The first ICIs for metastatic melanoma were first approved by the FDA in 2011, and since then ICIs have become the standard medical treatment for various cancers. However, several irAEs have been reported. (1) These adverse events occur when the suppression of the immune system is impaired by ICIs, presenting symptoms similar to autoimmune diseases, and appearing in all organs, such as the skin, respiratory system, nervous system, endocrine system, and digestive system. Although ICIs-induced diarrhea and colitis occur in approximately 20% of the patients on PD-1 inhibitors, the frequency of severe toxicities is low with only 2–5% of patients. The frequency of irAE colitis differs between CTLA-4 inhibitors and PD-1 / PD-L1 inhibitors. CTLA-4 inhibitors cause more frequent gastrointestinal symptoms than PD-1/PD-L1 inhibitors; ipilimumab causes symptoms in approximately 40% of cases, and severe inflammation occurs in 10–15% of cases. (2–4) PD-1 inhibitors were used in this case, and it is a rare case in terms of the frequency of severe disease. The symptoms of irAE colitis are similar to those of ulcerative colitis; moreover, there is a substantial overlap between irAE colitis and inflammatory bowel disease, both endoscopically and histologically. Therefore, the treatment of irAE colitis should be consistent with the treatment of ulcerative colitis.
The guidelines for irAEs have been published by the American Society of Clinical Oncology (ASCO), the European Society of Medical Oncology (ESMO) and others, which recommend treatment strategies based on a certain level of consensus. (2,3) These guidelines recommend that the patients with grade 2 or higher irAE colitis should discontinue treatment with ICIs and receive high-dose systemic corticosteroids, and steroid-resistant patients should be treated with infliximab. Recently, vedolizumab has been reported to be effective in cases that are refractory to steroids and infliximab. Mesalazine has also been reported to be effective for maintenance therapy. (5) Although the surgical treatment of refractory irAE colitis is not specified in the guidelines, it was managed equally with ulcerative colitis. For irAE colitis with perforation, a subtotal colectomy with ileostomy and sigmoidostomy is recommended because colonic lesions are generally extensive and segmental colonic resection is generally followed by a severe inflammation of the remaining colon in the postoperative phase. However, there have been some reports of deaths within a few days after subtotal colectomy. (6) Moreover, subtotal colectomy is considered invasive and risky for most patients with severe irAE colitis. In addition, there are case reports of ileostomy after colonic perforation in irAE colitis, and the prognosis was poor outcome. (7) In contrast, some studies have reported that ileostomy is effective in the management of refractory irAE. Moreover, in one case, a diverting loop ileostomy was performed to treat Fournier gangrene accompanied by refractory irAE colitis, and the patient reported that the colitis was also found to improved. (8,9)
In this case, prednisolone, infliximab, and mesalazine did not significantly improve the symptoms significantly, and leukocytapheresis was performed. Medical treatment improved the symptoms for a period of time, and the irAE colitis recurred and was extremely refractory to medical treatment; therefore, surgical treatment is considered; however, the performance status was 3. A subtotal colectomy was invasive and risky; therefore, we had to choose a less invasive surgery. Based on these reasons, an ileostomy was performed to rest the colon, and abdominal symptoms were resolved in the early postoperative period, and could the ICI treatment be resumed. However, a postoperative colonoscopy showed no improvement in the mucosal lesions, and stoma closure could not be performed.
The difference between irAE colitis and ulcerative colitis is that irAE colitis is a side effect of cancer treatment by ICIs. There is a conflicting view that the discontinuation of ICIs is preferable for severe irAE colitis; however, the continuation of ICIs is preferable for oncology. In other words, it is necessary to consider ways to avoid the discontinuation of ICIs in the treatment of irAE colitis, including surgical treatment. There is limited evidence on the safety of resuming treatment with ICIs in patients whose treatment was interrupted due to the onset of irAEs. In a report of 24079 irAE cases registered in the World Health Organization VigiBase database, 452 patients rechallenged ICIs and 130 (28.8%) patients showed the recurrence of irAEs, and colitis, pneumonia, and hepatitis were shown to have high recurrence rates. (10) In another report of 167 patients with irAE colitis, one-third of the patients who rechallenged ICIs relapsed with irAE colitis. (11) It has been reported that the risk-reward ratio of anti-PD-1 or the anti-PD-L1 rechallenge is within acceptable limits, and some have reported that the risk-reward ratio of anti-PD-1 or anti-PD-L1 rechallenge is within acceptable limits. (12) These reports suggest that a rechallenge therapy with ICIs is acceptable under close monitoring, although further studies are needed. Currently, the standard surgical procedure for treating ulcerative colitis is proctocolectomy. Appendicostomy, cecostomy, and ileostomy have been performed for ulcerative colitis since the early 1900s to restore the colon. Moreover, ileostomy was commonly performed for toxic colitis in patients with IBD. Subsequently, proctocolectomy was also performed; however, the mortality rate was high due to intraoperative fecal spillage. Turnbull et al. pointed out that fecal spillage occurs during colonic manipulation and proposed diverting ileostomy and decompression colostomy in 1971. (13) The colostomy, also known as Turnbull-Blowhole colostomy, decompresses the colon; however, it was reported to have the unexpected benefit of temporary remission of the disease in most patients. With improvements in the management, the Turnbull-Blowhole colostomy has rarely been performed, but has recently begun to be reevaluated. The Turnbull-Blowhole colostomy has been reported to be effective for toxic ulcerative colitis in pregnancy. (14,15) Since decompression of the colon was not necessary in this case, a Turnbull-Blowhole colostomy was not necessary; nevertheless, it may be an option for patients with irAE colitis complicated by toxic megacolon, who are in poor general condition.
We encountered a case in which ileostomy was performed for irAE colitis refractory to medical treatment, and the symptoms were alleviated. The indications for the treatment with ICIs are expected to expand and the number of such cases will increase in the near-future. The surgical treatment of irAE colitis also needs to be evaluated and analyzed.