Background and Aims: A number of Janus kinase (JAK) inhibitors (tofacitinib, filgotinib, upadacitinib) have been tested for moderate and severe Crohn’s disease (CD) in randomized control trials (RCTs). However, data on their comparative efficacy and tolerability is lacking. We aimed to study their performance comparatively, by means of network meta-analysis (NWM).

Methods: We searched the Pubmed/Medline, EMBASE, and Cochrane Library databases for relevant RCTs through March 2021 and data was extracted. A bayesian NWM was performed to investigate the efficacy and tolerability of the above JAK inhibitors and to explore their rank order in treating moderate and severe CD patients. The cumulative ranking probability for each intervention at the end of treatment period, was evaluated by means of surfaces under cumulative ranking (SUCRA) values.

Results: Four RCTs were entered into this NWM. They included 811 patients totally, randomized to 11 interventions, i.e. placebo, tofacitinib (1mg BID, 5mg BID, 10mg BID, 15mg BID), filgotinib 200 OD and upadacitinib (3 mg BID, 6 mg BID, 12 mg BID, 24 mg BID and 24mg OD). Two upadacitinib doses (6 mg BID and 24 mg BID) and filgotinib 200 OD, performed best as judged by the relevant forest plots, league matrixes, rankograms, SUCRA values (96.7%, 84,6 %and 78,7%, respectively) and the clustered ranking plots for efficacy and tolerability.

Conclusions: Upadacitinib 6 mg BID, upadacitinib 24 mg BID and filgotinib 200 OD performed better as induction therapies in comparison to control therapies. Consequently, these regimens may play a therapeutic role in CD and therefore they merit further evaluation with well-designed RCTs.


Crohn’s disease, JAK inhibitors, efficacy, tolerability, network meta-analysis