Comparative efficacy and safety of picankibart versus guselkumab and tildrakizumab for moderate-to-severe plaque psoriasis: a systematic review and individual participant data-anchored population-adjusted network meta-analysis

Introduction

The therapeutic landscape of moderate-to-severe plaque psoriasis has rapidly evolved with the emergence of targeted biologics inhibiting the interleukin (IL)-23/T-helper 17 pathway (1-3). IL-23 p19 inhibitors are now an important treatment option because they can provide high levels of skin clearance with relatively infrequent maintenance dosing (3). In China, several IL-23 p19 inhibitors have been approved for plaque psoriasis, including guselkumab, tildrakizumab, and, more recently, picankibart. As these options expand, clinicians and payers increasingly need comparative evidence within the same therapeutic class, rather than placebo-controlled efficacy data alone.

Picankibart is a novel, humanized monoclonal antibody targeting the IL-23 p19 subunit. Recent phase III clinical trials have demonstrated its robust efficacy and favorable safety profile in patients with moderate-to-severe plaque psoriasis (4). However, no direct head-to-head trial has compared picankibart with established p19 inhibitors such as guselkumab or tildrakizumab. This evidence gap is clinically relevant because the drugs differ in dosing interval, accumulated clinical experience, and expected efficacy profile. Comparative evidence is also important for formulary decisions in China, where the choice among biologics is influenced not only by efficacy and safety but also by dosing convenience and healthcare resource use.

Indirect comparisons are often used when head-to-head trials are unavailable, but their validity depends on the transitivity assumption (5). In psoriasis, this assumption may be threatened when trial populations differ in patient characteristics related to prognosis or treatment response. In the present analysis, body mass index and prior biologic exposure were prespecified as the patient-level effect modifiers because they were consistently available across the individual participant data (IPD) and aggregate-data sources. When such differences exist, conventional aggregate-level network meta-analysis (NMA) may yield biased estimates. To restore transitivity, population-adjusted indirect comparisons (PAICs)—such as matching-adjusted indirect comparison (MAIC) or simulated treatment comparison (STC)—are increasingly utilized. However, these pairwise methods are challenging to generalize to multi-arm networks without losing statistical efficiency or introducing selection bias. Leveraging IPD within a multilevel network meta-regression (ML-NMR) framework overcomes these limitations by directly modeling patient-level covariates, thereby helping to address measured transitivity violations across the entire evidence network.

ML-NMR provides a framework for combining IPD from one or more studies with aggregate data (AgD) from other trials (6,7). By modelling treatment-covariate interactions and integrating over the covariate distributions of aggregate-data studies, ML-NMR can reduce aggregation bias and generate estimates tailored to a defined target population. It does not replace a randomized head-to-head trial, but it can provide more informative indirect evidence than an unadjusted aggregate-data NMA when important baseline differences are present.

This study aimed to compare the induction-phase efficacy and safety of picankibart with National Medical Products Administration (NMPA)-approved IL-23 p19 inhibitors, guselkumab and tildrakizumab, in adults with moderate-to-severe plaque psoriasis. We used IPD from the picankibart phase III trial and AgD from comparator trials in an ML-NMR framework. We present this article in accordance with the PRISMA-NMA reporting checklist (8) (available at https://amj.amegroups.com/article/view/10.21037/amj-2026-0041/rc).

Methods

This systematic review and NMA was not registered in any prospective registry.

Data source and search strategy

We searched PubMed, Embase, Web of Science, China National Knowledge Infrastructure (CNKI), and Wanfang from database inception to May 28, 2026. Search terms combined disease-related terms, trial-design terms, and treatment names or developmental codes for picankibart, guselkumab, and tildrakizumab. The full search strategy is provided in Table S1.

No language restriction was applied during screening. The eligible studies identified by the search were published in English or Chinese. Records were managed and deduplicated using Zotero, version 7.0.

