DUBLIN – DBV Technologies SA hopes to move its Viaskin Milk (VM) desensitization therapy for IgE-mediated cow's milk protein allergy (CMPA) into phase III trials next year, on the back of phase I/II efficacy data it unveiled Sunday at the European Academy of Allergy and Clinical Immunology (EACCI) congress in Munich. That depends on FDA agreement, however, and the Paris-based firm plans to meet with the regulator to discuss its next steps later this year. The treatment missed the primary endpoint of the trial but at the same time showed definite efficacy trends.
The dose-finding study (n=198) compared three doses (150 mcg, 300 mcg and 500 mcg) of VM with placebo over a 12-month treatment period in children and adolescents, ages 2 to 17. The main efficacy endpoint was a minimum 10-fold change from baseline in the amount of cow's milk protein participants could receive without evidence of an allergic response, combined with the absolute amount of protein (cumulative reactive dose, or CRD) reaching a minimum of 1,444 mg. Responders could also be defined on the basis of a CRD greater than or equal to 1,444 mg of cow's milk protein at the end of the 12-month treatment period. Recruits were required to have an objective reaction to 300 mg or less of cow's milk protein at baseline.
The middle dose of 300 mcg elicited the best response rate, with 49 percent in that cohort (n=49) attaining a response vs. 30.2 percent for those in the placebo group (n=53). The result was not statistically significant, however (p=0.085). Dealing with a high placebo response rate "was the real challenge," David Schilansky, deputy CEO of DBV, told BioWorld.
The condition, although life-threatening, can resolve spontaneously. Most of those affected grow out of it after turning 5. The inherent variability of the condition also makes it difficult to detect an efficacy signal.
The per-protocol analysis was more forgiving. It yielded a placebo response rate of 28.9 percent (n=45) vs. 55 percent for the 300-mcg dose group (n=40). That result did cross the significance threshold (p=0.027).
Among younger participants, ages 2 to 11, the signal was stronger again. Those in the active treatment group (n=38) showed a response rate of 57.9 percent vs. 32.5 percent for the control group (n=40), according to the intent-to-treat analysis. The result was statistically significant (p=0.042). In the per-protocol analysis, the equivalent figures were 62.5 percent for the active treatment group (n=32) vs. 31.4 percent for the placebo group (n=35). That, too, was statistically significant (p=0.021).
The company also reported that participants in the 2 to11 age group exhibited an improvement with respect to placebo in the cumulative amount of cow's milk protein they could tolerate. In the 300-mcg group, that measure increased to a mean of 1,340.3 mg vs. 510.8 mg for placebo. In the intent-to-treat analysis, it increased to a mean of 1,322.4 mg vs. 565.6 mg for placebo.
The company has not divulged the equivalent figures for the entire study population.
So although the trial missed its primary endpoint, DBV has generated evidence that the treatment, which comprises an epicutaneous patch, has some level of efficacy. A big question to explore in its FDA discussions is to consider a phase III study in younger patients only, given their apparently higher levels of response to the therapy.
Last year, DBV produced another borderline result – for its lead product candidate, Viaskin Peanut, in development for peanut allergy. That result, DBV has long maintained, was due to an unnecessarily stringent statistical hurdle it adopted itself which it was unable to clear – and which is not clinically relevant. "It's very unfortunate," Schilansky said. (See BioWorld, Oct., 24 2017.)
It received a sympathetic hearing from the FDA, which earlier this year encouraged DBV to proceed with a BLA filing with the data in hand. "We had a positive pre-BLA meeting," Schilansky said. The company plans to file before the end of the year.
Although the company's stock has failed to recover the ground it lost on the phase III trial miss, DBV is still able to tap into supportive investors. It grossed $172.5 million in a share placing in March and is currently valued at about $1.4 billion, which reflects the scale of the markets it is addressing. Peanut allergy affects between 2 percent and 5 percent of children in the U.S., according to a range of estimates.
DBV has direct competition in peanut allergy desensitization therapy. Brisbane, Calif.-based Aimmune Therapeutics Inc. is also on track for a BLA filing later this year, for AR-101, which met the primary and secondary endpoints of a phase III trial earlier this year. (See BioWorld, Feb. 21, 2018.)
CMPA is also a large indication, even if it is not as prevalent as lactose intolerance, a difficult-to-diagnose condition that primarily causes gastrointestinal symptoms. CMPA is more serious and more obvious, as even a small exposure to milk can cause severe swelling, rash and, in a minority of cases, anaphylaxis. It is difficult to manage through dietary control.
"It's potentially even worse than peanut allergy, because avoiding accidental exposure is almost impossible," Schilansky said. The aim of therapy is not curative. "It's really about lowering the risk of accidental exposure."