FDA Panel Fails to Support Novel Opioid for Depression

Pauline Anderson


November 02, 2018

Members of a US Food and Drug Administration (FDA) advisory committee agreed overwhelmingly that the company behind a proposed new combination opioid product for treating major depression did not provide sufficient efficacy data.

Meeting delegates were concerned about data inconsistencies, use of a questionable study design, and a switch that was made to an outcome measure, among other issues.

Most members thought the safety data that were presented for the product were adequate.

Members of the joint Psychopharmacologic Drugs Advisory Committee and Drug Safety and the Risk Management Advisory Committee met to discuss the efficacy and safety, as well as the risk-benefit profile, of a new drug application (NDA) for combination buprenorphine (BUP) and samidorphan (SAM) sublingual tablets, submitted by Alkermes, Inc.

BUP is a partial agonist at the μ-opioid receptor and an antagonist at the κ-opioid receptor. SAM, a new molecular entity, is a potent μ-opioid receptor antagonist intended to mitigate the abuse liability of BUP.

The combination product would treat major depressive disorder (MDD) in patients who have had an inadequate response to commonly prescribed antidepressants. Research shows that for 2 of every 3 patients, response to standard antidepressants is inadequate.

If approved, BUP/SAM would be the first drug in an entirely new class of MDD drugs. No other opioids have been formally evaluated for the treatment of depression.

The NDA filing for BUP/SAM is based on results from a clinical efficacy and safety package that included data from more than 30 clinical trials and more than 1500 patients with MDD.

Efficacy was evaluated in four randomized, double-blind, placebo-controlled studies: one phase 2 (202) and three phase 3 (205, 206, and 207) studies.

Efficacy Results

One of the four studies (206) had a placebo run-in design. The other three studies used a novel sequential parallel comparison design (SPCD).

This design is aimed at reducing the high placebo response rates that often plague studies of antidepressant drugs and to increase statistical power.

The SPCD design includes two stages. In the first, patients are randomly assigned to receive either drug or placebo. The randomization ratio is skewed so that far more patients are placed in the placebo group.

The second stage includes only patients from the placebo group in the first stage who did not respond to treatment. These patients are reassigned to either drug or placebo.

All studies included patients with MDD who had had an inadequate response to antidepressant therapy (ADT). Patients remained on background ADT.

The BUP/SAM product comes in doses with equal amounts of BUP and SAM. For example, the proposed 2/2 dose includes 2 mg of BUP and 2 mg of SAM.

The studies used various tools to evaluate BUP/SAM efficacy, including the Hamilton Rating Scale for Depression (HAM-D17) and the Montgomery-Åsberg Depression Rating Scale (MADRS). The MADRS is typically a 10-item diagnostic questionnaire used to measure the severity of depressive episodes in patients with mood disorders.

Two of the four studies had positive results.

The first, Study 202, randomly assigned 142 patients and compared both the 2/2 dose and the 8/8 dose with placebo. The primary endpoint was thhe score on the HAM-D17 at week 4 (a secondary endpoint was the score the MADRS-10).

This study showed that for the 2/2 dose, the difference with placebo in the HAM-D17 was significant at the end of treatment (EOT) (P = .014) The comparison was not significant for the 8/8 dose.

In study 202, a single patient responded with the 2/2 dose. That response was unusually strong. The applicant maintained that this patient did not drive efficacy results.

Four Studies

The second positive trial, study 207, randomly assigned 407 patients and compared doses of 2/2 and 1/1. Outcomes included average score of an abridged six-item MADRS (MADRS-6) (an average of values from each patient from week 3 to EOT), MADRS-10 average score (week 3 to EOT), and MADRS-10 score at EOT.

In study 207, there was a significant difference in MADRS-6 average score (P = .018 vs placebo for the 2/2 dose). For the MADR-10 average score for this dose vs placebo, P = .026.

According to the sponsor, the treatment effects in these studies were consistent with other FDA-approved adjunctive therapies for MDD.

The only such therapies are atypical antipsychotics, which are associated with serious and sometimes treatment-limiting toxicities. All approved antidepressants, including those approved for adjunctive therapy, work via monoaminergic mechanisms such as serotonin, norepinephrine, and dopamine.

A meta-analysis that combined data from all four studies (202, 205, 206, and 207) demonstrated statistically significant improvement vs placebo for both common endpoints (MADRS and HAM-D17), according to data presented by the sponsor.

The efficacy continued during a 1-year treatment period in a long-term, open-label safety study (208).

The FDA took issue with some of the studies presented by the sponsor.

Semhar Ogbagaber, PhD, statistician, Office of Translational Sciences, Center for Drug Evaluation and Research, said study 202 was not adequate or well designed. Positive results for the 2/2 dose depended on a single patient whose eligibility for the study could not be determined, he said. The FDA had recommended against including this patient in the trial.

Well Tolerated

As for study 207, the FDA did not agree with using the MADRS average score as a primary endpoint.

Studies for antidepressant therapies typically assess an efficacy endpoint at a specific time point following several weeks of treatment, partly because the treatment effect may require several weeks, and partly to provide evidence of durability of the treatment effect.

The FDA agreed that averaging scores from multiple weeks can decrease the variability in mood disorder symptoms that tend to fluctuate over time. However, this means that scores at the final time point carry less weight.

The FDA also took issue with the switch to use of an abridged six-item version of the MADRS diagnostic questionnaire for the primary endpoint in study 207.

