A new treatment for Alzheimer’s disease agitation episodes might be on the horizon, but a principal trial investigator skirting protocol could muddy the waters.
As BioXcel unveiled its phase 3 data for its orally dissolving dexmedetomidine (originally Pfizer’s Precedex) formulation in Alzheimer’s-related agitation, the company also disclosed some serious trial missteps in an Securities and Exchange Commission (SEC) filing.
After the FDA in December inspected a trial site that enrolled about 40% of study participants, the agency found three big red flags relating to adherence of the trial’s framework, the filing says. For one, the investigator failed to follow the informed consent plan for four trial subjects, Chief Medical Officer Robert Risinger, M.D., told investors on a conference call.
In “certain instances” the investigative plan wasn’t followed and sufficient case histories weren’t maintained for some patients. For example, the site reported a serious adverse event for one patient outside of the standard 24-hour time period.
What’s worse, BioXcel alleges that the principal investigator fabricated email correspondence about another serious adverse event from the placebo arm to make it seem like it had reported it in the appropriate timely manner and received confirmation of receipt.
The company got wind of that in May and “promptly initiated” an investigation, recently confirming the cover-up, it said in the filing. According to BioXcel, the serious adverse event was timely entered into the data system despite having not been reported to the correct authority, the pharmacovigilance safety vendor, within the required 24 hours.
In its investigation, which is still ongoing, the company determined that the fake emails were provided to the FDA during its inspection. It’s still looking into data integrity and protocol adherence at the site and is working on securing an independent third party to audit the site’s data, which “may uncover new findings” on the accuracy of the trial findings and if it can be used for marketing authorization, the filing noted.
On top of the possible data inaccuracies, the phase 3 trial revealed a missed secondary endpoint.
The study found a 7.5 reduction from baseline on the Positive and Negative Syndrome Scale-Excitatory Component (PEC) score at two hours with a 60-mg dose, compared to placebo's 5.4. The drug also reduced agitation symptoms at one hour during the first agitation episode but didn’t change from baseline PEC score at 30 minutes, which was a key secondary endpoint.
Meanwhile, the 40-mg dose didn’t meet the primary endpoints at all. However, the 60 mg maintained a PEC reduction through repeated dosing. All 443 episodes treated with the dose had a PEC reduction from pre-dose versus placebo at 1 and 2 hours.
While the filing emphasizes that a potential approval may be impacted by the situation and a completely new trial might be necessary, Risinger “can’t see” a scenario in which the entire data set is unusable. An analysis excluding every patient from the site in question showed “a similar effect, at least a directional effect” of its positive efficacy ratings.
Still, investors were not happy, with shares falling more than 60% by late morning Thursday on the news.
The treatment has been granted breakthrough designation and the company is working closely with the FDA on “every aspect of this trial, down to the nitty-gritty details,” Risinger said. BioXcel noted in its release that it plans to “develop a path” to potential filing during the second half of this year.