Inclusion and exclusion criteria

Population: we included studies enrolling adult patients (aged ≥18 years) with a diagnosis of moderate-to-severe chronic plaque psoriasis who were candidates for systemic therapy or phototherapy. To ensure the transitivity of the network and the homogeneity of the study populations, trials focusing primarily on non-plaque forms were excluded. Similarly, trials restricted exclusively to patients with site-specific psoriasis were excluded. However, data examining these difficult-to-treat areas were eligible for inclusion if they were derived from subgroups within studies of the general moderate-to-severe plaque psoriasis population. Studies enrolling patients with mild disease were also excluded.

Interventions and comparators: the index intervention was picankibart. Comparators were limited to IL-23 inhibitors approved by NMPA of China for the treatment of moderate-to-severe plaque psoriasis—specifically guselkumab and tildrakizumab—administered according to their NMPA-approved dosing regimens. Placebo was included as a common comparator to anchor the network. Studies evaluating biosimilars not treated as distinct therapeutic entities were excluded.

Outcomes: the primary efficacy outcome was the proportion of patients achieving Psoriasis Area and Severity Index (PASI) 90 at the end of induction. The primary safety outcome was the incidence of serious adverse events (SAEs). Secondary efficacy outcomes included PASI 75, static Physician Global Assessment (sPGA) 0/1, and Dermatology Life Quality Index (DLQI) 0/1. Overall adverse events (AEs) were assessed as a secondary safety outcome.

Relative treatment effects were expressed as risk ratios (RRs) with 95% credible intervals (CrIs). To improve clinical interpretability, absolute risk differences (RDs) with 95% CrIs were also estimated from the posterior predictive distributions. For efficacy outcomes, positive RDs indicate higher response rates than placebo; for safety outcomes, positive RDs indicate higher event rates.

Study design and language: eligible studies were phase III randomized, placebo-controlled trials. We did not include phase I, II, II/III, or IV studies. This restriction was prespecified to improve comparability across interventions, because phase III trials are more likely to use confirmatory trial designs, regulatory-relevant dosing regimens, and consistent induction-phase assessment windows. Earlier-phase and dose-ranging studies were excluded because they frequently include exploratory doses, variable dosing schedules, and smaller or more selected populations. Including such studies could introduce additional clinical and design heterogeneity into an already sparse evidence network. Multiple publications or reports from the same trial were treated as a single study to avoid double counting. When overlapping reports were identified, we extracted data from the most complete and most recent source available for each prespecified outcome.

Data collection and risk of bias appraisal

Two reviewers independently assessed the risk of bias of each included trial using the Cochrane Risk of Bias 2 (RoB 2) tool. The assessed domains included the randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. Each domain and the overall judgment were rated as low risk, some concerns, or high risk according to RoB 2 guidance, with disagreements resolved through discussion with a third reviewer. We collected details on study design, treatment protocols, patient characteristics, and clinical outcomes. To support the ML-NMR analysis, we specifically extracted baseline characteristics for all trial arms. We used IPD when available; otherwise, we extracted the reported AgD from the publications.

Efficacy and safety data were extracted at the end of the induction period, which was the pre-specified timepoint for all outcomes in this analysis. If specific numbers were not listed in the text or tables, we extracted data points from figures using digitization software.

Certainty of evidence

The certainty of evidence for the PAICs was evaluated using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework adapted for NMA. We assessed the certainty of evidence across five distinct domains: study limitations (risk of bias), inconsistency, indirectness, imprecision, and publication bias.

Study limitations were informed by the overall risk of bias assessments using the RoB 2 tool for the trials contributing to the respective network nodes (9). Given the sparse nature of our evidence network, which lacked closed loops, formal statistical testing for global inconsistency (discrepancies between direct and indirect evidence) was not feasible; therefore, judgments on inconsistency primarily relied on the evaluation of clinical and methodological heterogeneity. Indirectness was evaluated based on the transitivity assumption. The application of the ML-NMR approach aimed to mitigate indirectness related to aggregation bias by adjusting for cross-trial baseline disparities and aligning AgD with the target IPD population. Imprecision was judged based on the width of the 95% CrIs and whether these intervals crossed the line of no effect. Finally, publication bias was evaluated qualitatively, as the small number of included studies precluded the use of formal statistical tests (e.g., funnel plots).

