– The use of interleukin inhibitors for treatment of rheumatologic diseases doubles a patient’s risk of serious infections, according to a comprehensive systematic review and meta-analysis of 74 randomized, placebo-controlled trials presented by Jawad Bilal, MBBS, at the 2020 Rheumatology Winter Clinical Symposium.

Dr. Jawad Bilal, rheumatology fellow at the University of Arizona, Tucson. Bruce Jancin/MDedge News

Dr. Jawad Bilal

The meta-analysis, which incorporated 29,214 patients with a variety of rheumatic diseases, demonstrated that patients receiving interleukin (IL) inhibitors had a 1.97-fold increased risk of serious infections, a finding accompanied by a high degree of statistical certainty. The number-needed-to-harm was 67 patients treated for a median of 24 weeks in order to generate one additional serious infection.

“That number-needed-to-harm is a significant finding because having a serious infection means by definition you’re getting admitted to the hospital and receiving IV antibiotics,” Dr. Bilal observed in an interview.

The meta-analysis also found that IL inhibition was associated with a 2.35-fold increased risk of opportunistic infections and a 1.52-fold higher risk of developing cancer, both findings with statistical significance (P =.03) but only moderate certainty because fewer of those events were captured in the trials compared to the numbers of serious infections, explained Dr. Bilal of the University of Arizona, Tucson.

For opportunistic infections, the number-needed-to-harm was 250 patients treated with an IL inhibitor for a median of 54 weeks in order to result in one additional opportunistic infection. For cancer, the number-needed-to-harm was 250 for a median of 24 weeks.

Dr. Bilal noted that while the IL inhibitors are drugs of established efficacy in rheumatologic diseases, their safety has not previously undergone anything approaching the comprehensive scrutiny carried out in this meta-analysis. The meta-analysis, which included all published placebo-controlled randomized trials and their extension studies, employed rigorous methodology in accord with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) statement and the GRADE approach to data analysis. Studies of IL inhibitors in patients with dermatologic and GI diseases were excluded from the meta-analysis.

He offered a caveat regarding the cancer risk findings: “Our analysis showed that the cancer risk is increased, but the results are not conclusive because we only had a few years of data. With cancer, you really need at least 8-10 years of data. So the real-world experience with the interleukin inhibitors in the large registries is what’s going to tell if the cancer risk is really increased or not. In the meantime, we all have to be cautious.”

The number of serious infections collected in the meta-analysis afforded sufficient statistical power for the investigators to break down differential risks based on individual drugs and indications. Among the drugs associated with significantly increased risk of serious infections were anakinra, with an odds ratio of 2.67, compared with placebo; secukinumab with an OR of 2.43; and tocilizumab with an OR or 1.76. Ustekinumab and ixekizumab were associated with 2.57- and 3.89-fold increased risks, respectively, but the number of rheumatology patients treated with those two biologics wasn’t large enough for those findings to achieve statistical significance.

Rheumatoid arthritis patients who received an IL inhibitor rather than placebo had a 1.98-fold increased risk of serious infection, while those with psoriatic arthritis had a 2.21-fold increased risk. Patients treated for SLE had a 6.44-fold increased risk, and those with juvenile idiopathic arthritis had a 5.37-fold higher risk, but the margins for error were such that those results weren’t statistically significant.

“I think this study is going to help clinicians and patients when they’re trying to weigh the risks and benefits of IL inhibitors, especially if they already have risk factors, like a recent history of serious infection or a history of cancer or of opportunistic infection,” Dr. Bilal commented.

A study limitation was that he and his coinvestigators had to lump together the various IL inhibitors in order to gain statistical power, even though the drugs work differently, he noted.

Dr. Bilal reported having no financial conflicts regarding his study, the full details of which have been published (JAMA Netw Open. 2019 Oct 2. doi: 10.1001/jamanetworkopen.2019.13102).