Efficacy and Safety of Guselkumab in Real-World Evidence: A Systematic Review and Meta-Analysis

Abstract

Background: Guselkumab, a selective anti-IL-23p19 monoclonal antibody, has shown high efficacy in randomized trials for moderate-to-severe psoriasis and active psoriatic arthritis (PsA). Real-world evidence is essential to assess treatment performance in broader and more heterogeneous patient populations. Objectives: To synthesize the available real-world evidence on the effectiveness and safety of guselkumab in psoriasis and PsA, and to quantify pooled clinical responses across clinically relevant follow-up windows. Methods: A systematic review and meta-analysis was conducted according to PRISMA 2020. PubMed/MEDLINE, MEDLINE, and Web of Science were searched for observational real-world studies of guselkumab in adults with psoriasis and/or PsA. Primary pooled outcomes were PASI 90 and PASI 100 in psoriasis and DAPSA < 14 in PsA. Random-effects meta-analysis of proportions was performed using a logit transformation. Cohort overlap was explicitly assessed and overlapping publications were not allowed to contribute concurrently to the same pooled estimate. Results: Thirty-four studies were included (29 psoriasis, 5 PsA). In psoriasis, pooled PASI 90 response rates were 50.8% (95% CI 46.8–54.8) at 12–16 weeks, 68.4% (95% CI 66.3–70.4) at 20–28 weeks, 71.2% (95% CI 64.9–76.8) at 36–60 weeks, and 77.1% (95% CI 74.9–79.3) at ≥96 weeks. Pooled PASI 100 response rates were 49.8% (95% CI 47.5–52.2) at 20–28 weeks and 49.7% (95% CI 47.1–52.3) at 36–60 weeks. In PsA, pooled DAPSA < 14 response rates were 56.9% (95% CI 23.0–85.3) at 20–28 weeks and 69.5% (95% CI 62.5–75.7) at 48–60 weeks. Safety reporting was heterogeneous, but serious adverse events and discontinuations due to adverse events were uncommon. Conclusions: Real-world evidence supports guselkumab as an effective treatment for psoriasis and a clinically useful option for PsA, with a safety profile broadly consistent with the known trial experience. Interpretation remains limited by observational designs, heterogeneous denominators, and inconsistent safety reporting.