OP0230 EFFICACY AND SAFETY OF GUSELKUMAB IN PATIENTS WITH ACTIVE PSORIATIC ARTHRITIS WHO DEMONSTRATED INADEQUATE RESPONSE TO TUMOR NECROSIS FACTOR INHIBITION: WEEK 24 RESULTS OF A PHASE 3B, RANDOMIZED, CONTROLLED STUDY

Abstract

Background:Guselkumab (GUS), a selective monoclonal antibody targeting the interleukin-23p19 subunit, has demonstrated efficacy in 2 pivotal Ph3 psoriatic arthritis (PsA) studies (DISCOVER-1,1 DISCOVER-22).Objectives:Evaluate GUS efficacy and safety in PsA patients (pts) with inadequate response (IR) to tumor-necrosis-factor inhibition (TNFi) through Week24 (W24) of the Ph3b COSMOS study.Methods:In this randomized, double-blind, placebo (PBO)-controlled trial, 285 pts with active PsA (≥3 swollen & ≥3 tender joints) who demonstrated lack of benefit or intolerance to 1-2 TNFi were randomized 2:1 to subcutaneous GUS 100mg (n=189) or PBO (n=96) at W0, W4, then every 8 weeks (Q8W) through W44 (with PBO crossover to GUS at W24). At W16, pts who met early escape (EE) criteria (<5% improvement in both tender & swollen joint counts) also could switch from PBO to GUS. The primary efficacy endpoint was ACR20 response at W24 among randomized, treated pts. Pts missing ACR20 data at W24 or who met treatment failure criteria (including meeting EE criteria at W16) were considered nonresponders (NRs). Subgroup analyses were performed to assess consistency of primary treatment effect based on demographics, disease characteristics, and medication use at baseline. Prespecified sensitivity analyses included ‘Per-Protocol’ (PP) (excluded pts with major protocol deviations) and ‘EE-Correction’ (included pts incorrectly routed to EE) analyses. Adverse events (AEs) were summarized by treatment received.Results:Baseline characteristics were similar across GUS and PBO pts, though a higher proportion of females and more severe joint symptoms were seen in the GUS group. At W24, 44.4% of GUS vs 19.8% of PBO pts achieved ACR20 (p<0.001) (Figure). GUS was superior to PBO for all major secondary endpoints. Efficacy was consistent across subgroups defined by baseline characteristics, including in pts who discontinued prior TNFi use due to inadequate efficacy (84% GUS vs 81% PBO) and safety (16% GUS vs 19% PBO) (Table). 20 pts (12 GUS, 8 PBO) were incorrectly routed to EE. Results of PP (48.8% vs 23.8%) and EE-correction (48.1% vs 19.8%) sensitivity analyses were consistent with the primary analysis (Figure). AEs were similar between GUS- and PBO-treated pts (Table).Table 1.Baseline characteristics of, and adverse events reported by, randomized and treated COSMOS ptsGUS 100 mg Q8W (N=189)PBO (N=96) Age, y4949 Sex, Female54%46% Duration of PsA, y8.38.7 Body mass index, kg/m22931a Swollen (0-66) / tender (0-68) joint count10 / 219 / 18 Pt pain / Pt global arthritis / Physician global disease, 0-10 cm VAS6.5 / 6.5 / 6.96.0 / 6.2 / 6.4 Health Assessment Questionnaire-Disability Index, 0-31.3b1.2 C-reactive protein, mg/dL1.2b1.2 Methotrexate use at baseline56%53% Psoriatic body surface area, %17.913.4 Number of prior TNFi: 1 / 288% / 12%89% / 11% Reason for prior TNFi discontinuation: Efficacy / Safety84% / 16%* 81% / 19%*Pts with ≥1 AE / SAE37% / 3%48% / 3%Pts with ≥1 infection / serious infection18% / 0%20% / 0%Pts with ≥1 AE leading to study agent discontinuation2%2%Pts with ≥1 malignancy0.4%0Pts with ≥1 injection-site reaction2%1%Data shown are mean or %. aN=95; bN=188. *Missing for 1 pt. SAE – serious adverse events; VAS – visual analog scaleConclusion:In this Ph3b, placebo-controlled study of PsA pts with IR to 1-2 TNFi, GUS 100 mg Q8W elicited a significantly higher ACR20 response rate vs. PBO at W24; results of prespecified sensitivity and subgroup analyses were consistent. GUS safety in TNF-IR PsA pts through W24 is consistent with the favorable GUS safety profile in psoriasis and biologic-naïve PsA pts.3