DOI: 10.1111/jdv.16521 JEADV ORIGINAL ARTICLE Efficacy of risankizumab in patients with moderate-to- severe plaque psoriasis by baseline demographics, disease characteristics and prior biologic therapy: an integrated analysis of the phase III UltIMMa-1 and UltIMMa-2 studies B. Strober,1,2,* A. Menter,3 C. Leonardi,4 K. Gordon,5 J. Lambert,6 L. Puig,7 H. Photowala,8 M. Longcore, T. Zhan,8 P. Foley9 8 1 Yale University, New Haven, CT, USA 2 Central Connecticut Dermatology Research, Cromwell, CT, USA 3 Baylor Scott and White, Dallas, TX, USA 4 Central Dermatology, Richmond Heights, MO, USA 5 Medical College of Wisconsin, Milwaukee, WI, USA 6 Ghent University, Ghent, Belgium 7 Hospital de la Santa Creu i Sant Pau, Barcelona, Spain 8 AbbVie, Inc., North Chicago, IL, USA 9 St. Vincent’s Hospital Melbourne, Probity Medical Research, Skin Health Institute, The University of Melbourne, Melbourne, VIC, Australia *Correspondence: B. Strober. E-mail: brucestrober30@me.com Abstract Background Risankizumab is a humanized IgG monoclonal antibody that selectively inhibits interleukin-23 through binding the p19 subunit. In Phase 3 trials, risankizumab demonstrated superior efficacy compared with adalimumab and ustekinumab in patients with moderate-to-severe plaque psoriasis. Here, we evaluated the impact of baseline character- istics on efficacy of risankizumab compared with ustekinumab in patients with moderate-to-severe plaque psoriasis. Methods This analysis included all patients initially randomized to risankizumab or ustekinumab from the repli- cate, double-blinded, randomized, placebo-controlled phase 3 trials, UltIMMa-1 (NCT02684370) and UltIMMa-2 (NCT02684357). Patients received either risankizumab (150 mg) or ustekinumab (weight-based; 45 or 90 mg per label) at weeks 0, 4, 16, 28 and 40. Efficacy was assessed as the proportion of patients achieving ≥90% improvement in Psori- asis Area and Severity Index (PASI 90) at weeks 16 and 52 by baseline patient demographics, disease characteristics and prior biologic exposure. Mean per cent improvement in PASI was calculated by body weight and body mass index at week 52. Missing efficacy data were imputed as non-responders for categorical variables and last observation carried forward for continuous variables. Logistic regression analyses assessed for interactions between treatment and five independent variables (age, sex, weight, baseline PASI score and presence of psoriatic arthritis) at both weeks 16 and 52. Results Baseline patient demographics, disease characteristics and prior biologic exposure were similar between patients randomized to risankizumab (n = 598) and ustekinumab (n = 199). At weeks 16 and 52, risankizumab demon- strated superior efficacy compared with ustekinumab across these patient characteristics (P < 0.01). Logistic regression analyses demonstrated that risankizumab was superior to ustekinumab at weeks 16 and 52 in all models tested (P < 0.0001 for all). Conclusions Risankizumab demonstrated consistent and superior efficacy compared with ustekinumab regardless of patient demographics, disease characteristics or prior biologic exposure. Received: 21 January 2020; Accepted: 31 March 2020 Conflict of interest Bruce Strober has served as an investigator (no direct payments made to Bruce Strober, MD, PhD) for Dermavant, AbbVie, the CORRONA Psoriasis Registry, and Dermira; has served as Scientific Director (consulting fee) for the CORRONA Psoriasis Registry; has served as a consultant (honoraria) for AbbVie, Almirall, Amgen, Arena, Aristea, Boehringer Ingelheim, Bristol-MyersSquibb, Celgene, Dermavant, Dermira, Janssen, Leo, Eli Lilly, Kyowa Hakko Kirin, Meiji Seika Pharma, Novartis, Pfizer, GlaxoSmithKline, UCB Pharma, Sun Pharma, Ortho Dermatologics, Regeneron, JEADV 2020, 34, 2830–2838 © 2020 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. RZB efficacy in patient subgroups 2831 and Sanofi-Genzyme; and has served as a speaker for AbbVie, Lilly, Janssen, and Ortho Dermatologics. Alan Menter has received grant support from AbbVie, Amgen, Boehringer Ingelheim, Celgene, Janssen Biotech, Inc., LEO Pharma, Merck, and Sienna; honoraria from AbbVie, Amgen, Boehringer Ingelheim, Eli Lilly, Janssen Biotech, Inc., LEO Pharma, Novartis, Sienna, and UCB; has served as an investigator for AbbVie, Amgen, Boehringer Ingelheim, Celgene, Eli Lilly, Janssen Biotech, Inc., LEO Pharma, Merck, Novartis, Sienna, and UCB; has served on the advisory board of AbbVie, Amgen, Boehringer Ingelheim, Janssen Biotech, Inc., LEO Pharma, and Sienna; has served as a consultant for AbbVie, Amgen, Eli Lilly, Janssen Biotech, Inc., LEO Pharma, Novartis, Sienna, and UCB; and has served as a speaker for AbbVie, Amgen, Janssen Biotech, Inc., LEO Pharma, Sienna, and UCB. Craig L. Leonardi, MD, FAAD, served as a consultant/ advisory board member for AbbVie, Amgen, Boehringer Ingelheim, Celgene