AB0556 COMPARING EFFICACY OF GUSELKUMAB VERSUS USTEKINUMAB IN PATIENTS WITH PSORIASIS ARTHRITIS: AN ADJUSTED COMPARISON USING INDIVIDUAL PATIENT DATA FROM DISCOVER 1&2 AND PSUMMIT TRIALS

Background: Guselkumab is an anti-interleukin (IL)-23 monoclonal antibody recently approved for a treatment of Psoriasis arthritis (PsA). In two large Phase III trials of patients with PsA (DISCOVER -1 & -2) guselkumab has shown to be superior versus placebo. In this indication no direct comparison is available between guselkumab and ustekinumab, a monoclonal antibody targeting IL-12 and IL-23. Indirect comparisons based on relative treatment effects versus a common comparator (placebo) only allow for analyses up to week 24 due to cross-over to active arms in available PsA trials. Objectives: To compare indirectly joint and skin efficacy of guselkumab versus ustekinumab up to week 52, using pooled patient-level trial data from DISCOVER 1&2 and PSUMMIT 1&2, adjusting for cross-trial population differences. Methods: Patient level data, including baseline characteristics and outcome data on American College of Rheumatology (ACR) response, Psoriasis Area Severity Index (PASI) response from the guselkumab arms of DISCOVER -1 & -2 were pooled with the data from the ustekinumab trials PSUMMIT -1&-2. Analyses were performed for bio-naive and bio-experienced populations separately. Differences in patient characteristics across trial populations were adjusted for using multivariate logistic regression, including: gender, age, body mass index, previous TNF use, disease duration, PASI level, number of swollen and tender joints. This method of indirect comparisons allows for analysis of comparative efficacy beyond controlled induction period and odds ratios’ resulting from this model were translated into predicted response rates for ustekinumab, assuming same patient population, as enrolled in the guselkumab trial arms. Results: Majority of baseline characteristics for patients on guselkumab (100mg q8w; 100mg q4w) were comparable to patients on ustekinumab 45/90mg, in both in bio-naive and bio-experienced group of patients. The probability of reaching a ACR 20 in both the bio-naive & bio-experienced population was significantly higher for guselkumab vs ustekinumab at weeks 52 for both dosing regimens of guselkumab (bio-naive ACR 20: q8w OR= 1.88 [1.28;2.76]), q4w (OR= 1.92 [1.29;2.86]; bio experienced ACR20 q8w OR= 2.72[1.17;6.31], q4w OR=4.77 [1.95;11.63]). Similarly guselkumab was superior over ustekinumab on PASI 90 outcome at week 52 in both bio-naive & bio-experienced patients with BSA ≥3 % at baseline (bio-naive: q8w OR= 2.59 [1.68;3.99]), q4w OR= 3.19 [2.03;5.00], and bio-experienced q8w OR= 3.96[1.39,11.27], q4w OR=13.10[4.18,41.04]). Figure 1 represents unadjusted pooled DISCOVER 1&2 trial results and estimated proportions of ustekinumab treated patient group achieving ACR 20 in bio-naive patient group up to week 52 using the method described above. Conclusion: An adjusted comparison using patient level data from pivotal Phase III studies demonstrates both dosages of guselkumab to be significantly more effective versus ustekinumab in both skin and joint outcomes in both bio-naive & bio experienced patients up to week 52. Disclosure of Interests: Joris Diels Shareholder of: Janssen, Employee of: Janssen, Pushpike Thilakarathne Employee of: Janssen, Agata Schubert Shareholder of: Janssen, Employee of: Janssen, Fareen Hassan Shareholder of: Janssen, Employee of: Janssen, Steve Peterson Shareholder of: Janssen, Employee of: Janssen, Wim Noel Shareholder of: Janssen, Employee of: Janssen.