Dermatol Ther (Heidelb) (2022) 12:167–184 https://doi.org/10.1007/s13555-021-00647-0 ORIGINAL RESEARCH Long-Term Benefit–Risk Profiles of Treatments for Moderate-to-Severe Plaque Psoriasis: A Network Meta-Analysis April W. Armstrong . Ahmed M. Soliman . Keith A. Betts . Yan Wang . Yawen Gao . Vassilis Stakias . Luis Puig Received: October 11, 2021 / Accepted: November 12, 2021 / Published online: December 4, 2021 Ó The Author(s) 2021 ABSTRACT identified through a systematic literature review through 2 May 2021. Bayesian network meta- Introduction: The long-term benefit–risk pro- analyses (NMAs) were conducted to compare files of licensed and investigational treatments the efficacy (Psoriasis Area and Severity Index for moderate-to-severe plaque psoriasis have [PASI] 75/90/100 [at least a 75/90/100% reduc- not been fully characterized. tion in PASI score from baseline] response) and Methods: Randomized controlled trials (RCTs) safety outcomes (any adverse event [AE], any of licensed and investigational treatments for serious AE [SAE], and AEs leading to treatment moderate-to-severe plaque psoriasis were discontinuation) of each treatment evaluated between weeks 48 and 56 after baseline. Surfaces under the cumulative ranking curves (SUCRAs) Supplementary Information The online version were calculated to evaluate the relative ranking contains supplementary material available at https:// doi.org/10.1007/s13555-021-00647-0. of treatments. The benefit–risk profiles of treat- ments were assessed by bidimensional plots of A. W. Armstrong (&) the NMA-estimated efficacy and safety Department of Dermatology, Keck School of outcomes. Medicine, University of Southern California, HC4 Results: In the efficacy NMA (N = 14 RCTs), the 2000 1450 San Pablo, Health Sciences Campus, Los relative rankings for PASI 75/90/100 responses Angeles, CA 90033, USA e-mail: aprilarmstrong@post.harvard.edu by weeks 48–56 were the highest for risankizu- mab (SUCRA: 98.5%) and bimekizumab (83.8% A. M. Soliman for dosing every 4 weeks [Q4W], 72.7% for AbbVie, Inc., 1 N. Waukegan Road, North Chicago, IL 60064, USA dosing Q4W then every 8 weeks). The PASI response rates did not differ significantly K. A. Betts  Y. Wang  Y. Gao between risankizumab and the two bimek- Analysis Group, Inc., 333 South Hope Street 27th izumab regimens. Additionally, risankizumab Floor, Los Angeles, CA 90071, USA was associated with significantly higher PASI V. Stakias response rates than brodalumab, guselkumab, AbbVie, Inc., 26525 N. Riverwoods Blvd, Mettawa, ixekizumab, secukinumab, ustekinumab, adali- IL 60045, USA mumab, and etanercept. In the safety NMAs L. Puig (N = 8 RCTs), risankizumab had the highest Department of Dermatology, Hospital de la Santa relative rankings for all three outcomes (SUCRA: Creu i Sant Pau, Universitat Autònoma de 92.1%, 82.0%, and 91.0% for any AE, any SAE, Barcelona, Carrer de Sant Quintı́, 89, 08041 Barcelona, Spain and AEs leading to treatment discontinuation, 168 Dermatol Ther (Heidelb) (2022) 12:167–184 respectively). Risankizumab had a significantly Although ixekizumab and bimekizumab lower rate of any AE than bimekizumab, ustek- had favorable efficacy profiles, both inumab, and secukinumab. treatments had lower rankings for safety Conclusions: Risankizumab was associated outcomes; thus, risankizumab was with the most favorable long-term benefit–risk associated with the most favorable long- profile for the treatment of moderate-to-severe term benefit–risk profile for the treatment plaque psoriasis. Although ixekizumab and of moderate-to-severe plaque psoriasis. bimekizumab had favorable efficacy profiles, both treatments had lower rankings for safety outcomes. INTRODUCTION Keywords: Long-term benefit–risk profile; Risankizumab; Bimekizumab; Ixekizumab; Psoriasis, an inflammatory skin disease, is esti- Network meta-analysis; Moderate-to-severe mated to affect approximately 0.84% of the plaque psoriasis global population [1], with higher prevalence in Australasia, Europe, and North America [2, 3]. Key Summary Points Symptoms of plaque psoriasis are multifactorial and vary in severity [4], necessitating a tailored Why carry out this study? approach to therapy. Mild-to-moderate plaque psoriasis can often be managed with topical Prior research has established that the therapies and phototherapy, while biologic long-term benefits and risks associated agents have high efficacy for moderate-to-severe with psoriasis treatments are important disease [5, 6]. Several novel biologic agents tar- considerations during treatment decision- geting the interleukin (IL)-17 pathway (ixek- making. izumab, brodalumab, secukinumab) and the IL- 23 pathway (risankizumab, guselkumab, til- The long-term benefit–risk profiles of drakizumab) have been recently approved by licensed and investigational treatments the United States (US) Food and Drug Admin- for moderate-to-severe plaque psoriasis istration (FDA) and the European Medicines have not been fully characterized. Agency (EMA) for the treatment of moderate-to- This study indirectly compared the long- severe plaque psoriasis. Furthermore, an inves- term efficacy and safety outcomes of tigational anti-IL-17 agent, bimekizumab, has licensed and investigational treatments shown promising data in several phase 3 clinical for moderate-to-severe plaque psoriasis trials (BE VIVID, BE READY, BE SURE, and BE based on network meta-analyses of RADIANT) [7–10] and was recently licensed by clinical trial data and characterized their the EMA [11]. long-term benefit–risk profiles. As there is no cure for psoriasis, long-term treatments and ongoing disease management What was learned from this study? are critical to control psoriasis symptoms and Risankizumab and bimekizumab had the related comorbidities [12–15]. Prior research has highest long-term Psoriasis Area and established that the long-term benefits and risks Severity Index response rates, and associated with psoriasis treatments are impor- risankizumab had the lowest long-term tant considerations during treatment decision- rates of safety events. making from both patients’ and physicians’ perspectives. For example, a systematic review by Sain et al. [16] reported that maintenance of treatment response was the most important attribute in several studies of the preferences of patients with psoriasis, and that the long-term Dermatol Ther (Heidelb) (2022) 12:167–184 169 risks of severe adverse events (AEs) such as the NMAs of clinical trial data and characterized 10-year risks of tuberculosis, lymphoma, and their long-term benefit–risk profiles. serious infections were of primary concerns to both patients and physicians. Data from both clinical trials and the real METHODS world have been routinely used to characterize and compare the long-term benefit–risk profiles Data Source of treatments for moderate-to-severe psoriasis. For example, Shear et al. [17] characterized the Phase 2, 3, or 4 randomized controlled trials benefit–risk profiles of novel treatments by (RCTs) of licensed (by the US FDA or the EMA) weeks 48–56 using a bidimensional plot of and investigational treatments for moderate-to- meta-analyzed long-term Psoriasis Area and severe plaque psoriasis were identified through Severity Index (PASI) response rates and long- a systematic literature review (SLR) through 2 term safety event rates based on network meta- May 2021. This SLR was originally conducted analyses (NMAs). The study found that risanki- on 4 December 2017 and updated on 17 zumab was associated with the most favorable September 2018 (reported by Armstrong et al. long-term benefit–risk profile compared with [20]), 19 February 2021, and 2 May 2021. The guselkumab, ixekizumab, secukinumab, ustek- searched databases include Embase, MEDLINE, inumab, and adalimumab [17]. Additionally, a and the Cochrane library. Additional searches real-world retrospective multi-country, multi- were conducted for the reference lists of inclu- cohort study by Torres et al. [18] compared drug ded studies, conference proceedings, previous survival, a composite metric capturing both health technology assessment submissions, and efficacy and safety, among biologic agents for clinical trial registries. moderate-to-severe psoriasis. The authors found The trial inclusion and exclusion criteria that the 18-month drug survival of the anti-IL- reported by Armstrong et al. [20], Armstrong 23 agents risankizumab and guselkumab was et al. [21], and Shear et al. [17] were modified to higher than that of ustekinumab, while that of accommodate both licensed and investigational secukinumab was lower versus ustekinumab treatments for moderate-to-severe plaque pso- [18]. In a systematic review and comparative riasis. The trials were required to: (1) be a phase meta-analysis of drug survival among biologics 2, 3, or 4 RCT for moderate-to-severe plaque for psoriasis, Mourad and Gniadecki [19] found psoriasis among adults who were eligible for that ustekinumab was associated with longer systemic therapies or phototherapy; (2) include drug survival than anti-tumor necrosis factor licensed treatments and dosages by the US FDA (TNF) and anti-IL-17 agents. However, there or the EMA or investigational treatments and were insufficient data for guselkumab, the only dosages assessed in phase 3 clinical trials; and anti-IL-23 agent considered in the study, to (3) report at least one of the efficacy outcomes facilitate further comparisons. (PASI 75, 90, and 100; indicating the propor- The clinical decision-making process in the tions of patients who achieved at least a 75, 90, real world requires clinicians to evaluate mul- or 100% reduction in PASI score from baseline) tiple therapeutic options simultaneously. To or safety outcomes (any AEs, any serious AEs inform this decision-making process, clinical [SAEs], and AEs leading to treatment discon- trials provide a rich set of data on both licensed tinuation) of interest by the end of the main- and investigational treatments, which can be tenance period (48–56 weeks from baseline). synthesized using established indirect compar- RCTs in which patients were crossed over from ison methods to ensure a comprehensive com- one treatment to another before the end of parison of the outcomes across treatments. In weeks 48–56, or were re-randomized based on the present study, we indirectly compared the certain efficacy criteria during the post-induc- long-term efficacy and safety outcomes of tion period, were excluded (for example, the licensed and investigational treatments for VOYAGE-2 study was excluded because it moderate-to-severe plaque psoriasis based on 170 Dermatol Ther (Heidelb) (2022) 12:167–184 rerandomized PASI 90 responders at week 28 to Statistical Analyses receive guselkumab or placebo [22]). As this is a post-hoc NMA of previously NMA Models published results of clinical trial data, no insti- Separate NMA models were fitted for the effi- tutional board review was required. The study cacy and safety outcomes. Specifically, a fixed- was performed in accordance with the Helsinki effects Bayesian probit NMA was implemented Declaration 1964 and its later amendments. to jointly model the PASI 75, 90, and 100 This article is based on previously conducted response rates, and fixed-effects Bayesian logis- studies and does not contain any new studies tic NMAs were applied to individually model with human participants or animals performed the rates of any AE, any SAE, and AEs leading to by any of the authors. treatment discontinuation [23]. The fixed-ef- fects models were selected because of the rela- Comparators tive lack of long-term randomized controlled data that could be included into the network. The study included regimens for anti-IL-23 Based on the efficacy and safety NMA mod- agents, anti-IL-17 agents, anti-TNF agents, and els, the proportions of patients achieving each an anti-IL-12/23 agent through 48–56 weeks of the efficacy and safety outcomes were after randomization. The anti-IL-23 agents described using posterior medians and 95% included guselkumab (100 mg at weeks 0 and 4, credible intervals (CrIs). Furthermore, the rela- then every 8 weeks [Q8W]) and risankizumab tive ranking of each treatment was assessed (150 mg at weeks 0 and 4, then every 12 weeks using the surface under the cumulative ranking [Q12W]). The anti-IL-17 agents included curve (SUCRA) [24], a summary statistic ranging bimekizumab (320 mg every 4 weeks [Q4W] or from 0% to 100%. In the efficacy NMA, a higher 320 mg at weeks 0, 4, 8, 12, and 16, then Q8W); SUCRA value indicated that a treatment had a brodalumab (210 mg at weeks 0, 1, and 2, then higher likelihood to have the highest PASI every 2 weeks [Q2W]); ixekizumab (160 mg at response rate among all treatments, i.e., the week 0, 80 mg Q2W until week 12, then 80 mg most efficacious treatment. In the safety NMAs, Q4W); and secukinumab (300 mg at weeks 0, 1, a higher SUCRA value suggested that a treat- 2, 3, and 4, then Q4W). The anti-TNF agents ment had a higher likelihood to have the lowest included adalimumab (80 mg at week 0, then safety event rates, i.e., the safest treatment. 