Corporation, Dermira, Eli Lilly and Company, Janssen Pharmaceuticals, Inc., Leo Pharma A/S, Ortho Dermatologics, Pfizer, Inc., Sandoz (a Novartis Company), UCB and Vitae receiving honoraria; as a speaker for AbbVie, Amgen, Celgene Corporation, Eli Lilly and Company, Novartis, Sun Pharmaceuticals, Ltd and UCB receiving honoraria; as a principal investigator for Actavis, Amgen, Boehringer Ingelheim, Celgene Corporation, Cellceutix, Coherus Biosciences, Corrona, Dermira, Eli Lilly and Company, Galderma Laboratories, L.P., Glenmark Generics, Inc., Janssen Pharmaceuticals, Inc., Leo Pharma, Inc., Merck, Novartis, Novella, Pfizer, Inc., Sandoz (a Novartis Company), Sienna Biopharmaceuticals, Stiefel a GSK company, UCB, and Warner Chillcott receiving other financial benefits (fee for service). Kenneth Gordon has received research grants from AbbVie, Bristol-Myers Squibb, Celgene, Janssen, and Novartis, and has served as a consultant for AbbVie, Almirall, Amgen, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Dermira, Eli Lilly, Janssen, Novartis, Pfizer, Sun, and UCB Pharma. Jo Lambert has received research grants from and has been an advisor/speaker for AbbVie, Celgene, Janssen, LEO Pharma, Eli Lilly, and Novartis. Lluıs Puig has received grants/research support (paid to the institution) from or participated in clinical trials for AbbVie, Amgen, Boehringer Ingelheim, Janssen, Leo Pharma, Lilly, Novartis, Pfizer, Regeneron, Roche, Sanofi, and UCB; received honoraria or consultation fees from AbbVie, Almirall, Amgen, Baxalta, Biogen, Boehringer Ingelheim, Celgene, Gebro, Janssen, Leo Pharma, Lilly, Merck-Serono, MSD, Mylan, Novartis, Pfizer, Regeneron, Roche, Sandoz, Sanofi, and UCB; and has participated in company-sponsored speakers' bureaus for Celgene, Janssen, Lilly, MSD, Novartis, and Pfizer. Huzefa Photowala, Michelle Longcore, and Tianyu Zhan are employees of AbbVie, Inc. and may hold stock or stock options. Peter Foley has received grant support from AbbVie, Amgen, Celgene, Janssen, Lilly, Merck, Novartis, Pfizer, and Sun Pharma. He has served as an investigator for AbbVie, Amgen, Boehringer Ingelheim, Celgene, Janssen, Lilly, Merck, Novartis, Pfizer, Sun Pharma, UCB Pharma, Valeant, BMS, Celtaxsys, CSL, Cutanea, Dermira, Galderma, Genentech, GSK, Leo Pharma, Regeneron Pharmaceuticals Inc, Roche, and Sanofi. He has also served on advisory boards for AbbVie, Amgen, Celgene, Janssen, Lilly, Merck, Novartis, Pfizer, Sun Pharma, UCB Pharma, Valeant, Galderma, GSK, Leo Pharma, and Sanofi. He has served as a consultant for Janssen, Lilly, Novartis, Pfizer, UCB Pharma, BMS, Galderma, Leo Pharma, and Roche. He has received travel grants from AbbVie, Janssen, Lilly, Merck, Novartis, Pfizer, Galderma, Leo Pharma, Roche, Sun Pharma, and Sanofi, and has served as a speaker for or received honoraria from AbbVie, Celgene, Janssen, Lilly, Merck, Novartis, Pfizer, Valeant, Galderma, GSK, Leo Pharma, and Roche. Funding source AbbVie sponsored the study, contributed to its design, data collection, analysis, and interpretation of the data, and par- ticipated in the writing, review, and approval of the manuscript. All authors had access to relevant data. Introduction treatments to demonstrate consistently high efficacy regardless Psoriasis is a chronic immune-mediated disease that affects an of patient demographics or prior biologic therapy. estimated 100 million people worldwide.1,2 Patients with moder- Recent preclinical and clinical findings demonstrated that ate-to-severe psoriasis are at increased risk of comorbidities, interleukin-23 (IL-23) is a regulatory cytokine critical for the including obesity.3,4 Although recent scientific advances have led development and maintenance of psoriatic inflammation.17 Pre- to highly efficacious treatments for moderate-to-severe psoriasis, clinical research established that IL-23 activates T helper and lower efficacy has been reported with biologic treatments in other cells to produce key cytokines and drive psoriatic inflam- patients with comorbid obesity or in those with prior biologic mation. In clinical trials of patients with moderate-to-severe pla- exposure and/or failure.5–16 Thus, there is a need for psoriasis que psoriasis, efficacy has been demonstrated by an IL-12/IL-23 JEADV 2020, 34, 2830–2838 © 2020 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology 2832 Strober et al. inhibitor, ustekinumab and, more recently, by the selective IL- study protocol) were approved by an institutional review board 23p19 inhibitors, guselkumab, tildrakizumab and risankizumab, or institutional ethics committee at each trial site, and all further supporting the role of IL-23 in psoriasis.18–23 patients provided written informed consent before participation. Risankizumab is a humanized IgG monoclonal antibody that binds with high affinity and specificity to the p19 subunit and