Canakinumab a Success in Systemic JIA, Study Confirms

Canakinumab (Ilaris) was effective in treating a majority of children with systemic juvenile idiopathic arthritis (JIA), a real-world Italian multicenter study found.

Among a cohort of 80 children with systemic JIA, 63.7% achieved clinically inactive disease off glucocorticoids at 6 months, as did 57.4% of the 68 with active disease at baseline, according to Fabrizio De Benedetti, MD, PhD, of Ospedale Pediatrico Bambino Gesù in Rome, and colleagues.

And at 12 months, 63.8% of those with active disease at baseline had clinically inactive disease off glucocorticoids, the investigators reported in Rheumatology.

Systemic JIA differs from other types of JIA in that it is characterized not only by arthritis but also quotidian fever, rash, lymphadenopathy, hepatomegaly, and splenomegaly. Conventional treatment has relied on glucocorticoids, but these agents are associated with significant adverse effects when used long term.

Previous experience has demonstrated efficacy of the interleukin (IL)-1 receptor antagonist anakinra (Kineret) in systemic JIA, but not all patients respond, and the drug, which blocks both IL-1α and IL-1β, is given as daily injections.

Clinical trials have also found efficacy for canakinumab, which selectively blocks soluble IL-1β in a “substantial” number of patients, according to the investigators. In addition, glucocorticoid withdrawal was successful in 15.6% to 36.4% of patients in these trials.

Little information is available, however, on the real-world efficacy and safety of canakinumab in systemic JIA. Therefore, De Benedetti and colleagues conducted a retrospective observational study of children from 15 pediatric rheumatology centers who initiated treatment with this medication before the end of 2019.

The primary endpoint was clinically inactive disease without glucocorticoid use at 6 months.

About two-thirds of patients were girls and mean age was 10.4 years. Almost all had previously been treated with glucocorticoids, one-third had used methotrexate, and 63% had previous exposure to biologic therapy. A total of 26 of the 80 patients had previously had macrophage activation syndrome (MAS).

At baseline, 56.3% had previously been treated with anakinra, and of those, 73.3% had discontinued because of a lack of efficacy and had active disease at the time of starting canakinumab. The remaining 26.7% of patients who switched from anakinra had stopped treatment with that medication because of pain from the daily injections. The median dosage of canakinumab was 4 mg/kg every 4 weeks.

Fever was present at the time of treatment initiation in 55.9%, and the median number of active joints was two.

At 6 months, 29 patients continued to have active disease, 14 of whom had previously received anakinra.

Of the 33 patients who had previously been given anakinra and switched to canakinumab because of persistently active disease, 19 had achieved clinically inactive disease at 6 months (57.6%), which was similar to the 57.1% rate among patients who had not previously had IL-1 inhibitor treatment. “This observation suggests that failure of one anti-IL-1 drug does not necessarily preclude use of another one,” the investigators noted.

On a univariate analysis, the number of active joints was higher among nonresponders, and the investigators chose a cutoff of five (based on the clinical definition of polyarticular JIA). The risk ratio for nonresponse at 6 months among patients with five or more active joints at baseline was 2.4 (95% CI 1.5-4.0, P=0.001), and the risk ratio for nonresponse among patients with a history of MAS was 2.0 (95% CI 1.2-3.3).

Responders typically initiated treatment with canakinumab earlier in the course of disease, at 13.9 months after symptom onset compared with 28.7 months among nonresponders (P=0.047).

On a logistic regression analysis, nonresponse was associated independently with number of active joints above five (OR 6.37, 95% CI 1.69-24.02, P=0.006) and with a history of MAS (OR 3.53, 95% CI 1.06-11.70, P=0.039).

Patients who had both of these factors had only an 11.8% probability of response (95% CI 0.0-26.9), whereas those with fewer than five active joints and no previous MAS had a 75.9% probability of response (95% CI 62.7-89.0).

No serious adverse events were reported. Lymphopenia developed in one patient, who withdrew from treatment. Six patients had nonserious infections and one developed herpes zoster; all responded to standard treatment. Two patients developed episodes of MAS, for a rate of 2.9 cases per 100 patient-years.

The response rates in this real-world study were higher than in the earlier clinical trials, which was likely because the disease severity in the trials was worse, with a baseline number of active joints of 10 rather than the two reported in this cohort.

A recent study found that CXCL9 levels were higher among canakinumab nonresponders. This ligand has been linked with interferon-γ, which has increasingly been seen as having a pathogenic role in MAS. Further studies will be needed to more fully explore the immunobiology of MAS in systemic JIA, the researchers noted.

Limitations of the study included its relatively small size and retrospective design.