Abstract Background Crohn’s disease (CD) and ulcerative colitis (UC) are chronic inflammatory bowel diseases (IBDs) with distinct clinical consequences. Distinct from patients with UC is that nearly 25% of patients with CD patients have or will develop perianal disease. Predictors for the development of new perianal disease or complications from existing perianal disease are not well described but such occurrences are included as adverse events (AEs) in clinical trials for IBD. We performed a systematic review and meta-analysis of perianal AEs in pivotal trials of therapies for IBD. Methods We conducted a search of PubMed, EMBASE, MEDLINE, Medline Plus, and CDSR library databases for registry and pivotal Phase II-III clinical trials of therapies for IBD. Safety data from clinicaltrials.gov was reviewed to identify incidence of perianal AEs using the terms "abscess" and "fistula." Reported perianal AEs included the following: anal abscess, fissure, fistula; perianal abscess, fistula; perineal abscess; perirectal and rectal abscess; subcutaneous (gluteal) abscess. Results Perianal AE incidence across all trials reviewed was 0.47% (91/19,327 patients). Perianal AEs were more likely to occur in patients with CD (0.77%, 70/9,046) than UC (0.20%, 21/10,281). Perianal AEs occurred more frequently in populations of drug-exposed patients than in patients exposed to placebo. In drug-exposed patients with UC, 0.22% (16/7,415) of patients experienced a perianal AE vs 0.17% (5/2,935) of patients exposed to placebo. In drug-exposed patients with CD, 0.76% (55/7,271) of patients experienced a perianal AE vs 0.61% (15/2,465) of patients exposed to placebo. The results are summarized in Table 1. Conclusion We found that perianal AEs occur in pivotal UC and CD clinical trials. Expectedly, the frequency of perianal AEs was higher in trials of CD than those for UC. However, the development of perianal AEs in trials of UC suggests that these patients may have been misclassified and have CD. Available data do not allow distinction between development of a perianal AE in a patient with known perianal disease or progression of undiagnosed clinical or subclinical perianal disease, nor does it allow for distinction in a patient not responding to the experimental agent. The finding that perianal AEs occurred more frequently with drug exposure than placebo may be due to fistula healing with consequential abscess formation or unique mechanistic effects that worsen perianal disease. These results highlight a probable rate of misclassification of patients with UC in pivotal trials, and suggest the possibility of subclinical perianal disease in some patients. Further clarification and study of these AEs and their implications are warranted.
Read full abstract