Background/Purpose: A prior analysis found that patients (pts) with psoriatic arthritis (PsA) treated with guselkumab (GUS), an IL-23p19 subunit inhibitor, were approximately 2× more likely to remain persistent with on-label treatment up to 24 months (M) vs first-time subcutaneous (SC) TNFi users.1 Real-world evidence comparing long-term on-label persistence between GUS and SC TNFi in biologic-naïve and biologic-experienced populations is lacking.

Methods: IQVIA PharMetrics® Plus database (1/1/2011-6/30/2023) was used to analyze treatment persistence among biologic-naïve and -experienced adults with active PsA who initiated GUS/SC TNFi (index date=first claim for GUS/SC TNFi). Analyses were performed separately for biologic-naïve (no prior pre-index biologic claims) and biologic-experienced (≥1 prior pre-index biologic claim) pts. Pts had ≥12M insurance eligibility and no claims for confounding diseases or index medications pre-index date. Baseline characteristics were summarized for the 12M pre-index period. On-label persistence was defined as continued use of index therapy without dose modification (escalation/reduction) per FDA-approved labeling. Propensity score overlap weighting was applied to balance baseline characteristics between the GUS and SC TNFi cohorts. On-label persistence over 24M was compared via weighted Kaplan-Meier (KM) curves and Cox proportional hazards model. For biologic-naïve pts, further adjustment for prior conventional synthetic (cs) and targeted synthetic (ts) disease-modifying antirheumatic drug (DMARD) use was performed.

Results: Among 3,294 pts, 361 GUS pts (49.5 years; 59.8% female) and 2,171 SC TNFi pts (48.3 years; 59.5% female) were biologic-naïve, and 443 in GUS cohort (50.8 years; 61.2% female) and 319 in SC TNFi cohort (48.5 years; 64.3% female) were biologic-experienced. Weighted baseline characteristics were balanced between cohorts, except for prior csDMARD/tsDMARD use among biologic-naïve pts. For biologic-naïve pts, the median time-to-discontinuation was 22.4M for GUS and 9.2M for SC TNFi; weighted KM on-label persistence rates at 6M/12M/18M/24M: 86.2%/73.1%/62.9%/48.9% for GUS vs 63.8%/43.8%/35.3%/28.4% for SC TNFi (all log-rank p < 0.001). For biologic-experienced pts, the median time-to-discontinuation was 18.4M for GUS and 7.4M for SC TNFi; weighted KM on-label persistence rates at 6M/12M/18M/24M: 76.9%/57.7%/50.6%/39.5% for GUS vs 58.1%/40.3%/31.1%/23.3% for SC TNFi (all log-rank p < 0.001). At 24M, GUS cohort pts were ~2× more likely to remain persistent vs SC TNFi cohort in both biologic-naïve (HR: 2.36, 95% CI: 1.88-2.98; p < 0.001) and biologic-experienced pts (HR: 1.86, 95% CI: 1.46-2.37; p < 0.001).

Conclusion: Pts initiating GUS demonstrated statistically significantly higher on-label persistence through 24M than those initiating SC TNFi in both biologic-naïve and biologic-experienced pts with active PsA. Higher long-term on-label persistence may improve disease management outcomes, including functional status and quality of life. Reference: 1Mease PJ, et al. Poster. CCR-West 2024; Sept 26-29, 2024; San Diego, CA, USA.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/real-world-on-label-treatment-persistence-through-24-months-in-biologic-naive-and-biologic-experienced-patients-with-psoriatic-arthritis-comparison-of-guselkumab-versus-subcutaneous-tumor-necrosis-fa/