vs. placebo after 48 weeks.

Details may paint a more hopeful, or at least explanatory, picture. Although the placebo-adjusted NSAA benefit at 48 weeks missed statistical significance (p=0.37), at every point after four weeks until the 48-week mark, patients on the drug did better than placebo. At 48 weeks, the drug arm of 20 patients turned up a betterment of 1.7 on the NSAA vs. baseline (p=0.009); placebo patients, numbering 21, gained an improvement of 0.9 points vs. baseline (p=0.14). But patients were stratified by age, and those in the 6-to-7 group started off much differently in terms of NSAA scores between the active drug and placebo subjects. Chief Scientific Officer Louise Rodino-Klapac said the mismatch made it “virtually impossible” to see a treatment win in the group – a factor that may have skewed the overall results.

In the prespecified analysis of 16 participants ages 4 to 5 at the time of treatment, the drug group showed a statistically significant 4.3-point improvement on NSAA total score at 48 weeks after treatment compared to a 1.9-point improvement in the age-matched placebo group (p=0.0172). The functional status at baseline for participants in the 4-to-5 age group was balanced across the placebo and treatment cohorts. A statistically significant imbalance (p=0.0046) in baseline NSAA total score, though, was evident in the 25-subject cohort of patients ages 6 to 7 and older, which put milder DMD participants in the placebo arm of 13 patients than in the drug-treated arm of 12.

No new safety signals

CEO Douglas Ingram said Sarepta has “had no additional discussions with any of the regulators about this. As it stands right now, we are going to focus on the completion of Study 102. Remember, this is, in fact, an interim analysis,” and the study remains blinded. The final readout of the trial will take place when all patients, including crossover subjects, have gone the full 48 weeks. “We'll do trajectory analysis and a number of other analytics that are prespecified there,” he said. “We're also focusing on Study 103, and getting the expression-level information out of [that phase I experiment] to see the performance of our commercial material. That will be coming up in the not-too-distant future, which will tell us a lot about the therapy.” Study 103 has enrolled and dosed 11 participants so far. Biomarker and safety results are due in the second quarter.

Meanwhile, what’s known of Study 102 was put through the wringer. During a conference call with investors hosted by Sarepta, Cantor Fitzgerald analyst Alethia Young cited ups and downs in apparent efficacy across the time period covered by the trial. “In between 12 and 24 weeks, it seemed like in the placebo, there was a huge jump up with wide error bars,” she said, and speculated that a better outcome might have been perceptible at the 12-week juncture. Ingram noted that “on that 24-week time point, you'll see the strange increase or a momentary increase for the placebo group. We can't say a ton about it yet, because we haven't done all of the analysis,” but the research was taking place “in the middle of the most difficult period of the pandemic.”

In the plus column, Study 102’s findings back the favorable safety and tolerability profile of SRP-9001, with no new signals identified. In line with previously reported data, no clinical complement activation was observed. Eighty-five percent of participants in the treatment group experienced at least one treatment-related adverse event (AE) compared to 43% in the placebo group. Among participants with treatment-related AEs, 82% were mild or moderate in severity, and four subjects ended up with serious treatment-related AEs, including three participants in the treatment group (two cases of rhabdomyolysis, two transaminase elevations) and one in placebo (rhabdomyolysis).

Dyne in line, too, with ASOs

What does it all mean, at least as of today? Opinions vary. SVB Leerink’s Joseph Schwartz wasn’t ready to blame the baseline disparity among the 6- to 7-year-olds. “We do agree that the imbalance was striking,” he wrote in a report. “However, the company failed to provide any additional analysis at this time to substantiate if the imbalance was indeed causative and not simply correlative.” Cowen’s Ritu Baral sounded more open to the causative idea, saying in her report that the lack of NSAA symmetry “likely” is the culprit. “We think this study had a high chance of success had both trial arms been adequately balanced at baseline in terms of functional measure and age,” she wrote.

Mizuho’s Difei Yang kept the faith. “Overall, we remain confident that SRP-9001 gene therapy works, based on all the data presented by Sarepta to date, and we actually gained confidence” in the drug, thanks to the results in 4- to 5-year-old subjects. “We believe timelines are likely to shift in terms of registration pathway, and we updated our model to reflect a later launch in 2024 (vs. 2022 previously),” she said in a report, dropping her price target to $158 from $168.95. Yang held on to her 70% probability of success estimate.

DMD typically leaves affected boys wheelchair-bound by their midteens, and life expectancy is 26 years, with death often arriving in the form of cardiac and pulmonary trouble. Among the high-profile contenders in the disease is New York-based Pfizer Inc., which recently disclosed its phase III trial called Ciffreo with gene therapy candidate PF-06939926. Expected to enroll 99 ambulatory patients, ages 4 through 7 across 55 sites in 15 countries, Ciffreo began dosing in Barcelona, Spain, on Dec. 29.

Baird’s Brian Skorney said he believes the Sarepta compound owns a safety advantage over Pfizer’s, though he allowed in a report that “our call on Study 102 was as wrong as could be, with results missing in a very big way.” He bought the imbalance thesis but conceded “an elongated timeline and less certain benefit make the competitive dynamic with Pfizer more concerning.” Ultimately, SRP-9001 will probably succeed, he said.

Lesser known and earlier-stage but intriguing in the DMD space is Dyne Therapeutics Inc., of Waltham, Mass., with its Force platform, designed to deliver antisense oligonucleotides (ASOs) directly to muscle tissue as a way to get around the limitations of non-targeted, RNA-based approaches. Dyne is conjugating an ASO via linker to an antibody fragment targeting transferrin receptor 1 (TfR1), highly expressed on muscle cells. The company plans INDs in DMD, myotonic dystrophy type I and facioscapulohumeral muscular dystrophy between the fourth quarter of this year and the fourth quarter of next. Sarepta’s Exondys 51 (eteplirsen), approved for DMD in September 2016, and Vyondys 53 (golodirsen), approved for the indication in December 2019, have proved the value of skipping specific dystrophin exons that harbor mutations. FORCE-DMD can selectively deliver exon skipping phosphorodiamidate morpholino (PMO) oligomers using the same TfR1 antigen-binding fragment that has gained about eight times the dystrophin expression as compared with naked PMO in animal models. Among Dyne’s fans is Jefferies analyst Eun Yang, who started coverage in October 2020 with a buy rating and $47 price target. Shares (NASDAQ:DYN) closed at $27.13, up $3.91, or 16.8%. The stock bears a 52-week high of $27.78. Founded with seed money by Atlas Venture in 2018, Dyne the following year completed a $50 million series A round with help from Forbion and MPM Capital.