Study design selectively inhibits IL-23.24,25 Exposure-response analyses of Detailed methodology has been reported previously for the repli- Phase 2 and 3 trials demonstrated that, following initial doses at cate, multi-part phase 3, multinational, multicenter, randomized, weeks 0 and 4, the every 12-week dosing of risankizumab double-blinded, placebo- and active comparator-controlled tri- (150 mg) led to maximal efficacy at weeks 16 and 52.26 In phase als, UltIMMa-1 and UltIMMa-2.22 Here, we report on patients 3 trials (IMMhance, UltIMMa-1, UltIMMa-2, and IMMvent), with moderate-to-severe plaque psoriasis who were initially ran- risankizumab demonstrated superior efficacy compared with domized to either risankizumab (150 mg) or ustekinumab placebo, ustekinumab and adalimumab at week 16 that was sus- (weight-based; 45 or 90 mg) in Part A (Fig. 1). Randomization tained compared with adalimumab and ustekinumab at weeks was stratified by baseline weight (≤100 kg vs. >100 kg) and prior 44 and 52, respectively.22,23,27 Overall, risankizumab has demon- exposure to TNF inhibitor (yes vs. no). Patients initially random- strated early and sustained high skin clearance with 12-week ized to receive risankizumab or ustekinumab received 150 mg dosing in patients with moderate-to-severe psoriasis.22,23 To risankizumab or weight-based ustekinumab (45 mg for patients date, however, the association of baseline disease or prior treat- with body weight ≤100 kg or 90 mg for patients with body ment variables on risankizumab’s efficacy has not been reported. weight >100 kg) subcutaneously at weeks 0, 4, 16, 28 and 40. Here, using integrated data from UltIMMa-1 and UltIMMa-2, we assessed the impact of baseline patient demographics, disease Assessments characteristics and prior biologic exposure on the efficacy of Efficacy was analysed using pooled data from UltIMMa-1 risankizumab compared with ustekinumab in patients with and UltIMMa-2 in all patients initially randomized to either moderate-to-severe plaque psoriasis. risankizumab or ustekinumab using an intention-to-treat analy- sis. Efficacy was assessed by demographics and baseline charac- Methods teristics [age, sex, body mass index (BMI), weight, baseline PASI score, baseline sPGA score and presence of psoriatic arthritis] as Patients well as prior treatment experience (prior biologic exposure, TNF This integrated analysis included all patients initially random- inhibitor exposure, IL-17 inhibitor exposure and prior biologic ized to risankizumab or ustekinumab in UltIMMa-1 or failure, including failure of 1 or ≥2 biologics and primary or sec- UltIMMa-2. In both trials, adult patients (≥18 years of age) were ondary failure). Efficacy outcomes included the proportion of eligible if they had stable moderate-to-severe chronic plaque patients achieving ≥90% improvement in PASI compared to psoriasis for the preceding ≥6 months (with or without psoriatic baseline (PASI 90) at weeks 16 and 52. Additional efficacy out- arthritis) with body surface area (BSA) involvement of 10% or comes were the proportion of patients achieving sPGA score of greater, Psoriasis Area and Severity Index (PASI) 12 or greater, clear or almost clear (sPGA 0 or 1), or sPGA score of clear and static Physician’s Global Assessment (sPGA) score 3 or (sPGA 0), and PASI 100. greater. Patients were required to be candidates for systemic Further efficacy analyses were conducted in subgroups at week therapy, phototherapy and treatment with ustekinumab (accord- 52 using the categories of BMI (<25, 25–<30, ≥30), weight ing to local label). Prior treatment exposure to IL-17 and (≤100 kg and >100 kg), weight quartiles, and weight deciles in tumour necrosis factor (TNF) inhibitors was allowed depending order to fully explore the effect of weight on efficacy. Mean per on timing prior to randomization; prior treatment exposure to cent improvement in PASI score from baseline to week 52 was ustekinumab or other IL-23 inhibitors was not permitted. Prior analysed for risankizumab and ustekinumab using last observa- biologic exposure was self-reported; prior biologic failure was tion carried forward. defined as any patient with prior biologic exposure who failed to respond to biologic treatment (primary failure) or failed due to Statistical analysis loss of response or lack of tolerability (secondary failures). A Patients with missing efficacy data for categorical variables were complete list of inclusion and exclusion criteria is included in handled by non-responder imputation and for continuous vari- Table S1, Supporting Information). ables with last observation carried forward. Categorical variables The trials were conducted in accord with the Good Clinical were tested using the Cochran-Mantel-Haenszel risk difference Practice Guideline, as defined by the International Conference estimate. Per cent change in PASI scores were