Background
Immune checkpoint inhibitors (immunotherapy, ICI) can often cause adverse events, including immune-mediated colitis (IMC). Evidence that the gut microbiota may play a role in the pathogenesis and resolution of IMC is increasing. However, the long-term benefit of faecal microbiota transplant (FMT) remains to be investigated, especially in the context of ICI resumption.
purpose
We report 2 unique case patients who responded to FMT for their IMC per study protocol (Case Patient 1 in NCT03819296; Case Patient 2 in NCT04038619 did not receive steroids or biologics prior to FMT); later successfully resumed their ICI treatment; and achieved cancer remission during the study period. FMT was performed via colonoscopy using 50 g/250 ml of stool from a single healthy donor for 1 dose. Stool material was provided from the Stool Bank of the University of Texas School of Public Health.
Patient Case 1
A 68-year-old woman with stage IV non-small cell lung cancer presented with Common Terminology criteria for adverse event grade 3 diarrhea after receiving atezolizumab for 1 year (Figure 1). Colonoscopy with biopsy showed normal mucosa consistent with chronic active colitis. Her diarrhea initially improved with a steroid taper and she was maintained on mesalamine and cholestyramine. ICI was relaunched. Diarrhea recurred 9 months later and evaluation showed erythema, friability, and exudate at endoscopy with histology demonstrating severe chronic active colitis and ulceration (Figure 2, A). ICI was withheld and IMC was managed with a steroid taper and concomitant vedolizumab, as she developed side effects of sinus congestion, leg swelling, and debilitating joint pain after a single dose. Similar symptoms recurred after switching to ustekinumab, which was discontinued. Cancer surveillance and assessment demonstrated disease progression. Given the critical need to restart ICI therapy, FMT was performed. The patient remains in clinical remission from IMC despite restarting ICI for 11 months without the need for additional diarrhea medication. A restaging scan showed remission of the cancer, and a repeat colonoscopy was normal. Faecal calprotectin decreased from an initial 227 mcg/g to 68 mcg/g at last follow-up.


Patient Case 2
A 54-year-old woman with metastatic renal cell carcinoma developed Common Terminology criteria for the adverse event grade 3 diarrhea after 5 cycles of pembrolizumab and 1 cycle of nivolumab therapy (Figure 3). Infectious work-up was negative. Faecal calprotectin was significantly increased at elevations greater than 1000 mcg/g. Colonoscopy demonstrated diffuse ulcerative inflammation (Figure 2, B). Biopsy showed chronic active colitis. FMT was performed according to the study protocol as first-line therapy for IMC. Follow-up stools improved dramatically to 78.6 mcg/g since day 1 of FMT therapy with calprotectin, the patient remaining in clinical remission. Repeat endoscopy showing ulcer healing within 2 months. He successfully resumed nivolumab and subsequently became cancer free after elective primary cancer resection at a follow-up of 7 months.

discussion
IMC may interfere with cancer care in patients with advanced malignancy (1). Although its management is described in oncology guidelines (2), effective strategies for keeping IMC in remission while resuming ICI have not been well studied. Ustekinumab (3) and FMT (4) are now increasingly employed as standard therapy in refractory IMC. ICI-induced microbiome dysbiosis has previously been reported as one of the potential mechanisms of IMC (5, 6). Dubin and colleagues (7) found that the prevalence of bacteria increased Phylum Bacteridetes IMC was associated with inhibition of development, which was attributed to the stimulation of regulatory T-cell differentiation by these bacteria. They also found that decreased microbial diversity was associated with increased risk of colitis (7). The first case series with successful FMT treatment of refractory IMC showed high amounts of bacterial taxa in the stools of recipients believed to be protective against IMC, including Bacteroides (4). Our case report provided further encouraging evidence that FMT has a lasting steroid-sparing therapeutic effect for IMC, which enabled the patient to recover the benefit of restarting ICI to achieve better cancer outcomes. Further microbiome analysis will provide more knowledge about the role of FMT in the treatment and maintenance of IMC.
reference
- 1.
Abu-Sebih H ,Ali FS ,Nakash AR ,etc . Resumption of immune checkpoint inhibitor therapy after immune-mediated colitis. J Clin Oncol. 2019;37:2738–45. [PMID: 31163011] doi:10.1200/JCO.19.00320 CrossrefMedlineGoogle Scholar - 2.
Brahmar JR ,Laccheti c ,Snyder BJ ,etc . Management of immune-related adverse events in patients treated with immune checkpoint inhibitor therapy: American Society of Clinical Oncology clinical practice guidelines.. J Clin Oncol. 2018;36:1714–68. [PMID: 29442540] doi:10.1200/JCO.2017.77.6385 CrossrefMedlineGoogle Scholar - 3.
Thomas A.S ,Mother W ,Wang Y . Ustekinumab for refractory colitis associated with immune checkpoint inhibitors. N Engl J Med. 2021;384:581. [PMID: 33567198] doi:10.1056/NEJMc2031717 CrossrefMedlineGoogle Scholar - 4.
Wang Y ,Wisnowski DH ,Helmink BA ,etc . Faecal microbiota replacement for refractory immune checkpoint inhibitor-associated colitis. Nat Med. 2018;24:1804–8. [PMID: 30420754] doi:10.1038/s41591-018-0238-9 CrossrefMedlineGoogle Scholar - 5.
Peugeot RC ,Wong M ,Martin A . Influence of gut microbiota on immune checkpoint inhibitor-associated toxicity. There Adv Gastroenterol. 2019;12:1756284819870911. [PMID: 31555343] doi:10.1177/1756284819870911 CrossrefMedlineGoogle Scholar - 6.
Anderson R ,Theron AJ ,Report BL . Immunopathogenesis of immune checkpoint inhibitor-associated adverse events: role of the gut microbiome and Th17 cells. Frontal immunol. 2019;10:2254. [PMID: 31616428] doi:10.3389/fimmu.2019.02254 CrossrefMedlineGoogle Scholar - 7.
Who is Dubin? ,Callahan MK ,Ren B ,etc . Gut microbiome analysis identifies melanoma patients at risk of checkpoint-blockade-induced colitis. reduce the nat. 2016;7:10391. [PMID: 26837003] doi:10.1038/ncomms10391 CrossrefMedlineGoogle Scholar