ֱ

Safety Questions Dog Novel IL-6 Blocker for RA

— Efficacy robust, but an apparent excess risk of death

MedpageToday

An FDA advisory committee will have a difficult question to answer when it meets Wednesday -- whether an investigational biologic drug for rheumatoid arthritis targeting interleukin (IL)-6 should be approved despite data showing that 34 of 35 deaths among clinical trial participants were in patients receiving the agent.

The Arthritis Advisory Committee will review this and other for treatment of adults with moderately to severely active RA.

Sirukumab (proposed trade name Plivensia) would be the third anti-IL-6 monoclonal antibody to be approved, the others being tocilizumab (Actemra) and sarilumab (Kevzara). That it could be seen as a "me-too" product may give the safety issue added weight.

One of the phase III trials of sirukumab was and lead author Daniel Aletaha, MD, of the Medical University of Vienna, told ֱ, "It's an important study, because we were able to show with the new compound that you can further reduce disease activity in patients where lots of therapies have already failed. And in these times of treat-to-target, where we are setting a target for every patient that is quite ambitious -- low disease activity or remission -- we need to have new options for many patients."

Efficacy

Sirukumab's efficacy is not really in question. Primary data came from two phase III trials of adults with active RA who were randomized to receive placebo, sirukumab 100 mg every 2 weeks, or sirukumab 50 mg every 4 weeks. In one of these trials, patients had an inadequate response to conventional disease-modifying anti-rheumatic drug (DMARD) therapy, such as methotrexate or sulfasalazine, and had to have at least six swollen and tender joints. The trial was placebo-controlled for 52 weeks, with the option of early escape at week 18 or late escape at week 40. Co-administration of DMARDs was permitted.

The trial included 1,670 patients, most of whom were white women. Mean age was 53 and mean RA duration was 8.6 years. Mean number of swollen and tender joints were 15 and 25, respectively. Almost all had previously used methotrexate, and more than one-third had previously used a biologic.

At week 16, 55% of patients receiving the 50 mg dose achieved an ACR20 response, as did 54% of the 100 mg group compared with 26% of the placebo group (P<0.0001 for both versus placebo). Also at that time point, ACR50 responses were seen for 30% and 26% of the 50 mg and 100 mg groups compared with 11% of the placebo group (P<0.0001 for both), while 13%, 13%, and 4%, respectively, had ACR70 responses (P<0.0001).

At week 52, the mean change on the radiographic van der Heijde-modified Sharp score was 3.7 in the placebo group compared with 0.5 in both sirukumab groups (P<0.0001). Changes in erosion scores were 2.2 in the placebo group and 0.1 in both sirukumab groups (P<0.0001), while scores for joint space narrowing were 1.5 in the placebo group and 0.4 in both sirukumab groups (P<0.0001).

In the second trial, participants had an inadequate response to at least one anti-tumor necrosis factor (TNF) agent or intolerance to two or more anti-TNFs, and had at least four swollen and tender joints. The study included a 24-week placebo-controlled phase, a 28-week uncontrolled phase, and an additional 16 weeks of safety follow-up. Concomitant methotrexate was required.

This study included 878 patients, with the majority being women. Mean age was 55 and mean disease duration was 12 years. Mean number of swollen and tender joints were 16 and 28.

ACR20 responses were seen in 40% of the 50 mg group, 45% of the 100 mg group, and 24% of the placebo group at week 16 (P<0.0001 for both), while ACR50 responses were seen in 21%, 22%, and 9% (P<0.001 for both). ACR70 responses were seen in 3% of the placebo group, in 6% of the 50 mg group (P=0.07), and in 10% of the 100 mg group (P=0.0007).

On the Health Assessment Questionnaire Disability Index, changes from baseline in the first trial were -0.22 in the placebo group, -0.42 in the 50 mg group, and -0.45 in the 100 mg group (P<0.0001 for both) at week 16. On the Short Form 36 quality of life questionnaire, changes from baseline for the physical component were 2.6 in the placebo group, 5.5 in the 50 mg group, and 5.8 in the 100 mg group (P<0.0001 for both) at week 16, and on the mental component, the changes were 2.7 in the placebo group and 4.8 in both sirukumab groups (P=0.0001).

Janssen's proposed dosage for RA will be 50 mg, given subcutaneously every 4 weeks, and the proposed indication is similar to that of the other two IL-6 blockers.

"In terms of the efficacy profile, the product is robust," said Newman Yeilding, MD, head of immunology development at Janssen.

"We are not anticipating any major contention about the efficacy of the product," he said in an interview.

"In terms of safety, we anticipate there will be more discussion. One focus of the discussion will be an imbalance in mortality that occurs after the strictly controlled portions of our trials," Yeilding noted.

Safety

Indeed, that appears to be the major problem for sirukumab. As of February 2016, there were 35 deaths in the clinical development program, with 34 being in patients receiving sirukumab. The most common causes of death were cardiovascular events, serious infections, and malignancies.

During the placebo-controlled portion of the phase III trials, one patient receiving placebo died, of acute respiratory distress syndrome, as did one patient in the 50 mg group (sudden cardiac death) and one in the 100 mg group (myocardial infarction/hypertension).

In pooled analyses through 52 weeks of exposure, the mortality rates were 0.2, 0.5, and 0.8 per 100 patient-years for patients in the placebo, 50 mg, and 100 mg groups, respectively.

A briefing document prepared by FDA staff pointed out, however, that many patients in the active treatment groups through week 52 had early escape from the placebo group, and early escape patients typically had more severe disease. "Thus, these comparisons are not balanced by randomization," the document stated.

In all phase III trials of sirukumab, there were 29 major adverse cardiovascular events, with 21 in the 50 mg group and eight in the 100 mg group. Among patients with one or more cardiovascular events, seven had non-fatal myocardial infarctions, 14 had non-fatal strokes, and in 13, cardiovascular deaths occurred.

For serious infections, which were mostly pneumonia and cellulitis, the rates were 2.7, 5.3, and 4.7 per 100 patient-years through 52 weeks in the placebo, 50 mg, and 100 mg groups, respectively. The company noted that their proposed prescribing information includes a boxed warning regarding the risk of serious infections leading to hospitalization or death.

Malignancies were reported at rates of 0.64 per 100 patient-years for both sirukumab doses and 0.19 per 100 for placebo through 52 weeks. The malignancies most often were solid tumors such as of the breast and lung.

"Wide confidence intervals around treatment comparisons for serious rare events such as death, malignancy, and major adverse cardiovascular events indicate that the imbalances could be due to chance but also that relatively large increases in risks on sirukumab cannot be ruled out based on the data alone," FDA staff concluded.

Other events included decreases in neutrophils and liver function test abnormalities. For instance, by 52 weeks the mean changes in ALT were -1.1 IU/L in the placebo group, +11.1 IU/L in the 50 mg group, and +11.6 IU/L in the 100 mg group.

An increase from baseline was seen in low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, and triglycerides. For instance, at week 16 in the 50 mg group, there was an increase of approximately 21 mg/dL in LDL, of 7 mg/dL in HDL, and 37 mg/dL in triglycerides. These were slightly greater than the increases that have been reported with sarilumab and similar to what has been seen with tocilizumab.

There also were increased rates of gastrointestinal perforations, and the company suggested that the prescribing information could include a warning about this.

Nevertheless, Yeilding told ֱ, "Overall we believe we have a strong package, that the benefit-risk profile is strong, and that we have a solid story to tell the advisory committee."