compared between on Harmonisation of Technical Requirements for Pharmaceuti- risankizumab and ustekinumab using analysis of covariance with cals for Human Use, the Declaration of Helsinki and applicable study, stratum, baseline value and treatment in the model. Logis- local regulations. All study-related documents (including the tic regression analyses were conducted for each efficacy endpoint JEADV 2020, 34, 2830–2838 © 2020 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology RZB efficacy in patient subgroups 2833 Part A Part B Risankizumab 150 mg Weeks 0, 4, 16, 28, and 40 Study completers had Randomized 3:1:1 option to enter OLE Ustekinumab 45/90 mg Weeks 0, 4, 16, 28, and 40 Placebo Risankizumab 150 mg Week 0 and 4 Weeks 16, 28, and 40 Week 0 4 8 12 16 22 28 34 40 46 52 56 = Risankizumab Week 16 Endpoints: Week 52 Endpoints: = Ustekinumab • PASI 90 • PASI 90 • PASI 100 • PASI 100 = Placebo • sPGA 0/1 • sPGA 0/1 • SPGA 0 • SPGA 0 Figure 1 Study design for UltIMMa-1 and UltIMMa-2. OLE, open-label extension; PASI, Psoriasis Area and Severity Index; sPGA, static Physician’s Global Assessment. at weeks 16 and 52 to assess the impact on risankizumab efficacy Table 1 Baseline demographics and disease characteristics of compared with ustekinumab of five independent variables (age, the intention-to-treat population sex, baseline weight, baseline PASI and presence of psoriatic Baseline characteristics Risankizumab Ustekinumab† arthritis) and potential interactions between these five indepen- (N = 598) (N = 199) dent variables (Table S2, Supporting Information). The Bonfer- Age, years, mean (SD) 47.2 (13.6) 47.5 (14.1) roni correction was used to adjust the critical value of 0.05 for Male, n (%) 415 (69.4%) 136 (68.3%) statistical significance in order to account for the multiple com- Race, white, n (%) 455 (76.1%) 165 (82.9%) parisons. All statistical analyses were conducted using SASâ ver- Weight, kg, mean (SD) 90.0 (22.4) 90.4 (22.2) sion 9.4 (SAS Institute, Inc., Cary, NC, USA) or higher using the Weight >100 kg, n (%)‡ 169 (28.3%) 56 (28.1%) UNIX operating system. BMI, kg/m2, mean (SD) 30.5 (7.0) 30.4 (6.8) PASI, mean (SD) 20.6 (7.7) 19.2 (6.4) sPGA of severe, n (%) 114 (19.1%) 33 (16.6%) Results BSA involvement, %, mean (SD) 26.2 (15.6) 23.0 (13.6) Psoriatic arthritis status, n (%)§ 159 (26.6%) 50 (25.1%) Patient demographics and baseline disease Any prior biologic therapy, n (%) 222 (37.1%) 73 (36.7%) characteristics Prior TNFi exposure, n (%)‡ 134 (22.4%) 43 (21.6%) Overall, 797 patients were randomized to receive either 150 mg Prior IL-17i exposure, n (%) 111 (18.6%) 35 (17.6%) risankizumab (n = 598) or 45/90 mg ustekinumab (n = 199) at BMI, body mass index; BSA, body surface area; IL-17i, interleukin-17 the start of UltIMMa-1 or UltIMMa-2, and were included in this inhibitor; PASI, Psoriasis Area Severity Index; SD, standard deviation; integrated analysis. Demographics and baseline disease charac- sPGA, static Physician’s Global Assessment; TNFi, tumour necrosis teristics were similar between groups (Table 1). Overall, most factor inhibitor. were male (69.1%); mean age was 47.3 years [standard deviation †Weight-based dose per label. ‡Stratification factors at randomization. (SD): 13.7 years] and mean weight was 90.1 kg (SD: 22.3 kg). §Diagnosed or suspected. Mean PASI score was 20.2 (SD: 7.5) and mean BSA involvement [Correction added on 26 October 2020, after first online publication: Footnote was 25.4% (SD: 15.2%). Prior systemic non-biologic and symbols on Table 1 have been corrected in this version.] JEADV 2020, 34, 2830–2838 © 2020 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology 2834 Strober et al. Figure 2 Proportion of patients (non-responder imputation) achieving PASI 90 responses at week 16 (a) and week 52 (b) by baseline patient demographics, disease characteristics and prior biologic therapy (NRI). BL, baseline; NRI, non-responder imputation; PASI, Psoriasis Area and Severity Index; PsA, psoriatic arthritis; RZB, risankizumab; sPGA, static Physician’s Global Assessment; UST, ustekinumab. **P < 0.01, ***P < 0.001 compared with ustekinumab. JEADV 2020, 34, 2830–2838 © 2020 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology RZB efficacy in patient subgroups 2835 RZB RZB UST UST (a) Week 16: PASI 90 *** ** *** *** *** *** 97 *** *** ** *** *** *** *** *** *** *** 100 96 97 95 93 95 94 94 93 100 76.9 Mean PASI improvement (%) 87 86 88 86 77.6 70.0 84 85 80.2 75.6 81 82 73.3 75 75.5 74.3 73.1 80 80 37.5 37.5 Patients (%) from baseline 45.7 60 60 46.0 42.5 41.9 34.6 33.3 31.6 40 40 20 20 n = 138 n = 169 n = 287 n = 428 n = 142 n = 166 n = 153 n = 146 n = 152 n = 143 n = 376 n = 222 n = 134 n = 126 n = 111 n = 46 n = 63 n = 88 n = 55 n = 48 n = 49 n = 49 n = 51 n = 78 n = 58 n = 20 n = 26 n = 60 n = 73 n = 43 n = 35 n = 26 n = 18 n = 19 n=8 n=8 0 0 <25 25-<30 ≥30 ≤100 >100 