Guselkumab Succeeds for TNF Failures in Psoriatic Arthritis

Patients with psoriatic arthritis who had been treated unsuccessfully with one or more tumor necrosis factor (TNF) inhibitors showed high rates of efficacy with guselkumab (Tremfya) treatment, a phase IIIb study found.

At week 24, 44.4% of patients randomized to guselkumab showed 20% improvements in the criteria of the American College of Rheumatology (ACR20) compared with 19.8% of those given placebo (P<0.001), reported Laura C. Coates, MBChB, PhD, of the University of Oxford in England at the European League Against Rheumatism virtual congress.

Guselkumab is a monoclonal antibody that targets the p19 subunit of interleukin (IL)-23, and was shown to be effective for psoriatic arthritis in two pivotal phase III trials, DISCOVER-1 and DISCOVER-2. It has been approved for use in psoriatic arthritis and severe plaque psoriasis.

DISCOVER-1 included a subgroup of patients who had previously been treated with TNF inhibitors, which is a cohort of psoriatic arthritis patients who have significant unmet needs, Coates said.

To more fully explore the potential for guselkumab in this group of patients, Coates and colleagues conducted a trial known as COSMOS that included 285 patients who had either failed or been intolerant to anti-TNF treatment. Participants had active disease, with at least three swollen and tender joints.

Mean age was 49, and the sexes were evenly represented.

“These patients had significant disease activity across domains,” she said. Mean duration of psoriatic arthritis was 8.5 years, and mean swollen and tender joint counts were 10 and 20, respectively. Patient pain and physician global assessment were 6.5, psoriasis-affected body surface area averaged 15%, and Health Assessment Questionnaire Disability Index (HAQ-DI) scores averaged 1.3.

A total of 88% had previously used one TNF inhibitor, and the remainder had received two. More than half of patients were on methotrexate at baseline.

The reason for stopping the previous TNF inhibitor was inadequate response in 84% and adverse events (AEs) in 16%.

Patients were randomized (2:1) to subcutaneous guselkumab or placebo at weeks 0 and 4 and every 8 weeks thereafter through week 44, with patients in the placebo crossing over to the active treatment at week 24.

In addition, at week 16, patients with less than 5% improvement in their swollen and tender joint counts were permitted early escape into the active treatment arm. Because 20 patients (12 in the guselkumab arm) were incorrectly given early escape, a sensitivity analysis was conducted to account for this.

Improvements with guselkumab were observed as early as week 4, at which time ACR20 responses were seen in 19% of the guselkumab group compared with 4.2% of the placebo group. By week 48, ACR20 responses had been achieved by 57.7% of patients in the guselkumab group and in 54.9% of those who had initially received placebo and switched to the active treatment after week 24.

In the sensitivity analysis that corrected for early escape, the result on the primary endpoint of ACR20 was similar, at 48.1%, to the 44.4% in the overall main analysis.

Guselkumab also was superior to placebo on all major secondary endpoints that represented a wide range of psoriatic arthritis domains, Coates said. Physical function as evaluated on the HAQ-DI showed changes from baseline of -0.01 at week 24 in the active treatment arm compared with -0.18 in the placebo group (P=0.003), with further improvement at 1 year (-0.40 vs -0.25).

For health-related quality of life assessed on the Short Form-36 physical component score, the change at week 24 was 3.51 in the guselkumab group compared with -0.39 in the placebo group (P<0.001), while at week 48 the numbers were 7.02 and 4.62, respectively.

Complete skin clearance was observed in 30.8% of patients in the guselkumab group compared with only 3.8% in the placebo group, rising to 53.4% and 39.1% by week 48.

The more stringent ACR50 joint response criteria was met at week 24 by 19.6% of the active treatment group versus 5.2% of the placebo group; these numbers rose to 39.2% and 29.4% at week 48.

The AE profile was consistent with the established safety profile seen in psoriasis and in the previous psoriatic arthritis trials, and the rates of AEs did not increase through 1 year. The rate of serious AEs with guselkumab through 56 weeks of follow-up was 6.2/100 patient-years, while for placebo during the first 24 weeks the rate was 8.2/100 patient-years.

No patients experienced opportunistic infections, active tuberculosis, anaphylaxis/serum sickness-like reaction, inflammatory bowel disease, or death.

“The results of this trial show a durable and broad impact of targeting the p19 subunit of IL-23 in patients with active psoriatic arthritis who have previously had an inadequate response to one or two TNF inhibitors,” she concluded.