The agency concluded that the MADRS-6 could not replace the MADRS-10 for use as a primary endpoint because it excludes items that are important in MDD, including "reduced sleep," "reduced appetite," "concentration difficulties," and "suicidal thoughts."

As for the meta-analysis, the FDA said it did not meet the usual standard of adequate and well-controlled trials for substantial evidence.

Ogbagaber also had concerns related to the SPCD. There are unresolved statistical questions regarding the most appropriate method for analyzing the results of an SPCD study, he said.

BUP/SAM was generally well tolerated, Gary Bloomgren, MD, vice president of drug safety and [harmacovigilance, Alkermes, Inc, told meeting delegates.

The most common adverse events (AEs) included nausea, constipation, dizziness, vomiting, somnolence, fatigue, and sedation.

Mild to Moderate AEs

AEs were typically mild or moderate in severity. They tended to occur with treatment initiation and did not lead to treatment discontinuation, said Bloomgren.

BUP/SAM treatment was not associated with metabolic disturbances or motor disorders, which are key concerns regarding the use of atypical antipsychotics as adjunctive treatments for MDD.

There was no evidence of AEs commonly associated with opioids, such as respiratory depression, hypotension/orthostatic hypotension, or hepatic injury, in the BUP/SAM group relative to placebo, said Bloomgren.

In addition, there was no evidence of hypomania/mania, sexual dysfunction, or suicidal ideation or behavior — conditions that are often associated with antidepressants, he said.

Given the abuse liability of BUP, the sponsor carried out a human abuse potential study, using the Clinical Opiate Withdrawal Scale. The product showed low abuse potential, and there was minimal evidence of dependence or opioid withdrawal, said Bloomgren.

A concern of the FDA is the effectiveness of SAM in affecting BUP. Although Alkermes had argued that SAM negates the μ-opioid properties of buprenorphine, the FDA maintained that this has not been conclusively proven.

During the open public hearing segment of the meeting, all but one of the 13 speakers supported approval of this new drug application. Many stressed the need for innovation and not more "me too" drugs.

Many speakers voiced the need for more choices to treat depression, which can be chronic and debilitating and for which a partial response to treatment is more the rule than the exception. Some noted the link between depression and suicide.

At the end of the day, most meeting delegates were not convinced that the sponsor had provided "substantial" evidence of efficacy. They voted 20 to 3 to this effect.

Many agreed with the FDA that the SPCD design has not been determined to be statistically acceptable.

Link Between Pain, Depression, Abuse

Delegates were concerned about the inconsistency of outcomes across studies. They were confused about the definition of "substantial" evidence or called for additional confirmatory evidence.

There was some debate about the statistical impact of excluding the outlier in study 202.

"That one outlier bothered me," said one delegate. "I don’t feel very comfortable with eliminating one subject and it changes the significance of the study."

However, many delegates felt that analyses using averages (instead of EOT) had clinical relevance.

Some attendees pointed out that few patients in the study were older than 65 years. This was a concern, because older patients often have comorbidities. Some expressed concern about the homogeneity of populations in the studies in general.

Another problem related to the inclusion of patients with pain, a population many referred to as "vulnerable" and as being overtreated with opioids. Some meeting attendees felt the sponsor did not adequately address safety issues in this population.

Guest speaker Mark Sullivan, MD, PhD, professor of psychiatry and behavioral sciences, University of Washington Medical Center, provided an overview of the complex relationship between depression, pain, and substance abuse.

Depression and pain are similar with respect to neuroscience and with respect to response to medications, including antidepressants, said Sullivan.

Opioids are a common therapy for depression, but depression increases the risk for misuse and abuse of prescription opioids, he said.

Some experts at the meeting saw the lack of a dose response (the higher 8/8 dose was not significant, whereas the 2/2 dose was) as a red flag

Those who agreed that the sponsor had provided sufficient evidence of efficacy had caveats. One delegate stressed the need for a restrictive risk evaluation and mitigation strategy. One suggested limiting the supply of the drug to a few months.

Safety Data OK

With regard to safety data, the vote affirming that the evidence was sufficient was relatively close (13 to 10).

For the most part, delegates praised the company for its efforts to produce an innovative product and were satisfied that it had done the appropriate tests for abuse.

But some experts expressed concern about possible accumulation of the active metabolite.

"Physical dependence and withdrawal were not well characterized," said Jane B. Acri, PhD, chief, Medication Discovery and Toxicology Branch, National Institute on Drug Abuse, National Institutes of Health.

Some delegates wondered how easy it would be to distinguish the BUP from the SAM in the "real world." At least one questioned how tamper proof the new formulation would be.

Others noted the lack of information on possible interactions between this new product and other drugs, including alcohol.

"This population is at high risk for drug dependence in first place, so I think it's a risky proposition to introduce an opiate in that type of population if there are other alternatives," said one delegate.

One voter — Felipe A. Jain, MD, assistant clinical professor of psychiatry, University of California, San Francisco — pointed out that patients with treatment-resistant depression do not undergo treatment for only 5 or 6 weeks but for much longer, perhaps for life. So safety needs to be adequately studied over an extended period.

The sponsor promised continuous postmarketing monitoring, physician and patient education, and various safety checks, should the product be approved.

Source: webcast of FDA advisory committee meeting, November 1, 2018; background materials from the FDA and the sponsor.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.