1st 2nd 3rd 4th Naive Any TNFi IL-17i Any 1 ≥2 y y ar ar im nd BMI Weight (Kg) Weight quartiles† Pr co Se 6 Biologic exposure Biologic failure Figure 3 Mean PASI improvement (%) from baseline to week 52 for patients treated with risankizumab and/or ustekinumab by BMI and body weight quartiles (LOCF)†. BMI, body mass index; LOCF, (b) Week 52: PASI 90 last observation carried forward; PASI, Psoriasis Area and Severity *** *** *** *** *** *** * *** Index; RZB, risankizumab; UST, ustekinumab. **P < 0.01, 100 84.5 83.3 70.0 73.1 ***P < 0.001 compared with ustekinumab. Data expressed as mean 81.6 80.6 81.3 78.4 80.8 50.0 per cent PASI improvement  standard error (SE). †Weight quar- 80 Patients (%) tiles (minimum – maximum) were defined in kg as follows: 1st (43.5– 53.2 42.9 60 33.3 25.0 74.0), 2nd (74.2–86.8), 3rd (87.0–103.0) and 4th (103.2–170.0). 37.0 32.6 30.8 40 15.8 biologic therapy exposure was reported in 398 (49.9%) and 295 20 n = 376 n = 222 n = 134 n = 126 n = 111 n = 78 n = 58 n = 20 n = 26 n = 60 n = 73 n = 43 n = 35 n = 26 n = 18 n = 19 n=8 n=8 (37.0%) patients, respectively. 0 Naive Any TNFi IL-17i Any 1 ≥2 y y ar ar im nd Efficacy outcomes Pr co Se Risankizumab demonstrated superior efficacy compared with Biologic exposure Biologic failure ustekinumab regardless of patient subgroup (baseline demo- graphics, disease characteristics or prior biologic exposure). Across all patient subgroups analysed, a significantly greater pro- Figure 4 Proportion of patients (non-responder imputation) portion of patients receiving risankizumab achieved PASI 90 achieving PASI 90 responses at week 16 (a) and week 52 (b) by responses at week 16 (70.3–82.2%) and week 52 (77.6–85.9%) prior biologic treatment experience (NRI). IL-17i, interleukin-17 inhibitors; NRI, non-responder imputation; PASI, Psoriasis Area compared with those receiving ustekinumab (34.6–55.6% and and Severity Index; RZB, risankizumab; TNFi, tumour necrosis 30.8–56.3%, respectively, all P < 0.01; Fig. 2). Similar results factor inhibitor; UST, ustekinumab. Data expressed as per cent of were observed with sPGA 0 or 1, sPGA 0 (data not shown) and patients with 95% confidence intervals (upper bounds shown). PASI 100 responses at week 16 and 52 (Figs S1 and S2, Support- *P < 0.05, ***P < 0.001 compared with ustekinumab. ing Information). The per cent improvement in PASI for indi- vidual patients taking risankizumab overall and subgroups of patients with PASI of 18 or above at baseline, BMI of 30 or PASI 26 and PASI 33) show substantial PASI improvement in higher, weight of 100 kg or greater, and who failed one or more patients treated with risankizumab from baseline to week 52 biologics can be observed in time-lapse videos provided as (Fig. S4, Supporting Information). Videos S1–S5 (Supporting Information). A significantly greater proportion of patients receiving Patients taking risankizumab achieved statistically greater risankizumab achieved PASI 90 response at week 16 (73.1– mean improvement in PASI scores (per cent change from base- 80.2%) and week 52 (78.4–81.6%) compared with those receiv- line) at week 52 compared with patients receiving ustekinumab ing ustekinumab (41.9–46.0% and 32.6–53.2%, respectively, all by BMI (<25, 25–<30, ≥30), weight (≤100 kg or >100 kg) and P < 0.001) regardless of the prior biologic exposure (Fig. 4). For weight quartiles (Fig. 3). Mean PASI improvement (per cent patients with prior biologic failure, a numerically greater pro- change from baseline) among patients receiving risankizumab portion of patients receiving risankizumab achieved PASI 90 remained high (88–98%) across all weight deciles at week 52 response at week 16 (70.0–77.6%) and week 52 (70.0–84.5%) (Fig. S3, Supporting Information). Representative images of compared with those receiving ustekinumab (31.6–37.5% and heavier patients with different baseline disease severity (PASI 15, 15.8–50.0%, respectively) regardless of the number or type of JEADV 2020, 34, 2830–2838 © 2020 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology 2836 Strober et al. prior failures (Fig. 4). Similar results were observed with sPGA 0 BMI or body weight and in patients with prior biologic expo- or 1, sPGA 0 (data not shown) and PASI 100 responses (Figs S5 sure or failure. and S6, Supporting Information). There are limitations to this study inherent to its design. This was a post hoc subgroup analysis of two replicate phase Multivariate logistic regression analyses 3 trials. The self-report of prior biologic exposure could be Logistic regression analyses were conducted in order to assess subject to recall errors and biases. Another limitation is the whether risankizumab demonstrated superior efficacy com- lack of information on disease duration for patients enrolled pared with ustekinumab at either week 16 or 52 regardless of in these trials. Psoriasis disease duration has