Therapeutic Decisions in Inflammatory Bowel Disease in the SARS-Cov-2 Pandemic

Dear Editors:

We were very interested to read the recent report of the first formal analysis of the Surveillance Epidemiology of Coronavirus Under Research Exclusion (SECURE) registry of the worldwide experience of inflammatory bowel disease (IBD) outcomes during the severe acute respiratory syndrome coronavirus 2 (SARS-Cov-2) pandemic.¹ We feel that a number of important observations worthy of discussion arise from this manuscript, each with potential implications for patient management.

The safety of anti–tumor necrosis factor (anti-TNF) therapy during the pandemic, along with other conventional therapies, is of great interest to all IBD physicians. Current international guidance has taken an appropriately cautious stance regarding the potential risks of immunosuppression and its implications for patient behavior/social distancing.^{2 , 3} We note the large number of SECURE-IBD registry patients who have developed Coronavirus Disease 2019 (COVID-19) while taking an anti-TNF (43.4%), most (three-fourths) on monotherapy, and the absence in this group of a signal of harm. The authors demonstrate in exploratory analyses a 5-fold increase in severe complications for those on combination therapy.

Without further data from this registry, or from a case-control study, we would caution against drawing firm conclusions regarding safety. We suggest it is still possible that anti-TNF may yet be shown to increase susceptibility to COVID-19. We acknowledge that SECURE-IBD is not designed to determine this question, but it is interesting that there is a relative paucity of other biologic use represented in the registry. The impact of anti-TNF monotherapy on those with infection appears reassuring, and there is the potential for benefit perhaps dependent on the phase of the viral illness.⁴ As clinicians, we are inevitably concerned by the data presented regarding combination therapy that appears to be associated with severe complications. Interestingly the proportion of patients in SECURE-IBD on combination therapy is substantially lower (23%) than that described during anti-TNF induction in the UK PANTS (Personalised anti-TNF therapy in Crohn’s disease study) (adalimumab 53%, infliximab 62%),⁵ and lower than in our ongoing local practice (760 patients on anti-TNF; combination therapy on adalimumab 31%, on infliximab 65%). The SECURE-IBD data suggest that treatment-naïve patients requiring anti-TNF therapy during the pandemic may preferentially be started on the least immunogenic agent (in our practice, adalimumab over infliximab), and this emerges as an important focal point for further research.

The data presented in the manuscript implicating sulfasalazine/5-aminosalicylate (5-ASA) therapy with poor outcomes are worrying, counterintuitive, and surprising based on the lack of immune-modulatory activity. Many possible explanations require exploration. We propose that the observed association with 5-ASA may partly be a surrogate marker for underlying IBD activity, and note that the record of disease activity in the registry, for pragmatic reasons, is based on a physician global assessment. We observe the elevated (12%) intensive care unit/ventilator/death rate in those patients with reported moderate to severely active IBD, and that other studies have also reported a detrimental association between all IBD activity and COVID-19 outcomes.⁶ Similarly, it is possible that the more serious COVID-19 outcomes in patients on steroids may at least be partly confounded by an association with active IBD. Recent data have implicated active colitis as associated with increased colonic expression of the angiotensin-converting enzyme 2 epithelial cellular receptor for SARS-Cov-2,⁷ providing a possible mechanistic link. Of note neither 5-ASA nor steroids are reported to independently alter intestinal mucosal angiotensin-converting enzyme 2 expression.⁸

Overall, we congratulate the authors for their timely establishment of a critically important resource during the pandemic, and for addressing important issues. We point to the need for further data to substantiate these initial observations, and for detailed case-control and cohort-based studies to address the key issues in patient management. As data accrue, evidence-based alterations to current clinical guidelines will be of additional benefit to our patients, not only in relation to therapy decisions, but also to social distancing guidelines and their ability to safely perform their roles within society. At present these, and other, published data suggest that in the COVID-19 era, we should closely monitor for objective IBD activity so as not to delay effective management, using where possible biologic monotherapy and avoiding systemic steroids. Importantly, we feel these data should not negatively affect the appropriate use of 5-ASA in management algorithms.