DUBLIN – Shares in DBV Technologies Inc. plunged on news that its Viaskin peanut allergy therapy failed to reach statistical significance in a pivotal phase III trial in children. The most immediate beneficiary was Aimmune Therapeutics Inc., of Brisbane, Calif., whose stock rose by more than one-third Monday, with investor attention switching to its AR-101 oral desensitization therapy, which is currently undergoing a phase III trial in children and adults. A read-out is expected in the first quarter of next year.
Paris-based DBV did see an efficacy signal, and, on the advice of the FDA, plans to maintain its present trajectory. "The initial feedback from the FDA is to continue with our BLA planning process," Susanna Mesa, DBV's senior vice president for strategy told a conference call audience late Friday. "At this stage we do not intend to change any of our clinical trial programs as they are today."
Many investors voted with their feet, however, exiting the stock in large volumes during after-hours trading. Trading in the company's shares was temporarily halted Monday on the Euronext exchange in Paris.
The Pepites trial randomized 356 participants, between 4 and 11 years, in a 2:1 ratio to receive either a daily 250-mcg dose of peanut protein over 12 months, delivered through an epicutaneous patch, or placebo. The primary endpoint was the percentage of treatment responders, defined as those with post-treatment eliciting dose of 1,000 mg or greater in a double-blind food challenge when the baseline eliciting dose was over 10 mg or a post-treatment eliciting dose of 300 mg or more when the when the baseline eliciting dose was ≤10 mg.
Over one-third – 35.3 percent – of those in the treatment arm responded to therapy, whereas less than one-sixth – 13.6 percent – of those in the control arm did. "The lower bound of the [95 percent] confidence interval did not reach the 15 percent criterion that was in the finalized statistical analysis plan that was submitted to the FDA," DBV's chief scientific officer, Hugh Sampson, said on the call. "This confidence interval requirement is quite stringent for a trial of Pepite's size."
The company is hopeful that the data from Pepites and earlier studies could help its case, particularly as it has a good safety and tolerability profile, reflected in a high rate of compliance and a low rate of dropout from the study.
"I think it's important to emphasize that the totality of topline data suggests a significant treatment effect as measured by the response rate that is observed with Viaskin peanut allergy in this population," Sampson said.
Whether the FDA shares that view is an open question for now. "We believe there may be some room for flexibility and negotiation," Mesa said. In DBV's favor, the unmet medical need is substantial – and it's growing. Sampson said peanut allergy in children trebled between 1997 and 2008 in the U.S. and as many as 2 percent are now allergic. The company is not alone in believing that there are still some grounds for optimism. "We have reasons to believe that the FDA may entertain filing," noted Leerink analyst Dae Gon Ha. The study did reach statistical significance on secondary endpoints such as the percentage of treatment responders and on the change from baseline of the cumulative reactive dose of peanut protein. Leerink has trimmed the therapy's probability of success from 75 percent to 60 percent.
There are, moreover, some regulatory precedents that could also help DBV's case. The hay fever therapies Oralair and Grastek, marketed by London-based Stallergenes Greer plc and Copenhagen, Denmark-based ALK-Abelló A/S, respectively, and the house dust mite allergy therapy Odactra, which is also marketed by ALK, all gained FDA approval despite attaining less than 10 percent on the lower bound of the 95 percent confidence interval in their respective pivotal trials, Ha noted. "In particular, the same division that oversaw these products – Centers for Biologics Evaluation and Research (CBER) – is now adjudicating Viaskin peanut," he added.
Charles Duncan, senior research analyst at Piper Jaffray & Co. noted that the statistical analysis plan for Aimmune's ongoing phase III Palisade trial in peanut allergy also probably incorporates a 15 percent lower bound on the 95 percent confidence interval (CI). But it's not a concern. "We are confident that the Palisade program has been adequately, even conservatively, powered based on the phase II dataset and drop-out rate. Therefore we do not see this miss by DBV on the lower CI in a large peanut allergy phase III as increasing risk to the Palisade read-out," he wrote in an information note.
Shares in DBV (NASDAQ:DBVT) closed Monday at $28.32, down 41 percent, while Aimmune's stock (NASDAQ:AIMT) closed at $32.94, up 28.4 percent.