been shown to patient characteristics (age, sex, weight, baseline PASI score negatively impact efficacy of other biologic treatments for pso- and presence of psoriatic arthritis). All logistic regression riasis.18 Further clinical or real-world studies are warranted to models at week 16 and week 52 demonstrated that risankizu- assess the impact of psoriasis disease duration on the efficacy mab treatment led to and maintained superior efficacy, of risankizumab. Some strengths are the inclusion of an active respectively (all P < 0.001), compared with ustekinumab for comparator through 52 weeks, inclusion of a high percentage PASI 90, PASI 100 and sPGA 0 or 1 (Table S3, Supporting of historically difficult-to-treat patients (prior biologic therapy Information); this analysis yielded similar results for sPGA 0 exposure, high disease severity and high average BMI), and (data not shown) as for PASI 100. the inclusion of patients across the globe allowing for general- izability of the findings. An additional strength is the inclusion Discussion of the logistic regression analyses to control for the effect of In this integrated analysis, risankizumab treatment resulted in five independent covariates (age, sex, weight, baseline PASI statistically greater proportions of patients achieving high skin score and presence of psoriatic arthritis) and their potential clearance compared with ustekinumab at both weeks 16 and 52, interactions on the efficacy of risankizumab compared with regardless of baseline patient demographics, disease characteris- ustekinumab. These multivariate analyses aimed to better rep- tics, or prior biologic therapy exposure. Response rates with resent patients routinely seen in clinical practice where the risankizumab treatment were consistently high across all sub- presence of multiple factors could affect efficacy. Risankizu- groups analysed at both weeks 16 and 52. The proportion mab demonstrated superior efficacy compared with ustek- achieving PASI 90 or sPGA 0 or 1 on risankizumab treat- inumab in all models tested, which included the covariate of ment remained relatively stable across subgroups from week 16 weight (>100 kg) known to adversely affect efficacy of other to week 52, while risankizumab treatment showed an increase biologic treatments.5–12 in the proportion of patients achieving complete clearance In summary, risankizumab demonstrated superior efficacy (PASI 100) from week 16 to week 52 across subgroups analysed. compared with ustekinumab through 52 weeks of treatment Logistic regression analyses revealed that risankizumab demon- regardless of patient baseline demographics, disease characteris- strated superior efficacy compared with ustekinumab at weeks tics or prior biologic therapy exposure. These results support 16 and 52 even when accounting for effects of five independent risankizumab as a psoriasis treatment suitable for a wide variety covariates typically associated with reduced efficacy responses of patients given its high and durable skin clearance across all with other commonly used biologics. subgroups analysed. Despite the availability of multiple classes of biologics with different treatment targets, psoriasis remains undertreated par- Acknowledgements tially due to inadequate efficacy or loss of efficacy over time Medical writing support was provided by Daniel O’Brien, Ph.D., with current therapies.28 In clinical practice, this has resulted of AbbVie. Writing support was also provided by Janet in dose optimization in order to increase or maintain clinical Matsuura, Ph.D., of Complete Publication Solutions, LLC; this responses.29–31 In particular, patients with higher BMI or body support was funded by AbbVie. weight have demonstrated lower response rates with many fixed-dose biologic treatments,5,6,8–12 leading some biologics to References be used with weight-based dosing or more frequent dosing 1 Harden JL, Krueger JG, Bowcock AM. The immunogenetics of psoriasis: schedules in order to achieve comparable efficacy in these a comprehensive review. J Autoimmun 2015; 64: 66–73. patients.7,32,33 Another factor that adversely affects efficacy 2 World Health Organization. Global Report on Psoriasis. World Health with some biologics is prior exposure to or failure on a bio- Organization, Geneva, 2016. 3 Rapp SR, Feldman SR, Exum L, Fleischer AB, JrReboussin DM. Psoriasis logic.13–16 In this analysis, risankizumab achieved superior effi- causes as much disability as other major medical diseases. J Am Acad Der- cacy compared with ustekinumab in patients with higher BMI matol 1999; 41: 401–407. or body weight and any prior biologic exposure or failure. 