Tumor necrosis factor (TNF) antagonists are effective for induction and maintenance of perianal fistula response and remission in patients with Crohn’s disease (CD), a systematic review and meta-analysis shows in the December issue of Clinical Gastoenterology and Hepatology. However, there are few data on the effects on internal fistulae, and further studies of ustekinumab, vedolizumab, and stem cell therapies are needed for patients with fistulizing CD.
Fistulas (abnormal, tunnel-like connections linking the bowel to neighboring epithelial surfaces) have been reported to develop in 17%–50% of patients with CD. They produce persistent fecal or urinary seepage, pain, and infection, impairing quality of life and productivity. It is not clear how the fistulas form, but mucosal and transmural inflammation, and possibly luminal bacteria, are believed to contribute to their development. Risk factors include disease location and duration, rectal or colonic involvement, and young age at diagnosis.
Management of fistulas in patients with CD involves controlling infection and sepsis and inducing mucosal and fistula tract healing. Many patients either do not respond to medical therapy or fistulas recur. Adjunct surgical techniques have been developed to close fistula tracts, but approximately 20% of patients with perianal fistula ultimately require proctectomy, and those with internal fistulas often require surgical resection due to penetrating disease.
TNF antagonists are effective in treatment of fistulizing CD, and biologics with alternative mechanisms have been developed, such as inhibitors of IL12 and IL23 signaling or antibodies against integrins, but their efficacy in management of fistulizing CD is uncertain. Matthew J. Lee et al therefore performed a systematic review and meta-analysis of 27 randomized controlled trials, comprising 2019 patients with CD and fistulas, that evaluated the efficacy of medical therapies for fistulizing CD.
In their meta-analysis, Lee et al found that TNF antagonists were the only drugs proven to induce and maintain fistula response (6 trials) and remission (6 trials). These drugs were associated with an approximately 1.5-fold increase in odds of induction of fistula response, and a 2-fold increase in ods of induction of fistula remission, maintenance of fistula response, and maintenance of fistula remission.
A total of 43% (53 of 124) of patients receiving a TNF antagonist maintained response compared with 22% (28 of 129) of patients receiving placebo (see figure). The between-group difference in effect was statistically significant (relative risk, 1.97).
The largest effect estimates for induction and maintenance of fistula response and fistula remission were observed for infliximab. Direct comparisons among the TNF antagonists were not possible.
Thiopurines were not superior to placebo for induction of fistula response or remission, although oral tacrolimus might be effective in inducing a response. Unfortunately, the side effect profile associated with this agent has limited its use.
A post-hoc analysis of ustekinumab (inhibitor of IL12 and IL23 signaling) revealed a statistically significant (1.5-fold) increase in the likelihood of inducing fistula response. Although these results are preliminary, they warrant further investigation.
A pooled analysis of bone marrow-derived mesenchymal stromal cells and adipose-derived stem cells calculated a 30% increase in the likelihood of achieving fistula remission over placebo. However, high rates of placebo-induced remission were observed in the trial of adipose-derived stem cells, through surgical curettage and injection of saline in fistula tracts, indicating the importance of good adjunctive surgical techniques in the management of perianal fistulizing disease.
The combination of TNF inhibitors and antibiotics produced a statistically significant higher rate of induction of fistula response and remission, compared with a TNF antagonist alone.
Lee et al propose that future studies focus on the efficacy of combination therapy with immunosuppressives and the relationship between TNF antagonist trough levels and response status to optimize treatment for fistulizing CD.