FDA panel votes against sabizabulin EUA for COVID-19 at risk for ARDS

The FDA’s Pulmonary-Allergy Drug Advisory Committee voted 8-5, leaning against granting sabizabulin emergency use authorization to treat patients hospitalized with COVID-19 and at risk for acute respiratory distress syndrome.

Following this split decision, the panelists provided recommendations for an additional study that further evaluates the efficacy of the drug.

Supporting data

Sabizabulin (VERU-111, Veru Inc.) has been submitted to the FDA for emergency use authorization for the treatment of adults hospitalized with moderate to severe COVID-19 and at high risk for ARDS.

According to a presentation by Robert Busch, MD, MMSc, medical officer and FDA participant, this drug is a new molecular entity and has yet to be approved in the U.S. and worldwide.

The panel reviewed data from the international, multicenter, randomized, double-blind, placebo-controlled, parallel group study V3011902, which was submitted by Veru Inc. The study included 204 patients aged 18 years and older, of which 134 were given 9 mg of VERU-111 for 21 days at most and 70 were given placebo.

In order to be included in the study, patients had to fit in category 5 or 6 of the WHO ordinal scale. Patients in category 4 with at least one of the following designated comorbidities could also be included: had asthma, chronic lung disease, diabetes, hypertension or severe obesity (BMI 40); were aged 65 or older; primarily resided in a nursing home or long-term care facility; or were immunocompromised.

Overall, the study met its primary endpoint of all-cause mortality after 60 days, showing that 78.4% of patients receiving VERU-111 were alive at day 60 vs. only 58.6% of patients receiving placebo (risk difference = 19%; 95% CI, 5.8%-32.2%; OR = 2.77; 95% CI, 1.37-5.58).

However, in assessing the data, the FDA found several uncertainties that when considered together, could lead to questions about the results in such a small sample of patients.

These uncertainties, presented by the FDA, included: high placebo mortality for baseline severity, potential unblinding events with enteral tube administration, application of standard of care therapies, timing of enrollment compared with COVID-19 clinical course, effects of goals of care on all-cause mortality, efficacy results of other microtubule disruptors in COVID-19 and study population uncertainties.

Pre-vote discussion

In evaluating the all-cause mortality results, many panelists said the data from the study were impressive, but they questioned the small sample size and uncertainties that were found, specifically the potential unblinding.

“I do believe that some issues with the blinding could invoke questions in my mind about this [mortality rate], and I think given the small trial size, there could be easy shifts in these things depending upon what the baseline severity of disease might have been or any potential differences could have come through in the patient population,” Daniel L. Gillen, PhD, chancellor’s professor and chair in the department of statistics at University of California, Irvine, who ultimately voted no, said during the pre-vote discussion. “Because of that, my enthusiasm is certainly tempered, though again, on face value the point estimate itself is impressive here, but it does not rise to the level of what we mostly considered to be the standards that we would look for on something like this.”

Jennifer A. Schwartzott, MS, a patient representative who ultimately voted yes, questioned the true efficacy results and wondered if pre-standard of care, VERU-111 or a mix of care was the reason patients lived in this study.

“I am leaning towards suggesting another study because it puts the drug out there under strict conditions while collecting data for potential future use,” Schwartzott said. “I simply don’t feel we have enough data but feel that this has enough promise to deserve the future study.”

David H. Au, MD, MS, chairperson of the panel, professor of medicine at the University of Washington and director of the Center for Innovation for Veteran-Centered and Veteran-Driven Care in the VA Puget Sound Health Care System, summarized the panel’s discussion, pointing out that the panel showed concern over stability of the point estimates and size of the trial.

“We recognized as the discussion has been that the overall number of patients that would need to shift are not that large,” Au said. “Even if you use the largest number that was presented, it would still represent about 10% of the population. Overall, the number needed to change the outcome of the trial is relatively small, and it speaks to the underlying instability in the estimate.”

Voting rationale

When asked if the known and potential benefits of VERU-111 outweighed the known and potential risks in the specific patient population studied, five panelists voted yes and eight panelists voted no.

