25 Efficacy of IL-12/23 and IL-23 Antagonists for Moderate to Severe Crohn’s Disease in Advanced Treatment Failure Populations: A Network Meta-Analysis
Bibliographic record
Abstract
Background: Recent approvals of advanced therapies have broadened the treatment landscape for Crohn’s disease (CD), providing patients with more options. While earlier clinical trials often used responder re-randomization designs, some newer therapies have been evaluated in treat-through trials. Combining heterogenous trial designs violates core assumptions of transitivity in network meta-analysis (NMA) and may lead to biased comparative efficacy estimates, especially for therapies with long half-lives that may impact withdrawal to placebo response rates in maintenance. This study aimed to compare the efficacy of interleukin (IL)-23 and IL-12/23 inhibitors for the treatment of moderate to severe CD using NMA focused on treat-through trials in advanced treatment failure populations. Methods: Indirect estimates were produced between risankizumab (RZB) 600 mg intravenous (IV) at weeks 0, 4, and 8 followed by RZB 360 mg subcutaneous (SC) every 8 weeks (Q8W), guselkumab (GUS) 200 mg IV or 400 mg SC at weeks 0, 4, and 8 followed by GUS 100 mg SC Q8W (GUS100) or GUS 200 mg SC Q4W (GUS200), mirikizumab (MIR) 900 mg IV at weeks 0, 4, and 8 followed by MIRI 300 mg Q4W, and ustekinumab (UST) 6 mg/kg IV at week 0 followed by UST 90 mg Q8W. Outcomes evaluated included clinical remission (Crohn’s Disease Activity Index < 150), endoscopic response (≥50% reduction in Simple Endoscopic Score for CD [SES-CD] score from baseline), and endoscopic remission (SES-CD ≤4 and a ≥2-point reduction from baseline and no subscore >1 in any individual component) at approximately 1 year of treatment (48–52 weeks). The NMA was conducted in a Bayesian framework utilizing placebo as the primary reference comparator of the network; outcomes were modeled with a binomial likelihood and a risk difference (RD)-link function. A statistically significant difference between comparators was declared if the 95% credible intervals (CrI) between the 2 comparators excluded the null value on the RD scale (i.e., RD of 0). Both fixed effects and random effects models were produced, with fixed effects models selected on the basis of parsimony. Results: All assessed therapies demonstrated a significantly greater RD across all outcomes relative to placebo, with RZB demonstrating the highest RD compared to placebo (clinical remission: 0.520 [95% CrI: 0.398, 0.639]; endoscopic response: 0.549 [0.438, 0.659]; endoscopic remission: 0.317 [0.223, 0.413]). The RDs in clinical remission for RZB (0.199 [0.114, 0.282]), GUS200 (0.141 [0.045, 0.235]), and GUS100 (0.108 [0.013, 0.201]) were significant relative to UST. For endoscopic response, the RD for RZB was significant relative to GUS100 (0.141 [0.022, 0.259]), MIRI (0.172 [0.052, 0.292]), and UST (0.230 [0.151, 0.308]). Additionally, the RDs for GUS200 (0.166 [0.074, 0.256]) and GUS100 (0.090 [0.001, 0.179]) were significant relative to UST. For endoscopic remission, the RD for RZB was significant relative to MIRI (0.112 [0.008, 0.218]) and UST (0.154 [0.082, 0.226]); the RD for GUS200 was significant relative to UST (0.117 [0.037, 0.197]). Conclusions: In this NMA of IL-23 and IL-12/23 inhibitors for moderate to severe CD in advanced treatment failure populations, RZB demonstrated the highest RD of achieving the evaluated outcomes compared to other IL-23 and IL-12/23 inhibitors.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.044 | 0.052 |
| Meta-epidemiology (narrow) | 0.004 | 0.002 |
| Meta-epidemiology (broad) | 0.013 | 0.068 |
| Bibliometrics | 0.006 | 0.005 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".