The overall certainty of evidence for each outcome comparison was ultimately categorized into one of four levels: high, moderate, low, or very low.

Statistical analysis

All analyses were conducted using the multinma R package, version 0.9.1, within a Bayesian framework (6,7). Binary efficacy outcomes (PASI 75, PASI 90, sPGA 0/1, and DLQI 0/1) and safety outcomes (AEs and SAEs) were modelled using binomial likelihoods with a logit link. Placebo was used as the common reference treatment.

The ML-NMR model adjusted for two baseline covariates selected a priori: body mass index (BMI) and prior biologic exposure. BMI was divided by 10 before model fitting to improve numerical stability, and prior biologic exposure was modelled as a binary covariate. Treatment-covariate interactions were specified at the treatment-class level. For DLQI 0/1, patients with baseline DLQI scores of 0 or 1 were excluded from the analysis. The two picankibart dose groups were pooled for induction-phase analyses because both groups received the same induction regimen before week 16.

A fixed-effect ML-NMR model was used because the evidence network was sparse, contained no closed loops, and included few studies per treatment comparison. Under these conditions, a random-effects variance would be weakly identified and likely driven by prior assumptions rather than by the data. Clinical heterogeneity was addressed by adjusting for the two prespecified effect modifiers in the ML-NMR framework, although residual unmeasured heterogeneity cannot be excluded.

Population-adjusted estimates were generated for a target population defined by the covariate distribution of participants in the picankibart trial. The placebo baseline response was specified as a normal distribution on the log-odds scale centered on the observed IPD placebo rate with a 0.5 continuity correction and a standard deviation of 0.1. For DLQI 0/1, the pooled aggregate-data placebo rate was used because of the restricted baseline DLQI eligibility. Missing aggregate covariate summaries were imputed using weighted averages across available arms. Integration over aggregate-data covariate distributions used the empirical correlation structure estimated from the picankibart IPD, with 100 quadrature points per arm.

Weakly informative priors were assigned to model parameters: normal N(0, 1.5²) priors for intercepts and treatment effects, and normal N(0, 0.5²) priors for regression coefficients. Four Markov chain Monte Carlo chains were run with 8,000 iterations each, including 2,500 warm-up iterations. Convergence was assessed using trace plots and the potential scale reduction statistic (R-hat), with R-hat <1.05 considered acceptable.

RRs and RDs were derived from posterior predictive distributions and summarized as posterior medians with 95% CrIs. RRs were used as the primary effect measure. RDs were reported to aid clinical interpretation of absolute benefit or harm. In the Bayesian framework, a treatment difference was considered credible when the 95% CrI excluded the null value, defined as RR =1.0 for relative effects or RD =0 for absolute effects. This criterion is analogous to, but not identical with, a two-sided frequentist significance threshold of α=0.05.

Results

Study selection

The study selection process is summarized in Figure 1. The database search identified 3,178 citations, of which 1,617 unique records remained after duplicate removal. After title and abstract screening, 156 full-text articles were assessed for eligibility. Finally, eight randomized controlled trials (RCTs) reported in seven publications were included in the NMA (4,10-15).

Figure 1 PRISMA flow diagram of the literature search and study selection process. CNKI, China National Knowledge Infrastructure.

Study characteristics and risk of bias

The eight included trials were published between 2017 and 2024 and included 3,099 randomized patients. Of these, 401 received picankibart, 950 received guselkumab, 726 received tildrakizumab, and 1,022 received placebo. Five trials were multicenter international studies, two were conducted in China, and one was conducted in Japan. The active interventions included guselkumab 100 mg every 8 weeks, tildrakizumab 100 mg every 12 weeks, and picankibart 100 or 200 mg every 12 weeks.

