The FDA raised significant—but expected—concerns about Alkermes’ major depressive disorder drug ALKS-5461 in briefing documents released (PDF) ahead of an advisory committee meeting Thursday. The report highlighted issues with Alkermes’ data and the study design it used, as well as with the drug itself, which contains an opioid: buprenorphine.
"Our first assessment of the documents suggests that ALKS has a significant battle ahead of it," wrote Jefferies analysts on Tuesday. The news sent Alkermes’ stock down 6.5% in premarket trading that morning, but it has since rebounded.
ALKS-5461 is a once-daily pill combining buprenorphine and samidorphan, a new molecular entity designed to ward off the negative effects of the opioid—such as addiction—by blocking the mu-opioid receptor in the brain.
"The original design intention [of ALKS-5461] was to minimize the abuse liability, the respiratory suppression, the euphoria and the addictive liability of buprenorphine, a drug with known antidepressive properties. The basic idea is, can we take a known antidepressant—buprenorphine—and can we target it in a very specific way at the molecular level to address what we see as its principal liabilities?" said Alkermes CEO Richard Pops.
While Alkermes thinks it has "a whole lot of data directly addressing" these questions, FDA staff were unconvinced: "Although the Applicant has made several arguments that the other component of their drug, samidorphan, negates the μ-opioid properties of buprenorphine, this has not been conclusively proven."
Abuse potential aside, FDA staff took issue with Alkermes’ data. In April, the FDA issued a refuse-to-file letter telling the company it needed to see data from "additional well-controlled clinical trials" before it would review a filing for ALKS-5461. Two weeks later, the agency rescinded the letter after having "productive interactions" with Alkermes in which the company "clarified certain aspects of the NDA submission."
That said, the data in Alkermes’ submission is mixed and comes from a phase 2 study—which the FDA will not count as a pivotal study—as well as a trio of phase 3 studies, known as FORWARD 3, 4 and 5, that use a new, two-stage trial design that the FDA has not seen before.
Alkermes used the design, known as sequential parallel comparison design (SPCD), to minimize placebo response in the trials, Pops said.
"Based on FDA’s own data, drugs with known efficacy in the treatment of depression fail to separate from placebo 50% of the time," Pops said.
"That’s not because the drug’s not active, it’s because it’s incredibly difficult to test the efficacy of drugs because the placebo response is so high. Part of that has to do with the self-selection process that goes into enrolling in clinical trials for depression. It tends to be a motivated group of patients who have high placebo responses," he said.
While the SPCD design was designed to overcome this problem, the FDA took issue with a change in the primary endpoint and duration of the second stage of FORWARD-5. Of the three phase 3 studies, only FORWARD-5 succeeded. Unlike its predecessors, FORWARD-3 and FORWARD-4, this trial averaged out Åsberg Depression Rating Scale scores over multiple weeks, which Alkermes believes is a more accurate way of showing how well a drug is working over time. But there's no guarantee the FDA will see the data that way.
"We recommend using a single efficacy measure at the end of the study, and not an average over multiple time periods, as the primary efficacy endpoint," FDA staff said in the briefing documents.
ALKS-5461 represents hope for a space in which investment and innovation have lagged. If approved, it would be the first drug of its class and a viable option for patients for whom traditional depression drugs do not work. Only 29 active substances have been FDA-approved for major depression since 1959, and all of them work the same way—by modulating the brain’s monoamine neurotransmitter levels, including serotonin levels.
If all goes well at Thursday’s advisory committee meeting, the FDA will have until Jan. 31 to make its decision on ALKS-5461. With a required DEA scheduling period ...
(All Rights Reserved)