4 Lonnberg AS, Skov L, Skytthe A, Kyvik KO, Pedersen OB, Thomsen SF. Association of psoriasis with the risk for type 2 diabetes mellitus and obe- Thus, these data support the use of risankizumab without dose sity. JAMA Dermatol 2016; 152: 761–767. adjustment to treat all patients, including those with higher JEADV 2020, 34, 2830–2838 © 2020 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology RZB efficacy in patient subgroups 2837 5 Clark L, Lebwohl M. The effect of weight on the efficacy of biologic ther- for the treatment of patients with moderate to severe psoriasis with ran- apy in patients with psoriasis. J Am Acad Dermatol 2008; 58: 443–446. domized withdrawal and retreatment: results from the phase III, double- 6 Hsu S, Green LJ, Lebwohl MG, Wu JJ, Blauvelt A, Jacobson AA. Compa- blind, placebo- and active comparator-controlled VOYAGE 2 trial. J Am rable efficacy and safety of brodalumab in obese and non-obese patients Acad Dermatol 2017; 76: 418–431. with psoriasis: analysis of 2 randomized controlled trials. Br J Dermatol 22 Gordon KB, Strober B, Lebwohl M et al. Efficacy and safety of risanki- 2020.182: 880–888. zumab in moderate-to-severe plaque psoriasis (UltIMMa-1 and 7 Lebwohl M, Yeilding N, Szapary P et al. Impact of weight on the efficacy UltIMMa-2): results from two double-blind, randomised, placebo-con- and safety of ustekinumab in patients with moderate to severe psoriasis: trolled and ustekinumab-controlled phase 3 trials. Lancet 2018; 392: rationale for dosing recommendations. J Am Acad Dermatol 2010; 63: 650–661. 571–579. 23 Reich K, Gooderham M, Thaci D et al. Risankizumab compared with 8 Lee JE, Wang J, Florian J et al. Effect of body weight on risk-benefit and adalimumab in patients with moderate-to-severe plaque psoriasis dosing regimen recommendation of secukinumab for the treatment of (IMMvent): a randomised, double-blind, active-comparator-controlled moderate to severe plaque psoriasis. Clin Pharmacol Ther 2019; 106: 78– phase 3 trial. Lancet 2019; 394: 576–586. 80. 24 Krueger JG, Ferris LK, Menter A et al. Anti-IL-23A mAb BI 655066 for 9 Naldi L, Addis A, Chimenti S et al. Impact of body mass index and obe- treatment of moderate-to-severe psoriasis: safety, efficacy, pharmacoki- sity on clinical response to systemic treatment for psoriasis. Evidence netics, and biomarker results of a single-rising-dose, randomized, double- from the Psocare project. Dermatology 2008; 217: 365–373. blind, placebo-controlled trial. J Allergy Clin Immunol 2015; 136: 116– 10 Papp KA, Reich K, Blauvelt A et al. Efficacy of tildrakizumab for moder- 124.e7. ate-to-severe plaque psoriasis: pooled analysis of three randomized con- 25 McKeage K, Duggan S. Risankizumab: first global approval. Drugs 2019; trolled trials at weeks 12 and 28. J Eur Acad Dermatol Venereol 2019; 33: 79: 893–900. 1098–1106. 26 Khatri A, Suleiman AA, Polepally AR, Othman AA. Exposure-response 11 Prussick R, Unnebrink K, Valdecantos WC. Efficacy of adalimumab com- Relationship for Efficacy and Safety of Risankizumab in Patients with pared with methotrexate or placebo stratified by baseline BMI in a ran- Moderate-to-severe Plaque Psoriasis: Integrated Analyses of Phase 2 and domized placebo-controlled trial in patients with psoriasis. J Drugs 3 Clinical Trials. Presented at the American Academy of Dermatology Dermatol 2015; 14: 864–868. (AAD) meeting, Washington, DC, 2019. 12 Reich K, Puig L, Mallbris L, Zhang L, Osuntokun O, Leonardi C. The 27 Blauvelt A, Papp KA, Gooderham M et al. Efficacy and safety of risanki- effect of bodyweight on the efficacy and safety of ixekizumab: results from zumab, an interleukin-23 inhibitor, in patients with moderate-to-severe an integrated database of three randomised, controlled Phase 3 studies of chronic plaque psoriasis: 16-week results from the phase III IMMhance patients with moderate-to-severe plaque psoriasis. J Eur Acad Dermatol trial. Psoriasis Gene to Clinic, 8th International Congress; December 2, Venereol 2017; 31: 1196–1207. London, UK, 2017. 13 Mazzotta A, Esposito M, Costanzo A, Chimenti S. Efficacy and safety of 28 Menter A, Strober BE, Kaplan DH et al. Joint AAD-NPF guidelines of etanercept in psoriasis after switching from other treatments: an observa- care for the management and treatment of psoriasis with biologics. J Am tional study. Am J Clin Dermatol 2009; 10: 319–324. Acad Dermatol 2019; 80: 1029–1072. 