Across aggregate-data sources, mean or median age ranged from 39.0 to 48.3 years, the proportion of male patients ranged from 64.5% to 84.4%, mean BMI ranged from 25.4 to 31.5 kg/m², and psoriasis duration ranged from 12.7 to 19.1 years. Baseline characteristics are summarized in Table S2.

During induction, active-treatment PASI 75 and PASI 90 response rates ranged from 61.2% to 94.3% and from 34.5% to 80.3%, respectively. Corresponding placebo response rates ranged from 0.0% to 12.7% for PASI 75 and from 0.0% to 2.9% for PASI 90. Overall AE rates in active-treatment arms ranged from 44.3% to 66.8%, and SAE rates ranged from 0.0% to 3.2% (Table S3).

Seven of the eight trials were judged to have low overall risk of bias (Table S4). One trial was rated as having some concerns because of incomplete reporting of selected secondary endpoints that were not prespecified in the original protocol registration. No trial was judged to be at high overall risk of bias. Because this concern was limited to selective reporting of secondary outcomes and did not involve randomization, deviations from intended interventions, missing outcome data, or outcome measurement, no formal risk-of-bias-based sensitivity analysis was performed.

Clinical efficacy and safety

Induction-phase efficacy and safety versus placebo

At the end of induction, all three IL-23 inhibitors were associated with substantially higher efficacy response rates than placebo across PASI 75, PASI 90, sPGA 0/1, and DLQI 0/1 outcomes (Figure 2). The treatment effects were most consistent for PASI 75, PASI 90, and sPGA 0/1, for which credible intervals clearly excluded the null value.

Figure 2 Efficacy and safety outcomes of IL-23 inhibitors compared with placebo during the induction phase. Estimates of RRs and 95% CrIs derived from a fixed-effect ML-NMR model adjusted for two prespecified baseline covariates (BMI and prior biologic exposure) and referenced to the target population covariate distribution are presented for all interventions. For efficacy outcomes (PASI 75, PASI 90, sPGA 0/1, and DLQI 0/1), an RR >1 indicates a favorable effect for the biologic therapy. Conversely, for safety outcomes (AE and SAE), an RR <1 indicates a lower risk of events compared with placebo. Treatments are sorted in descending order based on the magnitude of the RR for PASI 90 response. 95% CrI, 95% credible interval; AE, adverse event; BMI, body mass index; DLQI, Dermatology Life Quality Index; IL-23, interleukin-23; ML-NMR, multilevel network meta-regression; PASI, Psoriasis Area and Severity Index; RR, risk ratio; SAE, serious adverse event; sPGA, static Physician Global Assessment.

For PASI 90, the population-adjusted risk ratios versus placebo were 25.88 for picankibart, 25.49 for guselkumab, and 13.05 for tildrakizumab. The corresponding absolute RDs were 63.6%, 62.6%, and 30.8%, respectively, suggesting similar induction-phase benefit for picankibart and guselkumab and a smaller effect for tildrakizumab. A similar pattern was observed for PASI 75 and sPGA 0/1, with picankibart and guselkumab showing comparable absolute benefits and tildrakizumab showing smaller, although still substantial, improvements versus placebo (Table 1). For DLQI 0/1, all active treatments were also superior to placebo, although uncertainty was greater. The absolute RDs for DLQI 0/1 were 60.9% for picankibart, 54.2% for guselkumab, and 28.3% for tildrakizumab. Overall AE rates were not credibly different from placebo. The risk ratios for any AE were close to unity for picankibart, guselkumab, and tildrakizumab, and the corresponding RDs were small with credible intervals including zero. SAEs were uncommon across trials, leading to imprecise estimates with wide credible intervals. Although tildrakizumab showed a lower estimated SAE risk than placebo, the absolute difference was less than 0.5 percentage points and should be interpreted cautiously because of sparse events.

The GRADE certainty of evidence was moderate for PASI 75, PASI 90, and sPGA 0/1, mainly because of indirectness inherent in the treatment comparisons. Certainty was lower for DLQI 0/1 because of wider credible intervals, and low for SAEs because of sparse events and imprecision (Table S5).

