Effect of Costimulatory Blockade With Abatacept After Ustekinumab Withdrawal in Patients With Moderate to Severe Plaque Psoriasis: The PAUSE Randomized Clinical Trial

Abstract

Importance: Psoriasis relapse may involve compensatory T-cell activation pathways in the presence of CD28-CD80/CD86 blockade with abatacept.

Objective: To determine whether costimulatory signaling blockade with abatacept prevents psoriasis relapse after ustekinumab withdrawal.

Design, setting, and participants: Psoriasis Treatment with Abatacept and Ustekinumab: a Study of Efficacy (PAUSE), a parallel-design, double-blind, placebo-controlled randomized clinical trial, was conducted at 10 sites in the US and Canada. Participant enrollment opened on March 19, 2014, and concluded on April 11, 2016. Participants were adults with moderate to severe plaque psoriasis and received ustekinumab in a lead-in phase. Those who responded to ustekinumab at week 12 were randomized 1:1 to either the continued with ustekinumab group (ustekinumab group) or the switched to abatacept group (abatacept group). Treatment was discontinued at week 39, and participants were followed up for psoriasis relapse until week 88. Statistical analyses were performed in the intention-to-treat (ITT) and safety samples from May 3, 2018, to July 6, 2021.

Interventions: Participants received subcutaneous ustekinumab at weeks 0 and 4 (45 mg per dose for those ≤100 kg; 90 mg per dose for those >100 kg). Participants randomized to the abatacept group at week 12 received subcutaneous abatacept, 125 mg weekly, from weeks 12 to 39 and ustekinumab placebo at weeks 16 and 28. Participants randomized to the ustekinumab group received ustekinumab at weeks 16 and 28 and abatacept placebo weekly from weeks 12 to 39.

Main outcomes and measures: The primary end point was the proportion of participants with psoriasis relapse (loss of ≥50% of the initial Psoriasis Area and Severity Index improvement) between weeks 12 and 88. Secondary end points included time to psoriasis relapse, proportion of participants with psoriasis relapse between weeks 12 and 40, and adverse events. The psoriasis transcriptome and serum cytokines were evaluated.

Results: A total of 108 participants (mean [SD] age, 46.1 [12.1] years; 73 [67.6%] men) were treated with open-label ustekinumab; 91 were randomized to blinded treatment. Similar proportions of participants in the abatacept group and the ustekinumab group relapsed between weeks 12 and 88 (41 of 45 [91.1%] vs 40 of 46 [87.0%]; P = .41). Median time to relapse from the last dose of ustekinumab was similar between groups as well: 36 weeks (95% CI, 36-48 weeks) in the abatacept group vs 32 weeks (95% CI, 28-40 weeks) in the ustekinumab group. Similar numbers and rates of adverse events occurred. Abatacept did not maintain suppression of the pathogenic IL-23-mediated psoriasis molecular signature in lesions after ustekinumab withdrawal, and serum IL-19 levels increased.

Conclusions and relevance: This parallel-design, double-blind randomized clinical trial found that abatacept did not prevent psoriasis relapse that occurred after ustekinumab withdrawal because it did not completely block the pathogenic psoriasis molecular pathways that led to relapse.

Trial registration: ClinicalTrials.gov Identifier: NCT01999868.

