Abstract Background Whilst upadacitinib has demonstrated efficacy in Crohn’s disease (CD) trials, predicting individual treatment response remains challenging. No validated prediction models exist to guide clinical decisions regarding upadacitinib therapy. This study aimed to evaluate real-world effectiveness of upadacitinib in Chinese CD patients and develop a practical prediction model for 12-week clinical remission using readily available baseline. Methods This multicentre retrospective study analysed 140 Chinese CD patients treated with upadacitinib (45 mg/day). The primary outcome was clinical remission (Crohn’s Disease Activity Index 150) at week 12. Predictors were selected from 28 candidate variables using Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression with 5-fold cross-validation. Model performance was assessed via discrimination (area under the receiver operating characteristic curve AUC), calibration (Hosmer-Lemeshow test), and clinical utility (decision curve analysis DCA). Internal validation employed 500 bootstrap replicates to calculate optimism-corrected metrics. Results The cohort was highly refractory (99.3% prior biologic exposure; 80.0% perianal disease). At week 12, 75.7% (106/140) achieved clinical remission. The final model incorporated three predictors: baseline albumin ≥35 g/L (odds ratio OR 5.28, 95% CI 2.08–14.35, p 0.001), bowel wall thickness (OR 1.30 per mm, 95% CI 1.05–1.67, p = 0.028), and upadacitinib as ≥ 3rd-line therapy (negative predictor; OR 0.25, 95% CI 0.06–0.85, p = 0.038). The model demonstrated good discrimination (AUC 0.774, 95% CI 0.691–0.857) and calibration (p = 0.979). Internal validation confirmed stability (optimism-corrected AUC 0.756). DCA revealed meaningful net benefit at threshold probabilities 20%. Adverse events occurred in 16.4% (predominantly acne, 8.6%), with no serious adverse events reported. Conclusion Upadacitinib demonstrated substantial real-world effectiveness in Chinese CD patients with refractory disease characteristics. We developed and internally validated the first clinical prediction model for upadacitinib response in CD, incorporating three readily obtainable parameters with good predictive accuracy. This tool may assist clinicians in identifying patients most likely to benefit from upadacitinib therapy and facilitate shared decision-making. External validation in independent cohorts is warranted before clinical implementation. References: 1. Torres J, Mehandru S, Colombel JF, Peyrin-Biroulet L. Crohn’s disease. Lancet. 2017;389(10080):1741-1755. 2. Ananthakrishnan AN. Epidemiology and risk factors for IBD. Nat Rev Gastroenterol Hepatol. 2015;12(4):205-217. 3. Xu L, He B, Sun Y, et al. Incidence of Inflammatory Bowel Disease in Urban China: A Nationwide Population-based Study. Clin Gastroenterol Hepatol. 2023;21(13):3379-3386.e3329. 4. Peyrin-Biroulet L, Sandborn W, Sands BE, et al. Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE): Determining Therapeutic Goals for Treat-to-Target. Am J Gastroenterol. 2015;110(9):1324-1338. 5. Chanchlani N, Lin S, Bewshea C, et al. Mechanisms and management of loss of response to anti-TNF therapy for patients with Crohn’s disease: 3-year data from the prospective, multicentre PANTS cohort study. Lancet Gastroenterol Hepatol. 2024;9(6):521-538. 6. Burisch J, Claytor J, Hernandez I, Hou JK, Kaplan GG. The Cost of Inflammatory Bowel Disease Care: How to Make it Sustainable. Clin Gastroenterol Hepatol. 2025;23(3):386-395. 7. Noor NM, Verstockt B, Parkes M, Lee JC. Personalised medicine in Crohn’s disease. Lancet Gastroenterol Hepatol. 2020;5(1):80-92. 8. Bonelli M, Kerschbaumer A, Kastrati K, et al. Selectivity, efficacy and safety of JAKinibs: new evidence for a still evolving story. Ann Rheum Dis. 2024;83(2):139-160. Conflict of interest: Wu, Hongzhen: No conflict of interest Chen, Miaoyan: No conflict of interest Su, Tao: No conflict of interest Chen, Xiaomin: No conflict of interest Wu, Luying: No conflict of interest Zhi, Min: No conflict of interest Prof. Yao, Jiayin: No conflict of interest
Wu et al. (Thu,) studied this question.