Sixteen years ago, Kim Witczak’s husband died by his own hand, turning her world upside down. He had just begun taking an antidepressant off-label for insomnia, and she believes an undisclosed side effect of the drug drove him to suicide. Compelled to act, she became an advocate for tougher safety standards. Witczak now sits on the FDA advisory committee that debates the merits of experimental psychiatric drugs that companies bring to the agency for approval.
There are plenty of psychiatric drugs on the market, many of them—Prozac, Zoloft, Lexapro—as brand-recognizable as soft drinks and cars. Finding the right one for the right person often requires guesswork—if they’re matched up at all.
Meanwhile, depression has become epidemic in the US, the leading cause of disability for people between 18 and 44, and suicide rates are on the rise.
“We don’t need more drugs, we need better drugs,” says Witczak.
Last November, more was not better for Witczak and her colleagues when a new antidepressant came before the committee. They voted 21 to 2 to recommend that the FDA reject it. Three months later, the FDA took the recommendation to heart.
There were several reasons for the failure of that medicine, known as ALKS-5461, including its composition. It contains a powerful opioid, buprenorphine, at a time when opioids are driving an all-time high in drug overdose deaths.
But another problem, less emotional and more technical, tripped up the drug’s owner. The Waltham, MA drug firm Alkermes (NYSE: [[ticker:ALKS]]) used an unusual clinical trial design meant to combat the placebo effect, which many in the psychiatric field believe is standing between better drugs and patients who need them. The FDA’s rejection cast doubt upon the design, still in use in more than a dozen trials, as a weapon against the placebo effect.
That’s big news because the effect, also called the placebo response, has been growing stronger over the years in clinical studies that randomly assign patients to either an active drug or placebo. When the effect is high, it’s hard to know if a drug just isn’t good enough, if there are errors in the data, or if the participants taking the placebo—an inert pill meant to make them believe they’re getting the real thing—fared unusually well because of their expectations.
It’s a testament to the power of our minds to improve our health, at least temporarily. Many factors boost placebo response. “Most people, whether they know it or not, are biased to believe that they will receive the active drug even if they are told that they have a 50 percent chance of getting placebo, and this ‘therapeutic bias’ increases the placebo response,” says John Krystal, the chair of the psychiatric department at the Yale School of Medicine in New Haven, CT.
“A Cottage Industry”
If a psychiatric drug doesn’t beat a placebo in at least one randomized study, it’s not likely to be approved. Problem is, a drug can outperform placebo in one trial but not in another. That’s why drug companies developing psychiatric drugs will often run at least two big trials in parallel like a chef cooking backup banquets in separate kitchens because there’s a good chance the first kitchen will go up in flames.
Even after spending the extra money on larger and additional trials, companies often end up asking the FDA to approve a drug with mixed results. Alkermes did it with ALKS-5461. New York-based Intra-Cellular Therapeutics (NASDAQ: [[ticker:ITCI]]) is doing the same in schizophrenia with its drug lumateperone. The FDA’s decision is due in late December. Intra-Cellular might also try for approval in bipolar disorder, where the firm ran parallel Phase 3 trials. One met its goals. In the other, lumateperone performed well but did not meet its goal because of high placebo response.
(Before the bipolar results came out, Intra-Cellular said earlier this year it was using a number of strategies to reduce placebo response. When asked to describe those strategies, chief medical officer Andrew Satlin told Xconomy via a spokeswoman that independent interviewers conducted “adjudication of diagnosis and severity of illness… to ensure study eligibility.”)
Placebo response takes a toll. In depression, even drugs that are eventually approved do not hit their goals in about half their trials, according to the FDA.
“Rates of depression are going up with every new generation,” says Maurizio Fava, executive vice chair of psychiatry at Massachusetts General Hospital in Boston and one of the world’s top depression researchers. “We’re allowing effective treatments to die.”
Fava says he has a solution: the clinical trial design that Alkermes employed in its ill-fated depression program. It’s called SPCD, sequential parallel comparison design, and it’s one of many ideas, ranging from artificial intelligence to complicated