Population-adjusted indirect head-to-head comparisons

PAICs were performed using the picankibart trial population as the target population. Picankibart showed similar induction-phase efficacy to guselkumab across PASI 75, PASI 90, sPGA 0/1, and DLQI 0/1. For PASI 90, the risk ratio for picankibart versus guselkumab was 1.02, and the placebo-adjusted absolute benefits were nearly identical: 63.6% for picankibart and 62.6% for guselkumab. In contrast, picankibart showed higher efficacy than tildrakizumab across multiple outcomes. The risk ratio for PASI 90 was 1.97 for picankibart versus tildrakizumab, with similar advantages observed for PASI 75, sPGA 0/1, and DLQI 0/1 (Table 2). For safety outcomes, no credible differences were detected between picankibart and either active comparator. Any AE rates were similar across treatments, and SAE comparisons were highly uncertain because of the small number of events. All head-to-head relative estimates are summarized in Table 2. Population-adjusted placebo-referenced absolute RDs are provided in Table 1. The evidence network of the included placebo-controlled trials is shown in Figure 3.

Figure 3 Evidence network of included placebo-controlled phase III randomized trials for induction-phase comparisons. Node size is proportional to the number of randomized participants, and edge labels indicate the number of included placebo-controlled phase III trials contributing to each active treatment versus placebo comparison.

Discussion

Managing moderate-to-severe plaque psoriasis increasingly requires class-specific comparative data to guide clinical and formulary decisions. By utilizing an IPD-anchored ML-NMR framework, this study provides a robust indirect evaluation of picankibart against its established p19 class alternatives. During the induction phase, picankibart demonstrated superior PASI 90 and PASI 75 response rates compared with tildrakizumab and maintained higher numerical point estimates than guselkumab. The short-term safety profile of picankibart was consistent with both placebo and active comparators, with no increased risk of AEs or SAEs.

These findings are consistent with the established efficacy hierarchy of IL-23 inhibitors reported in previous comprehensive network meta-analyses (16-18). In these unadjusted indirect comparisons, guselkumab consistently ranks among the most efficacious biologic agents for short-term psoriasis clearance, whereas tildrakizumab typically demonstrates lower PASI response rates. Recent systematic reviews incorporating newer oral therapies, such as deucravacitinib, further confirm the high efficacy ceiling maintained by injectable IL-23 targeted biologics (19,20). Our population-adjusted estimates corroborate this class-level trend while positioning picankibart alongside guselkumab in terms of clinical response.

Clinically, picankibart and tildrakizumab are administered as a 12-week (Q12W) maintenance regimen, contrasting with the 8-week (Q8W) interval required for guselkumab. The data suggest that picankibart provides the dosing convenience of a Q12W schedule without the compromise in short-term efficacy typically observed with tildrakizumab. From a healthcare decision-making perspective, this combination of high efficacy and extended dosing may have implications for patient adherence, treatment satisfaction, and healthcare resource utilization, particularly in settings where injection frequency affects patient compliance.

A primary methodological consideration in this analysis is the use of ML-NMR to address cross-trial heterogeneity. Traditional NMAs rely heavily on the transitivity assumption, which can be compromised by variations in patient baseline characteristics, including body mass index and prior biologic exposure, across different trials (17). By integrating IPD from the picankibart trial with AgD from comparator studies, the ML-NMR approach adjusted for these two prespecified patient-level effect modifiers. This adjustment mitigates aggregation bias and provides estimates tailored to the baseline profile of the target clinical cohort, offering a more precise comparison than standard unadjusted models.