40 mg Q2W starting week 1) and etanercept While the efficacy and safety of each treatment (50 mg twice weekly [BIW] until week 12, then can be compared numerically using the poste- weekly [QW]). Finally, the anti-IL-12/23 agent rior median for the proportion and the SUCRA was ustekinumab with a weight-based dosage value, neither measure can provide a formal (45 mg B 100 kg, 90 mg[100 kg at weeks 0 and statistical comparison between each pair of 4, then Q12W). treatments. Therefore, odds ratios were used to formally compare the proportions between each pair of treatments and were summarized using Outcomes posterior medians and 95% CrIs. Statistical sig- nificance was determined based on whether the The study examined both efficacy and safety 95% CrIs of odds ratios excluded 1. outcomes by the end of the pre-specified maintenance period (weeks 48–56 after base- Computation line). The efficacy outcomes were the propor- A Markov Chain Monte Carlo (MCMC) proce- tions of patients with PASI 75, 90, and 100 dure was used to draw the posterior samples for responses; the safety outcomes were the pro- the Bayesian NMA models. The MCMC proce- portions of patients who experienced any AE, dure was implemented with 5000 adaptation any SAE, or AEs leading to treatment iterations, 50,000 burn-in iterations, and 50,000 discontinuation. Dermatol Ther (Heidelb) (2022) 12:167–184 171 posterior sampling iterations. Three parallel Consistent with the PASI rankings, the chains were run with a thinning factor of 10. SUCRA values were the highest for risankizu- mab (98.5%), bimekizumab Q4W (83.8%), and Benefit–Risk Profiles bimekizumab Q4W then Q8W (72.7%). The Based on the NMAs, the benefit–risk profiles of SUCRA values were the lowest for etanercept each treatment were assessed through bidi- (0.1%), adalimumab (13.1%), and ustekinumab mensional plots of the SUCRA values of the (20.1%) (Table 1). PASI response rates and each of the safety out- There were no significant differences comes. Treatments located at the top right cor- between the PASI response rates of risankizu- ner of the graph were associated with the most mab, bimekizumab Q4W, and bimekizumab favorable benefit–risk profiles. Q4W then Q8W, with 95% probability (Table 2; R statistical software (version 3.6.2, the R ESM Table S2). However, risankizumab had Foundation for Statistical Computing, Vienna, significantly higher PASI response rates than Austria) and JAGS (version 4.3.0, Martyn brodalumab, guselkumab, ixekizumab, secuk- Plummer, University of Warwick, Coventry, inumab, ustekinumab, adalimumab, and etan- UK) were used for all statistical analyses. ercept. Bimekizumab Q4W had significantly higher PASI response rates than ixekizumab, secukinumab, ustekinumab, adalimumab, and RESULTS etanercept. Bimekizumab Q4W then Q8W, brodalumab, and guselkumab had significantly SLR and Study Selection higher PASI response rates than secukinumab, ustekinumab, adalimumab, and etanercept The SLR identified 689 publications for 215 (Table 2; ESM Table S2). RCTs through 2 May 2021 (Fig. 1). A total of 14 studies were included in the NMAs of long-term Long-term Safety efficacy and safety outcomes, while 201 were excluded. The data extracted from the studies In the long-term safety NMAs, a total of eight included in the long-term efficacy and safety trials for seven treatments were included NMAs are presented in Electronic Supplemen- (Fig. 2b). tary Material (ESM) Table S1. Any AE Long-term Efficacy The posterior median for the rate of any AE was the lowest for risankizumab (67.5%), followed A total of 14 RCTs connecting ten treatments by guselkumab (72.2%), adalimumab (72.9%), were included in the efficacy NMA (Fig. 2a). The secukinumab (76.6%), ustekinumab (76.9%), posterior median of the PASI response rates was ixekizumab (80.9%), and bimekizumab (two the highest for risankizumab (84.9% for PASI 90 dosages pooled) (82.3%). Consistent with these and 65.4% for PASI 100), bimekizumab 320 mg rankings, the SUCRA values were the highest for Q4W (81.3% and 59.9%, respectively), and risankizumab (92.1%) and were the lowest for bimekizumab Q4W then Q8W (79.4% and bimekizumab (two dosages pooled) (9.6%) and 57.3%, respectively), which were followed by ixekizumab (22.0%) (Table 1). brodalumab (78.6% and 56.1%, respectively), Risankizumab had a significantly lower rate guselkumab (77.3% and 54.4%, respectively), of any AE than secukinumab, ustekinumab, and ixekizumab (72.0% and 47.8%, respectively), bimekizumab (Table 3). Guselkumab had a sig- secukinumab (66.2% and 41.3%, respectively), nificantly lower rate of any AE than bimek- ustekinumab (55.1% and 30.5%, respectively), izumab. No other statistically significant adalimumab (50.8% and 26.9%, respectively), differences were identified. and etanercept (37.4% and 16.9%, respectively). 172 Dermatol Ther (Heidelb) (2022) 12:167–184 Fig. 1 Flow chart of study selection. e1 exclusion 1, e2 exclusion 2, i1 inclusion 1, i2 inclusion 2, i3 inclusion 3, NMA network meta-analysis Any SAE dosages pooled) (7.2%), and ixekizumab The posterior median for the rate of any SAE (10.5%). Consistent with these rankings, the was the lowest for risankizumab (posterior SUCRA values were the highest for risankizu- median: 4.4%), followed by adalimumab mab (82.0%) and the lowest for ixekizumab (5.4%), ustekinumab (5.7%), guselkumab (16.5%) and bimekizumab (two dosages pooled) (5.9%), secukinumab (6.9%), bimekizumab (two Dermatol Ther (Heidelb) (2022) 12:167–184 173 Fig. 2 Evidence network for the NMAs of PASI response b Safety events (any AE, any SAE, and AEs leading to (a) and safety events (b); any AE, any SAE, and AEs treatment discontinuation). The included trials were: leading to treatment discontinuation) by the end of the ECLIPSE [40], VOYAGE-1 [41], CLEAR [47], maintenance period (weeks 48–56). a PASI response. The CLARITY [44], IXORA-S [45], UltIMMa1 [46], included trials were: AMAGINE-2 [39], AMAGINE-3 UltIMMa2 [46], and BE RADIANT [10]. AE Adverse [39], ECLIPSE [40], VOYAGE-1 [41], CLEAR [42], event, BIW twice weekly, PASI Psoriasis Area and Severity FIXTURE [43], CLARITY [44], IXORA-S [45], Index, QW once every week, Q4W once every 4 weeks, UltIMMa1 [46], UltIMMa2 [46], IMMerge [37], BE Q8W once every 8 weeks, SAE serious adverse event VIVID [7], BE RADIANT [10], and BE SURE [9]. 