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


Within our cohort, the absolute rates of GIRAE were low. The occurrence of confirmed IBD with secukinumab appear to be related, but not exclusive, to patients with a diagnosis of AS. Symptom exacerbation is more likely in patients with a prior diagnosis of IBD and tends to occur within the first year of treatment initiation. There is, however, insufficient evidence to infer causal links between IBD and secukinumab.

The reported rate of all IBD events from pooled secukinumab trial data has previously reported overall rates of around 0.2% (95% Cl: 0.1, 0.4);[12] however, our cohort rates were 1.3% amongst de novo cases and pre-existing IBD flares. It is possible that in the real-world settings, the absolute risk for IBD is higher than that seen in clinical trials. The nature of undertaking trials, with associated inclusion and exclusion criteria, inevitably creates an element of selection bias. In our cohort, a priori IBD was not an exclusion criterion if clinically inactive at the time of enrolment. Unsurprisingly, it has been shown to be highly predictive of GIRAE and associated flares. IBD prevalence has been shown within our cohort to be <1% of PsA and 5% of AS patients, compared with background population rates of >3% and 7%, respectively.[13,17,18] This raises the possibility that clinicians within our sites are avoiding initiating secukinumab in selective at-risk patients. Pooled trial IBD rate characteristics are comparable to those in our cohort and does not seem to account for the differences in observed GIRAE (AS 3.1% and PsA 0.6%).[12]

In our four definite IBD cases, the two new cases occurred after a switch from a TNFi. Animal data on IL-17 biology supports a plausible mechanism by which secukinumab could negatively impact gut homeostasis. In addition to considering definite IBD events, we analysed GIRAE suggestive of probable and possible IBD events. These captured new-onset diarrhoea, which prompted treatment interruption and some degree of investigation. There is no published meta-analysis describing rates of non-serious GIRAE in secukinumab patients. However, in the breakdown of non-serious adverse events available for the MEASURE and FUTURE clinical trial programmes in AS and PsA, there are numerically more patients reporting diarrhoea on the active arm compared with placebo.[9,10] Diarrhoea is listed on the drug summary product characteristics as a recognized common side effect, affecting between 1% and 10% of people. Many medications can alter bowel habit and yet have no association with a risk of IBD. The challenge with secukinumab is that it appears to be associated with both diarrhoea and new-onset IBD or exacerbation of established IBD. The faecal calprotectin tests performed within our cohort did not seem to identify patients who were most likely to need an endoscopy. Of the four definite IBD cases, only one pre-existing IBD case returned positive. This creates a challenge in deciding which patients to investigate for IBD-related symptoms after secukinumab initiation and to what extent.

Based upon the combination of our real-world experience (in which five of seven patients with prior IBD had bowel disease symptoms/flares) and published literature, it seems prudent for clinicians to consider alternative immune modulatory strategies for AS and PsA in patients with an established IBD diagnosis. No causal link has been demonstrated previously relating to secukinumab dose regimens of 150 mg or 300 mg and our results shed no light on this. Observations in favour of causality include the temporal relationship between exposure and IBD events, and the improvement in symptoms on treatment withdrawal. This association was further highlighted by a patient who developed a flare after initiating secukinumab, but in whom symptoms resolved on withdrawal and recurred after a rechallenge.

In considering the limitations of this study, it is important to acknowledge the relatively small sample size. In addition, there is potential that bias may be present due to incomplete follow-up. Some GIRAE did not result in full investigation, precluding definite diagnoses of an IBD event. It is also possible that right censorship is present. Rates of IBD may not be reflected in the length of data collection, hypothesizing that probable and possible cases may eventually lead to IBD on timeframes >3 years.

The results presented have the advantage of reflecting real-world experience from multiple large hospitals, reflective of clinician prescribing in clinical practice. Event outcome data were rigorously collected, and there is high confidence in the diagnostic accuracy in terms of rheumatological diagnoses, treatment exposure and adverse event coding. Incorporated into the experience is a type of natural experiment model in the context of challenge then re-challenge with a drug, which adds weight to there being a biological link between IL-17 inhibition and GIRAE.