Several limitations should be considered when interpreting these results. First, the strict inclusion criteria—focusing exclusively on phase III trials of NMPA-approved IL-23 p19 inhibitors administered at standard doses—resulted in a sparse evidence network. This deliberate choice prioritized internal validity and dose-level homogeneity but limited the number of contributing studies. The absence of closed loops prevented formal inconsistency testing between direct and indirect evidence, and the limited number of studies necessitated the use of a fixed-effect ML-NMR model rather than a random-effects framework that would have accommodated residual heterogeneity beyond what the covariate adjustment addresses. Second, the analysis was restricted to the induction phase. Because placebo-controlled trials typically cross over placebo-assigned patients to active treatment after the induction period, long-term comparative efficacy could not be evaluated. Third, although the model adjusted for two prespecified clinical covariates (BMI and prior biologic exposure), the potential impact of other measured or unmeasured cross-trial differences (e.g., baseline PASI score, disease duration, concomitant medications, genetic polymorphisms, or geographic practice patterns) cannot be completely excluded. Fourth, formal publication bias assessment was limited to qualitative evaluation given the small number of studies per comparison; however, the comprehensive search across five databases (including two Chinese-language databases) and the inclusion of all identified eligible trials minimize this concern. Finally, IPD were available only from the picankibart trial; the availability of IPD from all included studies would have permitted even more refined population adjustment.

Conclusions

The IPD-anchored ML-NMR analysis indicates that picankibart is an effective and safe treatment for moderate-to-severe plaque psoriasis, demonstrating superior induction-phase efficacy compared to tildrakizumab and comparable efficacy to guselkumab. By maintaining high clinical response rates with a Q12W dosing interval, picankibart offers a practical therapeutic option within the IL-23 inhibitor class. These population-adjusted estimates can help inform clinical decision-making and formulary evaluations, particularly in the Chinese healthcare context where all three agents are available. Direct head-to-head trials and long-term real-world studies are needed to confirm these indirect findings and assess durability of response, maintenance-phase comparative effectiveness, and long-term safety.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the PRISMA-NMA reporting checklist. Available at https://amj.amegroups.com/article/view/10.21037/amj-2026-0041/rc

Peer Review File: Available at https://amj.amegroups.com/article/view/10.21037/amj-2026-0041/prf

