Post-induction IL-12/23 levels correlate with clinical response and remission after ustekinumab in a multicenter prospective cohort study of patients with Crohn's disease: Preliminary induction results of the AURORA study, including the Australia New Zealand IBD Consortium Cohort

Background and Aim: Ustekinumab (UST) is a monoclonal antibody targeting IL-12 and IL-23 through their shared p40 subunit. Although UST was approved for the treatment of moderate to severe Crohn's disease (CD) in Australia in September 2017, there are limited data investigating its utility in real-world clinical practice. Further, there are no prospective Australian studies to date. The aims of this study were to determine the clinical outcomes after UST treatment in patients with CD in a real-world setting and to investigate the association of clinical outcomes with IL-12/23 levels.

Methods: We conducted a nationwide, multicenter, prospective observational cohort study of UST therapy for moderate to severe CD meeting Pharmaceutical Benefits Scheme criteria for biologic treatment. Patients were recruited from 19 Australian inflammatory bowel disease (IBD) centers, including 12 sites within the Australia New Zealand IBD Consortium (ANZIBDC), between September 2019 and April 2022. Baseline clinical assessments were made at Study Visit 1 (SV1), and post-induction clinical assessments were made at Study Visit 2 (SV2). Clinical response and remission rates were analyzed after induction using two-item patient-reported outcome score (PRO2) definitions from STRIDE II guidelines. Univariate and multivariate logistic regression analyses were performed to identify predictors of clinical response and remission.

Results: The study recruited a total of 200 patients, with the preliminary induction data on 114 patients reported here. Patients had a mean age of 41.2 years, 42% were male, the mean duration of disease was 10.4 years, and 12% were active smokers. Fifty-five patients were biologic-naïve (48.2%), and 59 (51.8%) were previously biologic-exposed (infliximab, 29; adalimumab, 39; vedolizumab, 5). Fifty-nine patients (51.8%) received concomitant immunomodulators during induction. Clinical response was achieved in 73 patients (64.0%), and remission in 47 patients (41.2%). Clinical response correlated inversely with prior surgery, prior biologic exposure, and high baseline C-reactive protein (CRP) level. Clinical remission correlated inversely with longer disease duration, perianal disease, prior surgery, presence of extraintestinal manifestations, and high baseline CRP level. Compared with mean serum baseline IL-12p40 and IL-23p19 levels, post-induction (SV2) levels of both cytokines were significantly reduced (P < 0.0001) across the entire cohort. In patients determined to have clinically responded at SV2, no difference in IL-12p40 and IL-23p19 levels was observed at baseline compared with nonresponders (Fig. 1). However, after induction (SV2), patients determined to have clinically responded to treatment had significantly lower IL-12p40 and IL-23p19 levels (P = 0.0002) compared with baseline (SV1). Clinical nonresponders also had a significant reduction in IL-12p40 levels (P = 0.03); however, no significant reduction in IL-23p19 levels was observed. In patients who had clinically responded to treatment, a significant association between IL-12p40 and IL-23p19 levels and fecal calprotectin (FCP) levels was observed at SV2 (P = 0.004 and P = 0.005, respectively). No significant association between serum cytokine levels and FCP level was observed in clinical nonresponders. From the available UST levels (n = 67) at SV2, IL-12p40 and IL-23p19 levels were significantly inversely correlated across the entire cohort (P = 0.00008 and P = 0.00002, respectively); however, no significant difference between UST levels in clinical responders compared with nonresponders was observed.

Conclusion: This is the first study to show that decreases in serum IL-12p40 and IL-23p19 levels correlate with clinical response following UST induction for refractory CD.