Randomized controlled trials provide high internal validity but often have limited generalizability. Benchmarking real-world data against randomized controlled trial findings helps evaluate external validity. The objectives were to assess whether the results of the IXORA-S trial comparing ixekizumab and ustekinumab could be replicated using real-world data from the Shanghai Psoriasis Effectiveness Evaluation Cohort (SPEECH) and to explore potential reasons for differences between trial and observational findings.
We conducted a prospective cohort study of adults with moderate-to-severe plaque psoriasis enrolled in SPEECH (2021-2023) and treated with ixekizumab or ustekinumab. Propensity-score methods and multiple imputation were applied to address confounding and missing data. Outcomes were benchmarked against IXORA-S using agreement metrics and sensitivity analyses, including population trimming. The primary endpoint was the proportion of patients achieving a ≥90% reduction in the Psoriasis Area and Severity Index (PASI 90) at week 12.
Among 367 patients (ixekizumab n = 203; ustekinumab n = 164), ixekizumab achieved higher PASI 90 response rates at week 12 compared with ustekinumab. Effect estimates were consistent with IXORA-S in direction, magnitude, and statistical significance. Differences in baseline eligibility criteria and disease severity explained part of the efficacy-effectiveness gap.
The results of IXORA-S were successfully replicated using real-world data from SPEECH. This demonstrates that registry-based observational data can yield reliable estimates of treatment effects and may be used to investigate effectiveness questions not feasible to study in randomized trials, such as long-term outcomes and effects in underrepresented populations.