Secukinumab as a Potential Trigger of Inflammatory Bowel Disease in Ankylosing Spondylitis or Psoriatic Arthritis Patients

Ioana A. Onac; Benjamin D. Clarke; Cristina Tacu; Mark Lloyd; Vijay Hajela; Thomas Batty; Jamie Thoroughgood; Sandra Smith; Hannah Irvine; Diane Hill; Grace Baxter; Natalie Horwood; Suma Mahendrakar; Rizwan Rajak; Sian Griffith; Patrick D. W. Kiely; James Galloway


Rheumatology. 2021;60(11):5233-5238. 

In This Article


Patient Population

We undertook a retrospective cohort study. Ten centres from the south east of England participated. All paper or electronic records for patients that commenced secukinumab at each centre between 2016 and 2019 were reviewed. An anonymous data collection form was used to collate patient information. Questions sought to answer whether IBD had occurred prior to or after secukinumab initiation. Screening was defined as clear documentation of a prior diagnosis of IBD, gastrointestinal symptoms suggestive of IBD and/or faecal calprotectin was performed. All gastrointestinal adverse events were included for analysis (see Supplementary Data S1, available at Rheumatology online).

Assessment Measures

Baseline data collected included patient sex, age, diagnosis, length of secukinumab treatment, prior IBD diagnosis, existing IBD symptoms, gastrointestinal-related adverse events (GIRAE) suggestive of IBD that developed during treatment, action taken if GIRAE developed (faecal calprotectin/gastroenterology referral/other investigation including biopsy) and outcome if GIRAE was confirmed, including treatment decision or secukinumab discontinuation.

Definition of Outcome

The degree of GIRAE suggestive of IBD were defined amongst the cohort, categorizing by:

  1. Definite [objective evidence of IBD (biopsy proven), clear temporal association, resolution of symptoms on drug withdrawal, no alternative explanation felt more likely]

  2. Probable (as per definite, but without biopsy confirmation)

  3. Possible (gastrointestinal symptoms not fulfilling definite or probable criteria).

Statistical Analysis

A Cox proportional hazards survival model was used to compare event rates over time between diagnostic groups, with adjustment for age and sex. A sensitivity analysis evaluated the association with a prior IBD diagnosis and event rates. The results were reviewed graphically using Kaplan–Meier plots to confirm model assumptions. Statistical analyses were conducted in Stata 16 (Statacorp LLC, Texas, TX, USA).

The project was conducted as part of a service evaluation and in line with trust policies did not required ethical approval. All data were fully anonymized.