FDA advisors voted against approving sirukumab, a monoclonal antibody that targets interleukin (IL)-6, for patients with rheumatoid arthritis who experience an inadequate response, or are intolerant to, one or more disease-modifying anti-rheumatic drugs (DMARDs).
Wednesday's vote by the Arthritis Advisory Committee was 12-1 against approval, with the primary reasons being a safety signal for all-cause mortality in the clinical trials as well as a too-broad indication that would permit use of the drug after failure of a single DMARD such as methotrexate.
As detailed at the meeting and in documents posted earlier in the week, there had been 35 deaths from any cause among trial participants through February 2016, with 34 of these being among patients receiving the monoclonal antibody. In pooled analyses through 52 weeks of exposure, the mortality rates were 0.2 per 100 patient-years for patients receiving placebo, 0.5 per 100 for those receiving sirukumab, 50 mg every 4 weeks, and 0.8 per 100 for those given sirukumab, 100 mg every 2 weeks.
The most common causes of death were cardiovascular events, serious infections, and malignancies.
The committee also voted 12-1 against the safety profile of sirukumab being adequate to support approval. The one committee member who voted in favor, James Katz, MD, director of the rheumatology fellowship and training branch of the NIH's National Institute of Arthritis and Musculoskeletal and Skin Diseases in Bethesda, Md., said of his vote, "This drug doesn't scare me any more than the other biologics. I'm scared of all the biologics."
In general, the committee members indicated that the potential benefits of sirukumab don't outweigh the possible risks, which included not only mortality, but also earlier events such as serious infections and laboratory abnormalities.
Several commented on the proposed indication for sirukumab, specifically that it could be used for patients who had failed only one DMARD. For instance, Michael H. Weisman, MD, of the University of California Los Angeles, suggested that he might have voted in favor if the indication was for patients who were nonresponders to biologics, rather than just DMARDs. Mara Becker, MD, of the University of Missouri Kansas City, agreed, voicing concern that, if approved, sirukumab might be used early in the course of treatment.
There was considerable discussion about the trial design and how that may have "muddied the waters" of the safety analysis. Because clinical trials in RA today do not permit lengthy placebo-controlled phases out of concern that joint damage may be occurring even if the patient is responding clinically, the studies had the opportunity for early escape for nonresponders and then crossover with re-randomization for all patients.
This meant that nonresponding patients in the placebo group -- those with the worst disease -- subsequently could be overrepresented in the active treatment groups, creating potential bias in the analysis, according to representatives of Janssen, the company developing the product.
The mortality signal therefore "maybe could be just noise, but it also might be real," said David Felson, MD, of Boston University School of Medicine. Others echoed that concern, saying that they don't have enough information to clarify if it's a real safety signal or an artifact of trial design.
Felson also pointed out that rheumatologists now have "a very large armamentarium" that includes 16 biologics and nine small molecules. This also includes two other IL-6 inhibitors, tocilizumab (Actemra) and sarilumab (Kevzara) that have been approved. One committee member who voted "no," Maria Suarez-Almazor, MD, PhD, of the MD Anderson Cancer Center in Houston, suggested that her vote might have been different if the drug represented a completely novel mechanism of action.
The efficacy of sirukumab was not questioned, with all 13 members voting "yes" that the efficacy was adequate for approval. The responses were robust and consistent across endpoints and in sensitivity analyses, the group agreed. At week 16, 55% of patients receiving the 50-mg dose had 20% responses on the criteria of the American College of Rheumatology (ACR20) as did 54% of those given the 100-mg dose compared with 26% of the placebo group (P<0.0001 for both).
And efficacy was seen for both patients who had been inadequate responders to conventional DMARDs, such as methotrexate, in one trial that included 1,670 patients, and also in a second trial of 878 patients who were nonresponders to anti-tumor necrosis factor therapy.