Conflict of interest statement

Conflict of Interest Disclosures: Dr Korman reported receiving grants and personal fees from AbbVie, Eli Lilly, Leo Pharma, Principia, and Trevi; grants from Amgen, Celgene, Chemocentryx, Dermira, Menlo Therapeutics, Syntimmune, and XBiotech; and personal fees from Genentech, Janssen, Novartis, Regeneron, Sun Pharma, and UCB. Dr Feldman reported receiving grants from AbbVie, Amgen, Janssen, Lilly, and UCB and personal fees from AbbVie, Amgen, Janssen, Lilly, Novartis, Sun, and UCB outside the submitted work. Dr Boh reported receiving grants from Janssen during the conduct of the study; personal fees from Sun and UCB; and grants from AbbVie, Janssen, and Incyte outside the submitted work. Dr Helfrich reported receiving grants from Immune Tolerance Network during the conduct of the study and grants from AbbVie, Boehringer Ingelheim, Eli Lilly, and Novartis outside the submitted work. Dr Sofen reported receiving grants from Janssen outside the submitted work. Dr Bissonnette reported receiving grants from AbbVie, AnaptysBio, Arcutis, Aristea, Bausch Health, Boehringer Ingelheim, Bristol Myers Squibb, Dermavant, Eli Lilly, Escalier, Janssen, Leo Pharma, Nimbus, Regeneron, Sienna, and UCB outside the submitted work; personal fees from AbbVie, Almirall, Amgen, Arcutis, Boehringer Ingelheim, Bristol Myers Squibb, Dermavant, Eli Lilly, Janssen, Kyowa Kirin, Leo Pharma, and Pfizer; and being a shareholder with Innovaderm Research. Dr J. G. Krueger reported receiving grants from Immune Tolerance Network during the conduct of the study; money for his institution from Novartis, Pfizer, Amgen, Lilly, Boehringer Ingelheim, Innovaderm, Bristol Myers Squibb, Janssen, AbbVie, Paraxel, Regeneron, Allergan, Novan, Biogen MA, Sienna, UCB, Botanix, Incyte, Avillion, Execure, Nimbus, and Arista; personal fees from Novartis, Pfizer, Amgen, Lilly, Boehringer Ingelheim, AbbVie, Leo Pharma, Biogen Idec, Valeant, Aurigne, Allergan, Asana, UCB, Siena, Celgene, Nimbus, Sanofi, Sun Pharma, Allmiral, Arena, Bristol Myers Squibb, Ventyx, Aclaris, and Galapagos outside the submitted work. No other disclosures were reported.

Figures

Figure 1.. CONSORT Diagram

Figure 2.. Time to Psoriasis Relapse

Time to psoriasis relapse curves for the switched to abatacept group (abatacept group) and the continued with ustekinumab group (ustekinumab group) are displayed as the estimated survival function for interval-censored data. For participants who experienced a psoriasis relapse, the censoring interval was derived from the time points of the participants’ last 2 Psoriasis Area and Severity Index (PASI) evaluations, which were typically 4 weeks apart. For participants who did not experience a psoriasis relapse, the time to relapse was right-censored at the time point of the last PASI evaluation. For participants randomized to the abatacept group, the last dose of ustekinumab was administered at week 4. For participants randomized to the ustekinumab group, the last dose of ustekinumab was administered at week 28. ITT indicates intention to treat.

Figure 3.. Modulation of the IL-23–Mediated Psoriasis Molecular Signature Genes in Participants Who Were Eligible for Randomization (≥PASI 75)

A, Venn diagram shows the number of genes in the disease transcriptome (lesions vs nonlesions at week 0) and not in the disease transcriptome that were significantly modulated by ustekinumab (at week 12: fold change, ≥1.5; false discovery rate z-transformed log expression values of key genes in the interleukin (IL)–23–mediated psoriasis molecular signature in paired nonlesions (pink group; left) and lesions (orange group; center) at week 0 and resolving lesions (indigo group; right) at week 12. C, Individual participant data are ordered by baseline Psoriasis Area and Severity Index (PASI) from low to high for the 3 groups. The z score indicates the number of SDs higher or lower than the mean expression value for each gene. DEG indicates differentially expressed gene.

Figure 4.. Suppression of the Psoriasis Molecular Signature in Resolving Lesions and Serum IL-19 Levels After Ustekinumab Withdrawal in Participants Who Completed Treatment

A, Line plot shows the eigengene (weighted mean expression of genes in the psoriasis molecular signature module) value for lesions at week 0 and resolving lesions at weeks 12, 24, and 40 by treatment group. B to F, Line plots show the mean concentration of interleukin (IL)–19, IL-17A, IL-22, IL-10, and IL-2 levels in serum at weeks 0, 12, 24, and 40 by treatment group. Error bars display the 95% CIs. P values between treatment groups were determined by mixed model for repeated measures with baseline adjustment. aSignificant difference (P < .05) between treatment groups.