Funding: This study was supported by the Clinical Research Project of Gansu Provincial Hospital (project No. ZX-62000001-2023-331), and Hospital Pharmacy Special Research Funding of Zhejiang Pharmaceutical Association (No. 2023ZYY37).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://amj.amegroups.com/article/view/10.21037/amj-2026-0041/coif). All authors report funding support from the Clinical Research Project of Gansu Provincial Hospital (project No. ZX-62000001-2023-331), and Hospital Pharmacy Special Research Funding of Zhejiang Pharmaceutical Association (No. 2023ZYY37). The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Yuan F, Cao L, Lu L, et al. Epidemiological Analysis of Psoriasis in China from 2020 to 2022. J Cutan Med Surg 2025;29:29-32. [Crossref] [PubMed]
2. Armstrong A, Bohannan B, Mburu S, et al. Impact of Psoriatic Disease on Quality of Life: Interim Results of a Global Survey. Dermatol Ther (Heidelb) 2022;12:1055-64. [Crossref] [PubMed]
3. Shi Y. Chinese Guidelines for the Diagnosis and Treatment of Psoriasis (2023 Edition). Journal of Tongji University (Medical Edition) 2023;44:631-3.
4. Shi Y, Qin L, Li Y, et al. Efficacy and safety of picankibart in moderate-to-severe plaque psoriasis: A randomized, double-blind, Phase 3 CLEAR-1 trial. Br J Dermatol 2024;191:iii46-iii47.
5. Phillippo D, Ades T, Dias S, et al. NICE DSU Technical Support Document 18: Methods for population-adjusted indirect comparisons in submissions to NICE. Decision Support Unit, ScHARR, University of Sheffield: NICE Decision Support Unit; 2016.
6. Phillippo DM, Dias S, Ades AE, et al. Multilevel network meta-regression for population-adjusted treatment comparisons. J R Stat Soc Ser A Stat Soc 2020;183:1189-210. [Crossref] [PubMed]
7. Bayesian network meta-analysis of individual and aggregate data • multinma [Internet]. [cited 2025 Nov 24]. Available online: https://dmphillippo.github.io/multinma/
8. Page MJ, McKenzie JE, Bossuyt PM, et al. Updating guidance for reporting systematic reviews: development of the PRISMA 2020 statement. J Clin Epidemiol 2021;134:103-12. [Crossref] [PubMed]
9. Sterne JAC, Savović J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019;366:l4898. [Crossref] [PubMed]
10. Blauvelt A, Papp KA, Griffiths CE, et al. Efficacy and safety of guselkumab, an anti-interleukin-23 monoclonal antibody, compared with adalimumab for the continuous treatment of patients with moderate to severe psoriasis: Results from the phase III, double-blinded, placebo- and active comparator-controlled VOYAGE 1 trial. J Am Acad Dermatol 2017;76:405-17. [Crossref] [PubMed]
11. Ferris LK, Ott E, Jiang J, et al. Efficacy and safety of guselkumab, administered with a novel patient-controlled injector (One-Press), for moderate-to-severe psoriasis: results from the phase 3 ORION study. J Dermatolog Treat 2020;31:152-9. [Crossref] [PubMed]
12. Ohtsuki M, Kubo H, Morishima H, et al. Guselkumab, an anti-interleukin-23 monoclonal antibody, for the treatment of moderate to severe plaque-type psoriasis in Japanese patients: Efficacy and safety results from a phase 3, randomized, double-blind, placebo-controlled study. J Dermatol 2018;45:1053-62. [Crossref] [PubMed]
13. Reich K, Armstrong AW, Foley P, et al. Efficacy and safety of guselkumab, an anti-interleukin-23 monoclonal antibody, compared with adalimumab for the treatment of patients with moderate to severe psoriasis with randomized withdrawal and retreatment: Results from the phase III, double-blind, placebo- and active comparator-controlled VOYAGE 2 trial. J Am Acad Dermatol 2017;76:418-31. [Crossref] [PubMed]
14. Reich K, Papp KA, Blauvelt A, et al. Tildrakizumab versus placebo or etanercept for chronic plaque psoriasis (reSURFACE 1 and reSURFACE 2): results from two randomised controlled, phase 3 trials. Lancet 2017;390:276-88. [Crossref] [PubMed]
15. Yu C, Geng S, Yang B, et al. Tildrakizumab for moderate-to-severe plaque psoriasis in Chinese patients: A 12-week randomized placebo-controlled phase III trial with long-term extension. Chin Med J (Engl) 2024;137:1190-8. [Crossref] [PubMed]
16. Bai F, Li GG, Liu Q, et al. Short-Term Efficacy and Safety of IL-17, IL-12/23, and IL-23 Inhibitors Brodalumab, Secukinumab, Ixekizumab, Ustekinumab, Guselkumab, Tildrakizumab, and Risankizumab for the Treatment of Moderate to Severe Plaque Psoriasis: A Systematic Review and Network Meta-Analysis of Randomized Controlled Trials. J Immunol Res 2019;2019:2546161. [Crossref] [PubMed]
17. Sbidian E, Chaimani A, Afach S, et al. Systemic pharmacological treatments for chronic plaque psoriasis: a network meta-analysis. Cochrane Database Syst Rev 2020;1:CD011535. [Crossref] [PubMed]
18. Lebwohl MG, Carvalho A, Asahina A, et al. Biologics for the Treatment of Moderate-to-Severe Plaque Psoriasis: A Systematic Review and Network Meta-analysis. Dermatol Ther (Heidelb) 2025;15:1633-56. [Crossref] [PubMed]
19. Armstrong AW, Warren RB, Zhong Y, et al. Short-, Mid-, and Long-Term Efficacy of Deucravacitinib Versus Biologics and Nonbiologics for Plaque Psoriasis: A Network Meta-Analysis. Dermatol Ther (Heidelb) 2023;13:2839-57. [Crossref] [PubMed]
20. Chen J, Yin X, Kan X, et al. A systematic review and network meta-analysis comparing the efficacy and safety of deucravacitinib versus selected treatments for moderate to severe plaque psoriasis. Clin Rheumatol 2025;44:4025-32. [Crossref] [PubMed]