14 Papp KA, Gordon KB, Langley RG et al. Impact of previous biologic use 29 Brezinski EA, Armstrong AW. Off-label biologic regimens in psoriasis: a on the efficacy and safety of brodalumab and ustekinumab in patients systematic review of efficacy and safety of dose escalation, reduction, and with moderate-to-severe plaque psoriasis: integrated analysis of the ran- interrupted biologic therapy. PLoS ONE 2012; 7: e33486. domized controlled trials AMAGINE-2 and AMAGINE-3. Br J Dermatol 30 Carrascosa JM, Garcia-Doval I, Perez-Zafrilla B et al. Use of off-label 2018; 179: 320–328. doses is frequent in biologic therapy for moderate to severe psoriasis: a 15 Cassano N, Galluccio A, De Simone C et al. Influence of body mass index, cross-sectional study in clinical practice. J Dermatolog Treat 2015; 26: comorbidities and prior systemic therapies on the response of psoriasis to 502–506. adalimumab: an exploratory analysis from the APHRODITE data. J Biol 31 Esposito M, Gisondi P, Conti A et al. Dose adjustment of biologic thera- Regul Homeost Agents 2008; 22: 233–237. pies for psoriasis in dermatological practice: a retrospective study. J Eur 16 Gottlieb AB, Lacour JP, Korman N et al. Treatment outcomes with ixek- Acad Dermatol Venereol 2017; 31: 863–869. izumab in patients with moderate-to-severe psoriasis who have or have 32 Reich K, Puig L, Szepietowski JC et al. Secukinumab dosing optimiza- not received prior biological therapies: an integrated analysis of two Phase tion in patients with moderate to severe plaque psoriasis: results from III randomized studies. J Eur Acad Dermatol Venereol 2017; 31: 679–685. the randomised, open-label OPTIMISE study. Br J Dermatol 2020.182: 17 Hawkes JE, Yan BY, Chan TC, Krueger JG. Discovery of the IL-23/IL-17 304–315. signaling pathway and the treatment of psoriasis. J Immunol 2018; 201: 33 Langley RG, Lebwohl M, Krueger GG et al. Long-term efficacy and 1605–1613. safety of ustekinumab, with and without dosing adjustment, in 18 Papp KA, Langley RG, Lebwohl M et al. Efficacy and safety of ustek- patients with moderate-to-severe psoriasis: results from the PHOE- inumab, a human interleukin-12/23 monoclonal antibody, in patients NIX 2 study through 5 years of follow-up. Br J Dermatol 2015; 172: with psoriasis: 52-week results from a randomised, double-blind, pla- 1371–1383. cebo-controlled trial (PHOENIX 2). Lancet 2008; 371: 1675–1684. 19 Reich K, Papp KA, Blauvelt A et al. Tildrakizumab versus placebo or etan- ercept for chronic plaque psoriasis (reSURFACE 1 and reSURFACE 2): Supporting information results from two randomised controlled, phase 3 trials. Lancet 2017; 390: 276–288. Additional Supporting Information may be found in the online 20 Blauvelt A, Papp KA, Griffiths CE et al. Efficacy and safety of guselkumab, version of this article: an anti-interleukin-23 monoclonal antibody, compared with adalimumab for the continuous treatment of patients with moderate to severe psoria- sis: results from the phase III, double-blinded, placebo- and active com- Figure S1. Proportion of patients (non-responder imputation) parator-controlled VOYAGE 1 trial. J Am Acad Dermatol 2017; 76: 405– achieving sPGA 0/1 (a) and PASI 100 (b) responses at week 16 417. by baseline patient demographics, disease characteristics and 21 Reich K, Armstrong AW, Foley P et al. Efficacy and safety of guselkumab, an anti-interleukin-23 monoclonal antibody, compared with adalimumab prior biologic therapy. JEADV 2020, 34, 2830–2838 © 2020 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology 2838 Strober et al. Figure S2. Proportion of patients (non-responder imputation) Table S3. Logistic regression models tested comparing propor- achieving sPGA 0/1 (a) and PASI 100 (b) responses at week 52 tion of patients achieving efficacy thresholds on risankizumab by baseline patient demographics, disease characteristics and compared with ustekinumab at weeks 16 and 52. prior biologic therapy. Video S1. Time-lapse of individual patients’ percent improve- Figure S3. Mean PASI improvement from baseline to week 52 by ment in PASI through 52 weeks of risankizumab treatment. weight deciles in patients treated with risankizumab (LOCF)*. Video S2. Time-lapse of individual patients’ percent improve- Figure S4. Representative images from 3 patients (a–c) treated ment in PASI through 52 weeks of risankizumab treatment: sub- with risankizumab at baseline, week 4, week 16, and week 52. group of patients with PASI 18 or above. Figure S5. Proportion of patients (non-responder imputation) Video S3. Time-lapse of individual patients’ percent improve- achieving sPGA 0/1 (a) and PASI 100 (b) responses at week 16 ment in PASI through 52 weeks of risankizumab treatment: sub- by prior treatment experience. group of patients with body mass index of 30 or more. Figure S6. Proportion of patients (non-responder imputation) Video S4. Time-lapse of individual patients’ percent improve- achieving sPGA 0/1 (a) and sPGA 0 (b) responses at week 52 by ment in PASI through 52 weeks of risankizumab treatment: sub- prior treatment experience. group of patients with weight of 100 kg and above. Table S1. Inclusion and exclusion criteria. Video S5. Time-lapse of individual patients’ percent improve- Table S2. Independent variables and their interactions analyzed ment in PASI through 52 weeks of risankizumab treatment: sub- for their effects on risankizumab compared with ustekinumab group of patients who failed 1 or more biologics. using logistic regression. JEADV 2020, 34, 2830–2838 © 2020 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology