POS0072 CONSISTENT LONG-TERM GUSELKUMAB EFFICACY ACROSS PSORIATIC ARTHRITIS DOMAINS IRRESPECTIVE OF BASELINE PATIENT CHARACTERISTICS

Abstract

BackgroundPsoriatic arthritis (PsA) patients (pts) with differing baseline (BL) characteristics may vary in their response to treatment. In the phase 3 DISCOVER-1 and DISCOVER-2 studies, guselkumab (GUS) significantly improved joint symptoms, skin disease, enthesitis, dactylitis, physical function, and quality of life at Week (W) 24 in pts with PsA.1,2 Clinical responses across these disease domains were maintained or increased with GUS at W52,3.4 regardless of BL pt demographics, disease characteristics, or conventional synthetic disease-modifying antirheumatic drug (csDMARD) use.5 Durable efficacy with GUS through W100 across multiple disease domains was observed.6ObjectivesAssess both BL predictors of, and by BL pt subgroups, GUS efficacy across PsA disease domains through W100 of DISCOVER-2.MethodsBiologic-naïve adults with active PsA despite standard therapies were enrolled in DISCOVER-2 (swollen joint count [SJC] ≥5 & tender joint count [TJC] ≥5, C-reactive protein [CRP] ≥0.6 mg/dL). Pts were randomized 1:1:1 to GUS 100 mg every 4 weeks (Q4W); GUS 100 mg at W0, W4, then Q8W; or placebo (PBO).2 GUS effects on joint, skin, enthesitis, dactylitis, spinal pain, and disease severity endpoints (change in Disease Activity in PsA [DAPSA], SJC, and TJC scores; Psoriasis [PsO] Area Severity Index [PASI] score [among pts with BL IGA ≥2 and body surface area [BSA] with PsO ≥3%]; Leeds enthesitis index [LEI] score [among pts with enthesitis at BL]; dactylitis score [among pts with dactylitis at BL]; spinal pain score; and PsA Disease Activity Score [PASDAS], respectively) at W100 were evaluated for GUS-randomized pts, both by treatment group and by pooling pts across Q4W and Q8W treatment arms. A multivariate linear model adjusting for BL pt characteristics assessed associations between BL predictors of interest and changes in DAPSA, PASI, and LEI scores from BL to W100, and to assess least squares mean (LSM) changes and 95% confidence intervals (CIs) in all continuous endpoints from BL to W100 within subgroups of pts defined by BL sex, body mass index (BMI), PsA duration, SJC, TJC, CRP level, %BSA, PASI score, and csDMARD use.Results442 (90%) GUS-randomized pts completed study treatment through W100.6 Among the BL predictors of long-term GUS efficacy assessed (see above), only PsA duration (p=0.032), SJC (p<0.001), and TJC (p<0.001) were significant predictors of long-term (BL to W100) DAPSA score change; %BSA (p=0.002), PASI score (<0.001), SJC (p=0.008), and csDMARD use (p=0.014) were significant predictors of long-term PASI score change; and none significantly predicted long-term LEI score change among pooled GUS pts (Figure 1). However, statistically significant improvements from BL to W100 in DAPSA, PASI, and LEI scores were observed across all BL strata, including those indicating more extensive or severe disease, in pooled GUS Q4W+Q8W pts (Figure 1, all p<0.001) and within each dosing group. Similar improvements were observed for other continuous endpoints assessed (change in Psoriatic Arthritis Disease Activity Score [PASDAS], SJC, TJC, spinal pain, and dactylitis score).ConclusionGUS significantly improved PsA signs and symptoms through W100 across all BL pt subgroups evaluated, including pts with highly active disease, and regardless of dosing regimen.