174 Dermatol Ther (Heidelb) (2022) 12:167–184 (36.2%) (Table 1). No statistically significant profile compared with the other treatments for differences were identified (Table 3). moderate-to-severe psoriasis. While there have been many NMAs charac- AEs Leading to Treatment Discontinuation terizing the short-term efficacy and safety of The posterior median for the rate of AEs leading treatments for moderate-to-severe plaque pso- to treatment discontinuation was the lowest for riasis [12, 17, 20, 21, 25–34], only a few studies risankizumab (0.9%), followed by ustekinumab have examined those in the long term (2.2%), guselkumab (2.5%), secukinumab [17, 20, 21, 35, 36]. The results of the current (3.2%), adalimumab (3.4%), bimekizumab (two long-term efficacy and safety NMAs, including dosages pooled) (4.1%), and ixekizumab (4.3%). PASI response rates and safety event rates, are Consistent with these rankings, the SUCRA consistent with the findings of previously pub- values were the highest for risankizumab lished long-term NMAs and meta-analyses of (91.0%) and the lowest for ixekizumab (31.7%) licensed treatments for psoriasis. Specifically, and bimekizumab (two dosages pooled) (25.7%) prior studies have demonstrated that risanki- (Table 1). No statistically significant differences zumab was associated with the highest PASI were identified (Table 3). response rate and the lowest safety event rates [17, 20, 21, 35]. In addition, an NMA by Blau- Benefit–Risk Profiles velt et al. [36], comparing cumulative days at PASI 90 and 100 for biologic treatments for psoriasis, found that the cumulative days at Risankizumab had the top SUCRA values for PASI 90 and 100 were the highest for risanki- efficacy and safety outcomes, and thus had the zumab and ixekizumab, respectively. Further- most favorable benefit–risk profile (Fig. 3). more, the long-term benefit–risk profiles Guselkumab also had relatively high rankings of characterized in the present study are similar to both efficacy and safety outcomes. While the ones reported by Shear et al. [17], with bimekizumab and ixekizumab had favorable risankizumab being the most favorable. The benefit profiles, the safety rankings of both comparative long-term benefit–risk profiles in treatments were among the lowest. Adali- the present study based on clinical trial data are mumab, ustekinumab, and secukinumab had also similar to the ones reported by Torres et al. the lowest rankings of efficacy compared with [18] using drug survival as the metric, although the other treatments. that real-world study did not include investi- gational therapies. DISCUSSION Compared with the previously published long-term NMAs which focused on licensed This study utilized up-to-date clinical trial data treatments [17], the present study included to indirectly compare the long-term efficacy bimekizumab, an investigational treatment and safety of licensed and investigational ther- with safety and efficacy data available in the apies for moderate-to-severe plaque psoriasis publications from several phase 3 clinical trials and characterize their relative benefit–risk pro- [7–10], as one of the comparators. The PASI files. In the NMA of long-term efficacy, risan- response rate for bimekizumab was among the kizumab and bimekizumab were associated with highest, which was significantly higher than the top PASI response rates and relative rank- multiple previously licensed treatments for ings. Risankizumab was also associated with the psoriasis, including ixekizumab. Compared lowest rates and the highest relative rankings with the other treatments in the NMA, bimek- for any AE, any SAE, and AEs leading to treat- izumab had the lowest ranking for any AE and ment discontinuation in the NMA of long-term AEs leading to treatment discontinuation. The safety. These data suggest that risankizumab has NMA found that the rate of any AE for bimek- the most favorable long-term benefit–risk izumab was significantly higher than the rates for the two anti-IL-23 agents (risankizumab and guselkumab). Additionally, the BE RADIANT Table 1 Long-term (weeks 48–56) PASI 75/90/100 response and safety event rates (any AE, any SAE, and AEs leading to treatment discontinuation): posterior distribution of rates and SUCRA values based on the network meta-analyses PASI response Treatment Posterior median, % (95% CrI) SUCRAa PASI 75 PASI 90 PASI 100 Risankizumab 150 mg at weeks 0 and 4, then Q12W 92.9% 84.9% 65.4% 98.5% (90.5%, 94.9%) (80.9%, 88.3%) (59.4%, 71.0%) Bimekizumab 320 mg Q4W 90.8% 81.3% 59.9% 83.8% (87.7%, 93.3%) (76.5%, 85.4%) (53.3%, 66.2%) Dermatol Ther (Heidelb) (2022) 12:167–184 Bimekizumab 320 mg at weeks 0, 4, 8, 12, and 16, then Q8W 89.6% 79.4% 57.3% 72.7% (85.6%, 92.8%) (73.4%, 84.6%) (49.5%, 64.9%) Brodalumab 210 mg at weeks 0, 1, and 2, then Q2W 89.1% 78.6% 56.1% 68.5% (85.8%, 91.7%) (73.8%, 82.8%) (49.9%, 62.2%) Guselkumab 100 mg at weeks 0 and 4, then Q8W 88.2% 77.3% 54.4% 61.9% (84.3%, 91.4%) (71.6%, 82.3%) (47.4%, 61.4%) Ixekizumab 160 mg at week 0, 80 mg Q2W until week 12, then 80 mg Q4W 84.6% 72.0% 47.8% 46.6% (77.6%, 90.0%) (62.6%, 80.0%) (37.5%, 58.1%) Secukinumab 300 mg at weeks 0, 1, 2, 3, and 4, then Q4W 80.4% 66.2% 41.3% 34.6% (77.2%, 83.3%) (62.2%, 70.1%) (37.1%, 45.7%) Ustekinumab 45 mg B 100 kg, 90 mg [ 100 kg at weeks 0, and 4, then Q12W 71.4% 55.1% 30.5% 20.1% (69.3%, 73.4%) (52.9%, 57.2%) (28.4%, 32.7%) Adalimumab 80 mg at week 0, then 40 mg Q2W 67.7% 50.8% 26.9% 13.1% (58.4%, 75.9%) (41.2%, 60.4%) (19.4%, 35.5%) Etanercept 50 mg BIW until week 12, then QW 54.7% 37.4% 16.9% 0.1% (46.6%, 62.6%) (30.1%, 45.3%) (12.3%, 22.5%) 175 Table 1 continued 176 Safety events (any AE, any SAE, and AEs leading to treatment discontinuation) Treatment Any AE Any SAE AEs leading to treatment discontinuation Posterior median, % SUCRA Posterior median, % SUCRA Posterior median, % SUCRAa (95% CrI) (95% CrI) (95% CrI) Risankizumab 150 mg at weeks 0 and 4, then Q12W 67.5% 92.1% 4.4% 82.0% 0.9% 91.0% (57.8%, 75.5%) (2.4%, 8.0%) (0.2%, 4.0%) Guselkumab 100 mg at weeks 0 and 4, then Q8W 72.2% 75.0% 5.9% 56.7% 2.5% 58.0% (63.6%, 79.5%) (3.1%, 11.0%) (0.9%, 7.1%) Adalimumab 80 mg at week 0, then 40 mg Q2W 72.9% 67.6% 5.4% 62.1% 3.4% 38.1% (61.3%, 82.0%) (2.1%, 13.4%) (0.8%, 12.6%) Secukinumab 300 mg at weeks 0, 1, 2, 3, and 4, then 76.6% 42.9% 6.9% 35.2% 3.2% 39.1% Q4W (71.7%, 80.9%) (4.4%, 10.6%) (1.7%, 6.1%) Ustekinumab 45 mg B 100 kg, 90 mg [ 100 kg at 76.9% 40.7% 5.7% 61.3% 2.2% 66.4% weeks 0, and 4, then Q12W (74.5%, 79.2%) (4.5%, 7.1%) (1.5%, 3.2%) Ixekizumab 160 mg at week 0, 80 mg Q2W until week 80.9% 22.0% 10.5% 16.5% 4.3% 31.7% 12, then 80 mg Q4W (68.7%, 89.3%) (3.8%, 27.2%) (0.6%, 28.9%) Bimekizumab 320 mg at weeks 0, 4, 8, 12, and 16, then 82.3% 9.6% 7.2% 36.2% 4.1% 25.7% Q4W or Q8W (74.4%, 88.2%) (3.4%, 14.5%) (1.4%, 11.5%) AE Adverse event, BIW twice weekly, CrI credible interval, PASI Psoriasis Area and Severity Index, PASI 75, 90, 100 at least a 75%, 90%, or 100% decrease from baseline PASI, QW every week, Q2W every 2 weeks, Q4W every 4 weeks, Q8W every 8 weeks, Q12W every 12 weeks, SAE serious adverse event, SUCRA surface under the cumulative ranking curves a SUCRA measures the likelihood of a treatment being in the top ranks Dermatol Ther (Heidelb) (2022) 12:167–184 Table 2 Pairwise odds ratio of achieving PASI 90 and 100 response in the long term (weeks 48–56) PASI 90: posterior median (95% CrI) RIS 0.77 0.69 0.65 0.60 0.46 0.35 0.22 0.18 0.11 a a a a a a (0.53, 1.12) (0.46, 1.04) (0.45, 0.94) (0.42, 0.88) (0.28, 0.76) (0.26, 0.46) (0.17, 0.28) (0.12, 0.29) (0.07, 0.16)a 1.30 BKZ Q4W 0.89 0.84 0.79 0.59 0.45 0.28 0.24 0.14 a a a a (0.89, 1.89) (0.66, 1.22) (0.58, 1.23) (0.55, 1.13) (0.36, 0.99) (0.34, 0.59) (0.21, 0.37) (0.15, 0.37) (0.09, 0.20)a 1.46 1.12 BKZ Q4W then Q8W 0.95 0.88 0.66 0.51 0.32 0.27 0.15 a a a (0.97, 2.19) (0.82, 1.53) (0.62, 1.43) (0.60, 1.30) (0.39, 1.14) (0.37, 0.68) (0.23, 0.44) (0.17, 0.42) (0.10, 0.23)a 1.54 (1.06, 2.23)a 1.18 1.06 BRO 0.93 0.70 0.53 0.33 0.28 0.16 Dermatol Ther (Heidelb) (2022) 12:167–184 a a a (0.81, 1.72) (0.70, 1.60) (0.64, 1.36) (0.43, 1.15) (0.40, 0.72) (0.26, 0.43) (0.18, 0.44) (0.11, 0.24)a 1.65 1.27 1.14 1.08 GUS 0.75 0.58 0.36 0.30 0.18 a a a a (1.14, 2.40) (0.88, 1.83) 0.77, 1.68) (0.73, 1.57) (0.45, 1.26) (0.45, 0.73) (0.27, 0.48) (0.23, 0.40) (0.12, 0.25)a 2.20 1.69 1.51 1.43 1.33 IXE 0.76 0.48 0.40 0.23 a a a a (1.32, 3.63) (1.01, 2.79) (0.88, 2.57) (0.87, 2.33) (0.79, 2.21) (0.48, 1.19) (0.31, 0.73) (0.23, 0.71) (0.14, 0.39)a 2.87 2.21 1.97 1.87 1.74 1.31 SEC 0.63 0.53 0.31 a a a a a a a (2.18, 3.83) (1.69, 2.92) (1.46, 2.69) (1.40, 2.52) (1.37, 2.22) (0.84, 2.07) (0.54, 0.73) (0.37, 0.75) (0.23, 0.40)a 4.60 3.54 3.15 2.99 2.78 2.09 1.60 UST 0.84 0.49 a a a a a a a (3.52, 6.06) (2.70, 4.70) (2.28, 4.41) (2.33, 3.86) (2.09, 3.72) (1.38, 3.23) (1.38, 1.86) (0.58, 1.23) (0.36, 0.67)a 5.46 4.20 3.75 3.55 3.30 2.49 1.90 1.19 ADA 0.58 a a a a a a a (3.49, 8.58) (2.70, 6.56) (2.37, 5.98) (2.25, 5.61) (2.52, 4.33) (1.42, 4.43) (1.34, 2.69) (0.81, 1.74) (0.37, 0.90)a 9.43 7.26 6.47 6.13 5.70 4.30 3.28 2.05 1.73 ETA a a a a a a a a a (6.36, 14.06) (4.94, 10.74) (4.31, 9.82) (4.11, 9.20) (3.97, 8.25) (2.53, 7.33) (2.49, 4.33) (1.50, 2.81) (1.11, 2.69) 177 Table 2 continued 178 PASI 100: posterior median (95% CrI) RIS 0.79 0.71 0.68 0.63 0.48 0.37 0.23 0.19 0.11 (0.56, 1.11) (0.49, 1.03) (0.48, 0.95)a (0.45, 0.89)a (0.30, 0.78)a (0.29, 0.48)a (0.18, 0.30)a (0.12, 0.30)a (0.07, 0.16)a 1.27 BKZ Q4W 0.90 0.86 0.80 0.61 0.47 0.29 0.25 0.14 (0.90, 1.77) (0.68, 1.20) (0.61, 1.21) (0.57, 1.12) (0.38, 0.99)a (0.37, 0.61)a (0.23, 0.38)a (0.16, 0.38)a (0.09, 0.20)a 1.41 1.11 BKZ Q4W then Q8W 0.95 0.89 0.68 0.52 0.33 0.27 0.15 (0.97, 2.04) (0.84, 1.47) (0.65, 1.39) (0.62, 1.27) (0.41, 1.13) (0.39, 0.69)a (0.24, 0.44)a (0.17, 0.43)a (0.10, 0.23)a 1.48 (1.06, 2.07)a 1.17 1.05 BRO 0.93 0.72 0.55 0.34 0.29 0.16 (0.83, 1.65) (0.72, 1.54) (0.66, 1.33) (0.45, 1.14) (0.42, 0.73)a (0.27, 0.43)a (0.18, 0.45)a (0.10, 0.24)a 1.58 1.25 1.12 1.07 GUS 0.76 0.59 0.37 0.31 0.17 (1.13, 2.22)* (0.89, 1.75) (0.79, 1.61) (0.75, 1.52) (0.47, 1.24) (0.47, 0.74)a (0.28, 0.48)a (0.23, 0.41)a (0.12, 0.25)a 2.07 1.63 1.47 1.40 1.31 IXE 0.77 0.48 0.40 0.22 (1.29, 3.32)a (1.01, 2.63)a (0.89, 2.44) (0.88, 2.24) (0.80, 2.13) (0.50, 1.19) (0.32, 0.72)a (0.23, 0.70)a (0.13, 0.38)a 2.68 2.12 1.91 1.82 1.70 1.30 SEC 0.62 0.52 0.29 a a a a a a a (2.08, 3.47) (1.65, 2.73) (1.44, 2.53) (1.38, 2.40) (1.36, 2.12) (0.84, 2.00) (0.54, 0.72) (0.36, 0.75) (0.21, 0.39)a 4.30 3.40 3.06 2.91 2.72 2.08 1.60 UST 0.84 0.46 a a a a a a a (3.39, 5.49) (2.65, 4.39) (2.26, 4.14) (2.31, 3.68) (2.08, 3.56) (1.38, 3.13) (1.38, 1.86) (0.56, 1.24) (0.32, 0.65)a 5.15 4.07 3.66 3.49 3.26 2.49 1.92 1.19 ADA 0.55 a a a a a a a (3.31, 8.14) (2.61, 6.40) (2.32, 5.87) (2.21, 5.57) (2.46, 4.35) (1.42, 4.43) (1.34, 2.79) (0.81, 1.80) (0.34, 0.90)a 9.30 7.35 6.61 6.29 5.88 4.51 3.46 2.16 1.81 ETA a a a a a a a a a (6.23, 14.03) (4.95, 11.02 (4.37, 10.10) (4.16, 9.63) (4.04, 8.64) (2.63, 7.72) (2.57, 4.74) (1.53, 3.08) (1.12, 2.90) An odds ratio [ 1 indicates that the treatment in that column has a higher probability of achieving PASI response compared with the treatment in that row. An odds ratio \ 1 indicates that the treatment in that column has a lower probability of achieving PASI response compared with the treatment in that row ADA adalimumab, BKZ bimekizumab, BRO brodalumab, CrI credible interval, ETA etanercept, GUS guselkumab, IXE ixekizumab, PASI Psoriasis Area and Severity Index, PASI 90, 100 90% or 100% decrease from baseline PASI, Q4W every 4 weeks, Q8W every 8 weeks, RIS risankizumab, SEC secukinumab, UST ustekinumab a Denotes that the 95% CrI excludes 1 Dermatol Ther (Heidelb) (2022) 12:167–184 Dermatol Ther (Heidelb) (2022) 12:167–184 179 Fig. 3 Long-term benefit–risk profiles for licensed and event, BKZ bimekizumab, GUS guselkumab, IXE ixek- investigational treatments for moderate-to-severe plaque izumab, PASI Psoriasis Area and Severity Index, PASI 75, psoriasis. a SUCRA for PASI 75/90/100 responses vs. 90, 100 at least a 75, 90, or 100% decrease from baseline SUCRA for any AE. b SUCRA for PASI 75/90/100 PASI score, RIS risankizumab, SEC secukinumab, SUCRA responses vs. SUCRA for any SAE. c SUCRA for PASI surface under the cumulative ranking curves, UST 75/90/100 responses vs. SUCRA for AEs leading to ustekinumab treatment discontinuation. ADA Adalimumab, AE adverse trial of bimekizumab versus secukinumab Candida infection (until 20 weeks after the last reported that 21.2% and 4.6% of patients in dose of study drug administration) and that the those respective arms experienced Candida infections were evenly distributed between the infections during the period from weeks 0 to 48 two arms [37]. Sbidian et al. [12] also conducted [10]. Conversely, the 52-week IMMERGE trial of a risk–benefit analysis of treatments for moder- risankizumab versus secukinumab reported that ate-to-severe psoriasis, including bimekizumab, seven of 327 patients in total experienced and suggested that risankizumab and Table 3 Pairwise odds ratio of experiencing any AE, any SAE, and AEs leading to treatment discontinuation in the long terms (week 48–56) 180 Any AE: posterior median (95% CrI) RIS 1.26 (0.74, 2.16) 1.30 (0.69, 2.47) 1.58 (1.02, 2.48)a 1.61 (1.10, 2.39)a 2.05 (0.97, 4.41) 2.25 (1.25, 4.08)a 0.80 (0.46, 1.35) GUS 1.03 (0.73, 1.46) 1.26 (0.93, 1.71) 1.28 (0.88, 1.86) 1.63 (0.77, 3.49) 1.79 (1.09, 2.94)a 0.77 (0.41, 1.46) 0.97 (0.68, 1.37) ADA 1.22 (0.77, 1.94) 1.24 (0.75, 2.08) 1.58 (0.69, 3.64) 1.73 (0.94, 3.18) 0.63 (0.40, 0.98)a 0.79 (0.59, 1.07) 0.82 (0.52, 1.30) SEC 1.02 (0.82, 1.27) 1.29 (0.65, 2.59) 1.42 (0.96, 2.11) 0.62 (0.42, 0.91)a 0.78 (0.54, 1.13) 0.81 (0.48, 1.34) 0.98 (0.79, 1.22) UST 1.27 (0.67, 2.46) 1.39 (0.89, 2.19) 0.49 (0.23, 1.03) 0.61 (0.29, 1.29) 0.63 (0.27, 1.44) 0.77 (0.39, 1.53) 0.79 (0.41, 1.50) IXE 1.10 (0.50, 2.41) 0.45 (0.24, 0.80)a 0.56 (0.34, 0.92)a 0.58 (0.31, 1.06) 0.70 (0.47, 1.04) 0.72 (0.46, 1.12) 0.91 (0.42, 2.02) BKZ Any SAE: posterior median (95% CrI) RIS 1.25 (0.40, 3.84) 1.31 (0.72, 2.31) 1.35 (0.57, 3.20) 1.61 (0.79, 3.24) 1.67 (0.65, 4.29) 2.54 (0.74, 9.05) 0.80 (0.26, 2.48) ADA 1.04 (0.40, 2.75) 1.08 (0.52, 2.26) 1.29 (0.54, 3.11) 1.34 (0.46, 3.95) 2.04 (0.48, 8.97) 0.76 (0.43, 1.40) 0.96 (0.36, 2.51) UST 1.04 (0.55, 1.96) 1.23 (0.83, 1.85) 1.28 (0.61, 2.71) 1.94 (0.67, 6.04) 0.74 (0.31, 1.75) 0.92 (0.44, 1.92) 0.96 (0.51, 1.81) GUS 1.19 (0.73, 1.94) 1.24 (0.56, 2.73) 1.88 (0.54, 6.88) 0.62 (0.31, 1.27) 0.78 (0.32, 1.87) 0.81 (0.54, 1.21) 0.84 (0.52, 1.37) SEC 1.04 (0.56, 1.95) 1.58 (0.50, 5.26) 0.60 (0.23, 1.53) 0.75 (0.25, 2.19) 0.78 (0.37, 1.64) 0.81 (0.37, 1.78) 0.96 (0.51, 1.79) BKZ 1.52 (0.41, 5.84) 0.39 (0.11, 1.34) 0.49 (0.11, 2.08) 0.51 (0.17, 1.50) 0.53 (0.15, 1.86) 0.63 (0.19, 2.00) 0.66 (0.17, 2.43) IXE AEs leading to treatment discontinuation: posterior median (95% CrI) RIS 2.39 (0.56, 9.49) 2.69 (0.46, 15.10) 3.44 (0.74, 15.15) 3.63 (0.51, 25.40) 4.50 (0.78, 25.22) 4.77 (0.44, 59.29) 0.42 (0.11, 1.77) UST 1.13 (0.40, 3.16) 1.43 (0.83, 2.53) 1.52 (0.39, 6.03) 1.88 (0.68, 5.29) 1.98 (0.29, 17.08) 0.37 (0.07, 2.17) 0.89 (0.32, 2.49) GUS 1.27 (0.54, 3.07) 1.34 (0.56, 3.38) 1.67 (0.50, 5.71) 1.77 (0.20, 19.04) 0.29 (0.07, 1.35) 0.70 (0.40, 1.21) 0.79 (0.33, 1.86) SEC 1.06 (0.31, 3.70) 1.31 (0.56, 3.12) 1.38 (0.19, 12.81) 0.28 (0.04, 1.97) 0.66 (0.17, 2.57) 0.74 (0.30, 1.79) 0.94 (0.27, 3.27) ADA 1.24 (0.27, 5.60) 1.31 (0.12, 16.45) 0.22 (0.04, 1.28) 0.53 (0.19, 1.46) 0.60 (0.18, 2.01) 0.77 (0.32, 1.78) 0.81 (0.18, 3.67) BKZ 1.06 (0.12, 11.25) 0.21 (0.02, 2.27) 0.50 (0.06, 3.39) 0.56 (0.05, 4.99) 0.72 (0.08, 5.31) 0.76 (0.06, 8.04) 0.94 (0.09, 8.38) IXE An odds ratio [ 1 indicates that the treatment in that column has a higher probability of experiencing safety events compared with the treatment in that row. An odds ratio \ 1 indicates that the treatment in that column has a lower probability of experiencing safety events compared with the treatment in that row a Denotes that the 95% CrI excludes 1 ADA adalimumab, AE adverse event, BKZ bimekizumab, CrI credible interval, GUS guselkumab, IXE ixekizumab, RIS risankizumab, SAE serious adverse event, SEC secukinumab, UST ustekinumab Dermatol Ther (Heidelb) (2022) 12:167–184 Dermatol Ther (Heidelb) (2022) 12:167–184 181 bimekizumab had favorable short-term bene- lower rankings for safety outcomes compared fit–risk profiles based on PASI 90 response rates with risankizumab. and SAEs. However, the results for bimekizumab were considered to be of low certainty as the phase 3 trial data were not available for inclu- ACKNOWLEDGEMENTS sion at the time of the study’s conduct. This study is subject to several limitations, some of which are inherent to indirect com- Funding. Sponsorship for this study and the parisons. First, as with other NMAs, potential Rapid Service Fee were funded by AbbVie Inc. cross-trial differences in trial characteristics, patient characteristics, and outcome assessment Medical Writing. Editorial assistance in the (e.g., time points, AE assessment and collection) preparation of this article was provided by Dr. may undermine the comparability of outcomes Shelley Batts of Analysis Group, Inc. Support for across trials. Second, given the sparsity of the this assistance was funded by AbbVie Inc. networks, it was infeasible to implement a ran- dom-effects or meta-regression model to cap- Authorship. All authors meet the Interna- ture or explain the potential heterogeneities in tional Committee of Medical Journal Editors contrasts between treatments connected by (ICMJE) criteria for authorship for this article, more than one trial (e.g., risankizumab vs. take responsibility for the integrity of the work ustekinumab). Third, as the study synthesized as a whole, and have given their approval for clinical trial data, the results may not necessar- this version to be published. ily be generalizable to all patients with moder- ate-to-severe plaque psoriasis in the real world. Author Contributions. All authors con- As more long-term data are becoming avail- tributed to the study conception and design. able for treatments for moderate-to-severe pla- Data collection and analysis were performed by que psoriasis, future studies may leverage a Yan Wang and Yawen Gao. The first draft of the richer set of trial data to understand the impact manuscript was written by Yan Wang and all of cross-trial heterogeneities in patient and trial authors commented on previous versions of the characteristics on the comparative long-term manuscript. All authors read and approved the benefit–risk profiles. With a richer network, final manuscript. techniques such as random-effects NMA models and traditional and multilevel network meta- Prior Presentation. Portions of this research regression [38] could be considered. Addition- are planned to be presented at the 9th Interna- ally, validating the long-term benefit–risk pro- tional Congress, Psoriasis: from Gene to Clinic, files of the therapies considered in this study in occurring during 9–11 December, 2021, and the the real world will also augment the present 2022 American Academy of Dermatology (AAD) results. Annual Meeting, occurring during 25–29 March, 2022 in Boston, MA, USA. CONCLUSIONS Disclosures. April W. Armstrong serves as investigator and/or scientific advisor to AbbVie, Risankizumab and bimekizumab had the high- BMS, Incyte, Leo, UCB, Janssen, Lilly, Novartis, est long-term PASI response rates, and risanki- Ortho Dermatologics, Sun, Dermavant, Der- zumab had the lowest long-term rates of safety mira, Sanofi, Regeneron, Pfizer, and Modmed. events. Therefore, risankizumab was associated Ahmed M. Soliman and Vassilis Stakias are with the most favorable long-term benefit–risk employees of AbbVie and own AbbVie stock. profile. While bimekizumab and ixekizumab Luis Puig has served as investigator and/or had favorable efficacy profiles, both treatments consultant or paid speaker to AbbVie, Almirall, had higher rates of safety events/outcomes and Amgen, Biogen, Boehringer Ingelheim, Cel- gene, Janssen, Leo, Lilly, Merck, Novartis, Pfizer, 182 Dermatol Ther (Heidelb) (2022) 12:167–184 and UCB. Keith A. Betts, Yan Wang, and Yawen 2. Armstrong AW, Mehta MD, Schupp CW, Gondo Gao are employed by Analysis Group, Inc., GC, Bell SJ, Griffiths CEM. Psoriasis prevalence in adults in the United States. JAMA Dermatol. which received payment from AbbVie Inc. for 2021;157(8):940–6. participation in this research. 3. Parisi R, Iskandar IYK, Kontopantelis E, et al. Compliance with Ethics Guidelines. As this National, regional, and worldwide epidemiology of psoriasis: systematic analysis and modelling study. is a post-hoc NMA of previously published BMJ. 2020;369:m1590. results of clinical trial data, no institutional board review was required. The study was per- 4. Sampogna F, Gisondi P, Melchi CF, et al. Prevalence formed in accordance with the Helsinki Decla- of symptoms experienced by patients with different clinical types of psoriasis. Br J Dermatol. ration 1964 and its later amendments. This 2004;151(3):594–9. article is based on previously conducted studies and does not contain any new studies with 5. Menter A, Strober BE, Kaplan DH, et al. Joint AAD- human participants or animals performed by NPF guidelines of care for the management and treatment of psoriasis with biologics. J Am Acad any of the authors. Dermatol. 2019;80(4):1029–72. Data Availability. The datasets analyzed 6. Honma M, Hayashi K. Psoriasis: recent progress in during the current study are in the electronic molecular-targeted therapies. J Dermatol. supplementary materials of the published 2021;48(6):761–77. article. 7. Reich K, Papp KA, Blauvelt A, et al. Bimekizumab versus ustekinumab for the treatment of moderate Open Access. This article is licensed under a to severe plaque psoriasis (BE VIVID): Efficacy and Creative Commons Attribution-NonCommer- safety from a 52-week, multicentre, double-blind, cial 4.0 International License, which permits active comparator and placebo controlled phase 3 trial. Lancet. 2021;397(10273):487–98. any non-commercial use, sharing, adaptation, distribution and reproduction in any medium 8. Gordon KB, Foley P, Krueger JG, et al. Bimekizumab or format, as long as you give appropriate credit efficacy and safety in moderate to severe plaque to the original author(s) and the source, provide psoriasis (BE READY): a multicentre, double-blind, placebo-controlled, randomised withdrawal phase 3 a link to the Creative Commons licence, and trial. Lancet. 2021;397(10273):475–86. indicate if changes were made. The images or other third party material in this article are 9. Warren RB, Blauvelt A, Bagel J, et al. Bimekizumab included in the article’s Creative Commons versus adalimumab in plaque psoriasis. N Engl J Med. 2021;385(2):130–41. licence, unless indicated otherwise in a credit line to the material. If material is not included 10. Reich K, Warren RB, Lebwohl M, et al. Bimek- in the article’s Creative Commons licence and izumab versus secukinumab in plaque psoriasis. your intended use is not permitted by statutory N Engl J Med. 2021;385(2):142–52. regulation or exceeds the permitted use, you 11. European Medicines Agency. Product information: will need to obtain permission directly from the Bimzlex (bimekizumab). https://www.ema.europa. copyright holder. To view a copy of this licence, eu/en/medicines/human/EPAR/bimzelx#product- visit http://creativecommons.org/licenses/by- information-section. Accessed 2 Nov 2021. nc/4.0/. 12. Sbidian E, Chaimani A, Afach S, et al. Systemic pharmacological treatments for chronic plaque psoriasis: a network meta-analysis. Cochrane Data- base Syst Rev. 2020;1(1):Cd011535. REFERENCES 13. Mrowietz U. Implementing treatment goals for successful long-term management of psoriasis. J Eur 1. AlQassimi S, AlBrashdi S, Galadari H, Hashim MJ. Acad Dermatol Venereol. 2012;26(suppl_2):12–20. Global burden of psoriasis—comparison of regional and global epidemiology, 1990 to 2017. Int J Der- 14. Boehncke WH, Boehncke S, Tobin AM, Kirby B. The matol. 2020;59(5):566–71. ‘‘psoriatic march’’: a concept of how severe psoriasis Dermatol Ther (Heidelb) (2022) 12:167–184 183 may drive cardiovascular comorbidity. Exp Derma- 25. Mahil SK, Ezejimofor MC, Exton LS, et al. Com- tol. 2011;20(4):303–7. paring the efficacy and tolerability of biologic therapies in psoriasis: an updated network meta- 15. Takeshita J, Grewal S, Langan SM, et al. Psoriasis analysis. Br J Dermatol. 2020;183(4):638–49. and comorbid diseases: implications for manage- ment. J Am Acad Dermatol. 2017;76(3):393–403. 26. Signorovitch JE, Betts KA, Yan YS, et al. Compara- tive efficacy of biological treatments for moderate- 16. Sain N, Willems D, Charokopou M, Hiligsmann M. to-severe psoriasis: a network meta-analysis adjust- The importance of understanding patient and ing for cross-trial differences in reference arm physician preferences for psoriasis treatment char- response. Br J Dermatol. 2015;172(2):504–12. acteristics: a systematic review of discrete-choice experiments. Curr Med Res Opin. 2020;36(8): 27. Armstrong AW, Betts KA, Signorovitch JE, et al. 1257–75. Number needed to treat and costs per responder among biologic treatments for moderate-to-severe 17. Shear NH, Betts KA, Soliman AM, et al. Compara- psoriasis: a network meta-analysis. Curr Med Res tive safety and benefit–risk profile of biologics and Opin. 2018;34(7):1325–33. oral treatment for moderate-to-severe plaque pso- riasis: a network meta-analysis of clinical trial data. 28. Sawyer LM, Malottki K, Sabry-Grant C, et al. J Am Acad Dermatol. 2021;85(3):572–81. Assessing the relative efficacy of interleukin-17 and interleukin-23 targeted treatments for moderate-to- 18. Torres T, Puig L, Vender R, et al. Drug survival of IL- severe plaque psoriasis: a systematic review and 12/23, IL-17 and IL-23 inhibitors for psoriasis network meta-analysis of PASI response. PLoS ONE. treatment: a retrospective multi-country, multi- 2019;14(8): e0220868. centric cohort study. Am J Clin Dermatol. 2021;22(4):567–79. 29. Bai F, Li GG, Liu Q, Niu X, Li R, Ma H. Short-term efficacy and safety of IL-17, IL-12/23, and IL-23 19. Mourad AI, Gniadecki R. Biologic drug survival in inhibitors brodalumab, secukinumab, ixekizumab, psoriasis: a systematic review & comparative meta- ustekinumab, guselkumab, tildrakizumab, and analysis. Front Med. 2021;7:625755. risankizumab for the treatment of moderate to severe plaque psoriasis: a systematic review and 20. Armstrong AW, Puig L, Joshi A, et al. Comparison of network meta-analysis of randomized controlled biologics and oral treatments for plaque psoriasis: a trials. J Immunol Res. 2019;10:2546161. meta-analysis. JAMA Dermatol. 2020;156(3): 258–69. 30. Xu G, Xia M, Jiang C, et al. Comparative efficacy and safety of thirteen biologic therapies for patients 21. Armstrong AW, Soliman AM, Betts KA, et al. Com- with moderate or severe psoriasis: a network meta- parative efficacy and relative ranking of biologics analysis. J Pharmacol Sci. 2019;139(4):289–303. and oral therapies for moderate-to-severe plaque psoriasis: a network meta-analysis. Dermatol Ther. 31. Xue W, Saharia P, Gray E, et al. Efficacy of bro- 2021;11(3):885–905. dalumab for moderate to severe plaque psoriasis: a Canadian network meta-analysis. J Cutan Med 22. Reich K, Armstrong AW, Foley P, et al. Efficacy and Surg. 2020;24(6):561–72. safety of guselkumab, an anti-interleukin-23 mon- oclonal antibody, compared with adalimumab for 32. Tada Y, Watanabe R, Noma H, Kanai Y, Nomura T, the treatment of patients with moderate to severe Kaneko K. Short-term effectiveness of biologics in psoriasis with randomized withdrawal and retreat- patients with moderate-to-severe plaque psoriasis: a ment: Results from the phase III, double-blind, systematic review and network meta-analysis. placebo- and active comparator-controlled VOY- J Dermatol Sci. 2020;99(1):53–61. AGE 2 trial. J Am Acad Dermatol. 2017;76(3): 418–31. 33. Warren RB, Gooderham M, Burge R, et al. Com- parison of cumulative clinical benefits of biologics 23. Dias S, Sutton AJ, Ades AE, Welton NJ. Evidence for the treatment of psoriasis over 16 weeks: results synthesis for decision making 2: a generalized linear from a network meta-analysis. J Am Acad Dermatol. modeling framework for pairwise and network 2020;82(5):1138–49. meta-analysis of randomized controlled trials. Med Decis Making. 2013;33(5):607–17. 34. Sawyer L, Fotheringham I, Wright E, Yasmeen N, Gibbons C, Holmen MA. The comparative efficacy 24. Rücker G, Schwarzer G. Ranking treatments in fre- of brodalumab in patients with moderate-to-severe quentist network meta-analysis works without psoriasis: a systematic literature review and network resampling methods. BMC Med Res Methodol. meta-analysis. J Dermatol Treat. 2018;29(6):557–68. 2015;15(1):58. 184 Dermatol Ther (Heidelb) (2022) 12:167–184 35. Yasmeen N, Sawyer LM, Malottki K, Levin L, double-blinded, placebo- and active comparator- Didriksen Apol E, Jemec GB. Targeted therapies for controlled VOYAGE 1 trial. J Am Acad Dermatol. patients with moderate-to-severe psoriasis: A sys- 2017;76(3):405–17. tematic review and network meta-analysis of PASI response at 1 year. J Dermatolog Treat. 2020. 42. Blauvelt A, Reich K, Tsai TF, et al. Secukinumab is https://doi.org/10.1080/09546634.2020.1743811. superior to ustekinumab in clearing skin of subjects with moderate-to-severe plaque psoriasis up to 1 36. Blauvelt A, Gooderham M, Griffith CE, et al. year: results from the CLEAR study. J Am Acad Cumulative clinical benefits of biologic treatments Dermatol. 2017;76(1):60-9.e9. for psoriasis over 1 year. Presented at American Academy of Dermatology Virtual Meeting Experi- 43. Langley RG, Elewski BE, Lebwohl M, et al. Secuk- ence (AAD VMX) on 23–25 April, 2021 (virtual). inumab in plaque psoriasis–results of two phase 3 trials. N Engl J Med. 2014;371(4):326–38. 37. Warren RB, Blauvelt A, Poulin Y, et al. Efficacy and safety of risankizumab vs. secukinumab in patients 44. Bagel J, Blauvelt A, Nia J, et al. Secukinumab with moderate-to-severe plaque psoriasis maintains superiority over ustekinumab in clearing (IMMerge): results from a phase 3, randomised, skin and improving quality of life in patients with open-label, efficacy assessor-blinded clinical trial. Br moderate to severe plaque psoriasis: 52-week results J Dermatol. 2021;184(1):50–9. from a double-blind phase 3b trial (CLARITY). J Eur Acad Dermatol Venereol. 2021;35(1):135–42. 38. Phillippo DM, Dias S, Ades AE, et al. Multilevel network meta-regression for population-adjusted 45. Paul C, Griffiths CEM, van de Kerkhof PCM, et al. treatment comparisons. J R Stat Soc Ser A Stat Soc. Ixekizumab provides superior efficacy compared 2020;183(3):1189–210. with ustekinumab over 52 weeks of treatment: Results from IXORA-S, a phase 3 study. J Am Acad 39. Lebwohl M, Strober B, Menter A, et al. Phase 3 Dermatol. 2019;80(1):70-9.e3. studies comparing brodalumab with ustekinumab in psoriasis. N Engl J Med. 2015;373(14):1318–28. 46. Gordon KB, Strober B, Lebwohl M, et al. Efficacy and safety of risankizumab in moderate-to-severe 40. Reich K, Armstrong AW, Langley RG, et al. plaque psoriasis (UltIMMa-1 and UltIMMa-2): Guselkumab versus secukinumab for the treatment results from two double-blind, randomised, pla- of moderate-to-severe psoriasis (ECLIPSE): Results cebo-controlled and ustekinumab-controlled phase from a phase 3, randomised controlled trial. Lancet. 3 trials. Lancet. 2018;392(10148):650–61. 2019;394(10201):831–9. 47. Thaçi D, Blauvelt A, Reich K, et al. Secukinumab is 41. Blauvelt A, Papp KA, Griffiths CE, et al. Efficacy and superior to ustekinumab in clearing skin of subjects safety of guselkumab, an anti-interleukin-23 mon- with moderate to severe plaque psoriasis: CLEAR, a oclonal antibody, compared with adalimumab for randomized controlled trial. J Am Acad Dermatol. the continuous treatment of patients with moder- 2015;73(3):400–9. ate to severe psoriasis: results from the phase III,