C.Diff Unmet Need Overcomes Bezlotoxumab Panel's Efficacy Concerns

Advisory committee votes 10-5 for approval of the monoclonal antibody for prevention of C. difficile recurrence; drug shows 10% absolute reduction in recurrence vs. placebo.

Despite reservations about Merck & Co. Inc.'s bezlotoxumab clinical trial design and efficacy rate, an FDA advisory committee voted 10 to 5 for approval of the monoclonal antibody for prevention of Clostridium difficile infection recurrence.

FDA's Antimicrobial Drugs Advisory Committee reviewed bezlotoxumab at its June 9 meeting. In explaining their reasons for recommending approval, committee members cited the unmet medical need for therapies to treat recurrence of C. difficile infection (CDI). There are no approved treatments for the indication. One panel member abstained from the vote.

Committee member Peter Weina, Walter Reed National Military Medical Center, said he voted for approval despite the lack of conclusive evidence of the drug's safety and efficacy. He cited the possibility that bezlotoxumab could prevent 8,000 cases of C. difficile infection per year and 3,000 deaths per year.

However, he said that one would expect bezlotoxumab to provide better than a 10% improvement. "You would expect something much more substantive, "Weina said. He suggested that the drug might work better for other isolates.

Limiting the treatment to those at high risk was suggested by several committee members.

In Merck's two pivotal Phase III trials, bezlotoxumab demonstrated an absolute reduction in CDI recurrence of approximately 10% and a relative reduction of approximately 40% compared to placebo in patients receiving standard of care antibiotic therapy for CDI.

During the panel discussion, Merck was asked why there was not a larger difference between bezlotoxumab and placebo.

"Obviously, we wish it was higher but at the same time we consider this to be a significant effect," a Merck representative responded. He noted that the effect is on par with other monoclonal antibodies and most vaccines, including influenza vaccines.

"We feel we have the right dose and we've demonstrated the efficacy of bezlotoxumab in these patients," he said. "Indeed, in the setting of unmet medical need, huge public health burden and the fact that there are no treatment options for these patients, we believe it's an appropriate agent for consideration."

Several panel members suggested that use of the drug be limited to high risk patients and that the company continue to collect safety data. Some members voiced concern about a cardiovascular safety signal and recommended that labeling include a warning about use in patients with heart problems.

The drug had an overall favorable safety profile with the rates of treatment-emergent adverse events, serious adverse events and deaths similar to placebo. But in the two Phase III trials there were numerically higher numbers of serious adverse events and deaths in bezlotoxumab-treated patients with baseline congestive heart failure compared to placebo.

While panelists raised various questions about the data and trial design, they did not seem to share FDA's concern about choice of endpoint.

In a briefing document prepared for the meeting, FDA questioned whether efficacy of the drug could be determined based on the primary efficacy endpoint used in the pivotal trials. The agency had recommended that Merck use the primary endpoint of global cure (sustained clinical response) rather than prevention of CDI recurrence over a 12-week follow-up period, but the company rejected the suggestion, saying that recurrence rate is the most specific endpoint for the indication (Also see "Merck Went Own Way On Bezlotoxumab Endpoint And FDA May Not Follow" - Pink Sheet, 7 Jun, 2016.).

In the pivotal Phase III trials (P001 and P002, also known as MODIFY I and MODIFY II) global cure was the secondary endpoint and clinical cure of the initial CDI episode was an exploratory endpoint.

The panel did not seem to share FDA's concerns about the endpoint. Panel member Thomas Moore, University of Kansas School of Medicine-Wichita Campus, who voted for approval, praised FDA for its dissection of the data and the sponsor for its "heavy lifting" in developing the drug. He said that as a panel member it was difficult to "walk into a disagreement between the sponsor and FDA," but added that he thought it was an "amicable disagreement."

The panel focused on other aspects of the data. Committee member Joan Hilton, University of California, San Francisco School of Medicine, said she voted for approval despite weaknesses in the study design. She said it would have been helpful if patients had been randomized to receive bezlotoxumab treatment after the initial infection was resolved. In the two Phase III studies bezlotoxumab was administered concomitantly with standard of care antibiotics for the baseline CDI episode.

Merck explained that the study design required administration of the study medication during the course of antibiotic therapy rather than afterward to ensure the antibody would be present during the full at-risk period.

Among those voting against approval, panel member Lindsey Baden, Brigham and Women's Hospital and Harvard Medical School, said substantial evidence of efficacy had not been presented and that the data is conflicting. He noted that only 800 patients in total were treated with the therapy, alluding to the bezlotoxumab-only arms in the Phase III trials.

"There is a cloud in my mind over the efficacy," Baden stated. There was a jump from Phase II to Phase III trials, which "in my mind are Phase IIb trials."

Study P001 is an adaptive, four-arm, factorial trial designed to evaluate actoxumab (anti-toxin A antibody), bezlotoxumab (anti-toxin B antibody), or actoxumab + bezlotoxumab versus placebo. Study P002 was identical in design except it did not include an actoxumab arm, had no interim analysis, and had an extended follow-up period through month 12 in a subset of patients.

In Study P001, 15.9% of subjects in the actoxumab plus bezlotoxumab group had CDI recurrence compared to 17.4% in the bezlotoxumab group and 27.6% in the placebo group. In Study P002, 14.9% in the actoxumab plus bezlotoxumab, 15.7% in the bezlotoxumab group and 25.7% in the placebo group had CDI recurrence.

Several antibiotic alternatives are in development for treatment of recurrent C. difficile infection, including Pfizer Inc.'s vaccine PF-06425090, Rebiotix Inc.'s probiotic RBX2660, Seres Therapeutics Inc.'s probiotic SER-109, and Shire PLC's probiotic VP20621, all in Phase II clinical trials (Also see "Antibiotic Discovery Hurdles: Could Data Sharing, Research Hub Solve Problem?" - Pink Sheet, 11 May, 2016.).