Several of the panelists who voted yes said their rationale for their vote was based on the language of the voting question.

“It was my impression that there were neither clearly known benefits nor clearly known harm or risks, but the potential benefits based upon the data that are available outweighed the potential risks based on the data that are available in the context of this patient population that is hospitalized for severe and critical illness under monitoring for a drug that’s going to be administered for a short interim,” Capt. Daniel S. Chertow, MD, MPH, FCCM, FIDSA, tenure-track investigator and head of the emerging pathogens section in the critical care medicine department at the NIH Clinical Center and in the laboratory of immunoregulation at the National Institute of Allergy and Infectious Diseases, said.

Edwin H. Kim, MD, MS, associate professor in the division of pediatric allergy and immunology at the University of North Carolina School of Medicine, who also voted yes, added, “I think the benefit of protecting against mortality — although maybe the magnitude is not as big because of some of these uncertainties — I think there is likely benefit to be had. The risks of a short course, 21-day in-hospital treatment, I think are going to be manageable, so the benefits do outweigh the risks, in my opinion.”

Panelists who voted no said they had concerns over the efficacy, small sample size and the uncertainties.

“I believe that we have a limited both efficacy and safety data set with a new molecule where we don’t have a full understanding yet of the mechanism of action,” Gillen said. “I don’t know if we’re ever guaranteed to know that completely, but certainly we’re far from it at this point.”

Pamela Shaw, PhD, senior investigator in the biostatistics division at Kaiser Permanente Washington Health Research, said she could not accurately judge the potential benefit in the target population with the data presented. She added that an additional trial with a sample close to 500 patients would be beneficial.

“I certainly agree that there is an impressive point estimate for the effect, but as has been said by a few colleagues already, it’s hard to know whether to believe that effect because of potential anomalies observed in the placebo mortality, some peculiarities of the viral burden patterns or potential imbalance factors between the groups in the setting of a small size, where only a small number of outcomes would need to be changed to influence results,” Scott E. Evans, MD, FCCP, ATSF, chairman ad interim and professor in the department of pulmonary medicine and Rebecca Meyer Brown and Joseph Mellinger Brown Chair in Basic Science Research at The University of Texas MD Anderson Cancer Center, said.

Steven D. Shapiro, MD, senior vice president for health affairs at the University of Southern California, who also voted no, pointed out the toll that COVID-19 has taken on the population and how this could be utilized in a future study.

“There are hundreds dying a day,” Shapiro said. “If they’re dying of ARDS and lower lung disease, it should be pretty easily achievable to get a larger study and improve it.”

With this split decision, panelists suggested a second study, and Au summarized their recommendations.

“There was a number of points around uncertainty about mechanism of action, and because there was this concern about mechanism of action, it led to questions around the dosing, the timing, the administration and the population,” Au said. “When considering the study design and the future execution of studies, that may be something that actually has more mechanistic orientation that includes biomarkers and indicators of the host immune response to help an understanding of the outcomes data to make sure that the data is consistent, which has been one of the of common themes throughout this discussion. Do we have internal validity, and do all the data converge on the same answer?

“There were also discussions to ensure that there’s proper heterogeneity of treatment effects particularly around demographic populations,” Au added. “The virus and its effect on mortality has continued to change over time, and that using criteria that may have provided some degree of homogeneity or estimated population effects may not necessarily be valid today. Thinking about whether or not the WHO classification, which is an ordinal scale, whether or not that’s really the correct way to think about the enrollment population, as well as thinking about why these patients may be dying today, as opposed to why they died 2 years ago. I think those are all incredibly insightful and valuable suggestions back to the agency.”

References:

• FDA. Meeting of the Pulmonary-Allergy Drugs Advisory Committee (PADAC). YouTube. https://www.youtube.com/watch?v=7uMTrZ1lSsY. Published Nov. 9, 2022. Accessed Nov. 9, 2022.
• FDA briefing document: EUA request 000113. https://www.fda.gov/media/162709/download. Accessed Nov. 8, 2022.