Fecal microbiota transplantation with ruxolitinib as a treatment modality for steroid‐refractory/dependent acute, gastrointestinal graft‐versus‐host disease: A case series

Abstract

Acute graft-versus-host disease (aGvHD), together with infections, constitute the two main early complications of allogeneic hematopoietic cell transplantation (alloHCT).1 The mainstay of first-line therapy in aGvHD is systemic administration of high-dose glucocorticoids, but only 40%–60% of patients respond to this treatment depending on the grade of severity of the disease.1 At present, there is no established standard-of-care second-line therapy. The high mortality rate of steroid-refractory/dependent (sr/d) aGvHD, especially in patients with grade III-IV lower gastrointestinal (GI) tract involvement,2 is a major drive for exploring novel therapeutic strategies. One of them, ruxolitinib, a selective Janus kinase (JAK1 and JAK2) inhibitor, is associated with significant general improvement, and improved remission rates in patients with sr/d aGvHD.3 A second, emerging therapy is fecal microbiota transplantation (FMT) (being the most effective method of gut microbiota replacement), which we recently assessed in a prospective, multicenter study regarding antibiotic-resistant bacteria decolonization in which sr/d aGvHD patients reached 57% of overall response rate.4 Taken together, considering the possibly complementary mechanisms of ruxolitinib and FMT, we progressed on from positive initial clinical observations to offer combination therapy of ruxolitinib and FMT to our patients with sr/d GI-aGvHD. To our knowledge, this is the first report on such combination therapy. CASE 1: A 66-year-old male, a recipient of alloHCT from a matched unrelated donor (MUD; see Table 1 for details), attended our clinic because of fatigue, dehydration, cachexia, and exacerbation of severe gastrointestinal symptoms. Despite thorough investigation, we could not identify other reasons for the patient's symptoms (i.e. > 2 L/day of loose stools with mucus and blood occasionally, with ileus and peritoneal signs). On the day +45 after alloHCT, the patient was diagnosed with possible (colonoscopy with biopsy) GI aGvHD stage IV MAGIC, IV Glucksberg, IBMTR D,5 and treatment was initiated with high-dose methylprednisolone (2 mg/kg). No clinical response was seen and mycophenolate mofetil was introduced with steroid tapering, but the patient's condition progressively worsened. We decided to treat the patient with ruxolitinib and FMT, but pancytopenia was the reason for postponing ruxolitinib. Because of GI tract colonization with antibiotic-resistant bacteria (ARB; Pseudomonas aeruginosa MBL), our patient qualified for intervention with an FMT procedure (as a part of a prospective clinical study assessing ARB decolonization rate by the use of FMT [NCT02461199]). The first FMT dose was associated with improvement of the patient's condition and diarrhea alleviation (Figure S1). The patient subsequently received a second and third FMT dose. We saw marked clinical improvement, including complete resolution of diarrhea and abdominal pain. Unfortunately, due to the patient noncompliance, diarrhea, ileus, and peritoneal signs recurred. As the last, fourth treatment line, we decided to add ruxolitinib. Following this, the patient's status stabilized, and he obtained clinical partial remission, although he still had diarrhea up to five stools a day (formally active disease, according to EBMT criteria5). He was lost to follow-up and died on day +264. Anti-Thymocyte Globulin Metothrexate, cyclosporine A, budesonide, methylprednisolone, mycophenolate mofetil, tacrolimus Klebsiella pneumoniae MBL, K. pneumoniae ESBL, Pseudomonas aeruginosa MBL, Enterobacter asburiae ESBL, Candida glabrata, C. crusei E. coli ESBL K. pneumoniae ESBL, C. difficile Metothrexate, cyclosporine A, budesonide, Methylprednisolone, tacrolimus CASE 2: A 52-year-old female, post-alloHCT from matched related donor (MRD), began to have diarrhea and extensive skin rash on day +17. She was finally diagnosed as overall stage IV MAGIC, Glucksberg IV, IBMTR D5 aGvHD (skin IV biopsy proven and gut IV biopsy proven). She was given methylprednisolone (2 mg/kg), budesonide, parenteral nutrition, and—due to steroid resistance—mycophenolate mofetil; temporary improvement occurred. Because of the deterioration of diarrhea, ruxolitinib was added (initially 2 × 5 mg, then 2 × 7.5 mg per day). Because of ARB colonization, the patient received FMT twice. After FMT, gastrointestinal symptoms resolved. The obtained response was consolidated with another FMT and intravenous methotrexate (15 mg). Ruxolitinib was continued. After all treatments were administered, the patient's diarrhea resolved, and the patient entered complete remission of gut aGvHD. Furthermore, her overall performance status significantly improved, and she was discharged. During follow-up, there was no aGvHD exacerbation and no chronic GvHD occurred, but the patient ultimately died due to relapse on day +398 after alloHCT. CASE 3: A 55-year-old female, post-alloHCT from MRD, started to have diarrhea with increasing intensity from day +11. Biopsies taken during colonoscopy revealed features of severe aGvHD. We diagnosed aGvHD in overall stage IV according to MAGIC criteria, Glucksberg IV and D according to IBMTR5 (skin II probable and gut IV proven). High-dose methylprednisolone (2 mg/kg) was introduced to the patient's treatment, after which all symptoms from the gastrointestinal tract alleviated. During the weaning of the steroid dose, the diarrhea gained in its intensity again (> 2 L of stool daily); additionally, the patient started to have severe, spastic abdominal pain with peritoneal signs. Again, high-dose steroids were administered, but after 10 days of therapy, there was no response (NR). Given this, we introduced second-line treatment with ruxolitinib; this was associated with a subsequent slow and gradual improvement in diarrhea severity, but without any progress in alleviation of abdominal pain. Even though the patient was treated for more than 14 days with ruxolitinib, we did not see any further improvement in the patient's symptoms. Suboptimal clinical improvement to the administered treatment (steroids and ruxolitinib) prompted us to combine ruxolitinib therapy with FMT, as we did in previous cases. After obtaining the Institutional Review Board (IRB) acceptance [KB/203/2020] for three FMTs every week, the first and second administration of FMT was performed. Following this, we observed rapid and significant improvements. A third FMT administration was performed, with further normalization of the patient's condition and stabilization of the frequency of patient's defecation to once a day. Finally, the CR of aGVHD was diagnosed. She is now in +351 day after alloHCT with aGvHD CR, without significant morbidity or further complications. CASE 4: A 22-year-old female, post-alloHCT from MUD, started to have diarrhea on day +31, after which we suspected gut GvHD. Colonoscopy with histopathology showed GI-aGvHD and MAGIC grade IV overall, IV Glucksberg, and IBMTR D5 was diagnosed. Therapy with methylprednisolone (2 mg/kg) and budesonide was initiated. After 10 days of steroids, the patient was assessed as having partial clinical response, but formally NR according to MAGIC and active disease according to EBMT.5 In light of the unsatisfactory result of therapy with high-dose steroids, ruxolitinib treatment (2 × 5 mg) was started, and IRB application for FMT was made (with the same protocol as above [KB/202/2020]). Finally, the first FMT was performed and from the next day, the volume of stool gradually decreased. The abdominal pain almost completely resolved. Then, second and third FMTs were performed. We observed general improvement of the patient's condition. Despite this, she continued to have up to five bowel movements a day, less profuse, more concentrated but still diarrheal stool with episodic abdominal pain. Therefore, a further three doses of FMT were planned, and after additional IRB approval [KB/202/2020], she obtained the fourth, fifth, and sixth FMT; after this, she had one stool a day, formed, without any abdominal pain. CR was assessed, and she is now in perfect condition, on day +326 without any morbidity, cGvHD and aGvHD recurrence, with ruxolitinib stopped. The main comment about the first case is that FMT and ruxolitinib were administered very late during the post-alloHCT course—our patient did not report diarrhea in post-transplant care, although it did occur. We independently observed early and satisfying improvement, but unfortunately, it lasted for only a short period, and ended rapidly because of our patient's noncompliance. Similar to the situation in case 1, case 2's performance status was advanced and severely complicated. We decided to perform FMT first, before ruxolitinib administration, with the intention of ARB decolonization and simultaneously, to initiate immunomodulatory effects by the gut microbiota, without possible cytopenias as a treatment complication. Our patient's general condition significantly improved and ruxolitinib could be safely added. She reached durable CR of gut aGvHD. Encouraged by previous observations, case 3 was treated with FMT and ruxolitinib intentionally early after the diagnosis of steroid resistance. We initially planned FMTs sessions (3×) in combination with ruxolitinib. This patient showed a particularly marked improvement, and reached complete remission of aGvHD just after FMT introduction, on ruxolitinib taken for several weeks. A possible explanation for such an outstanding response is the relatively short previous time-span from the diagnosis of steroid dependency, with treatment prior to the development of severe gut wall damage facilitating microbiota engraftment. The same strategy as in case 3 was applied in case 4, when ruxolitinib and FMT was administrated even earlier than in case 3. Undoubtedly, after ruxolitinib administration and the first three FMTs, the patient's condition improved. Eventually, despite visible improvements, complete remission was not reached. In order to deepen the response, we decided to continue our treatment with the next three FMTs combined with a higher dose of ruxolitinib that resulted in CR. After this initial experience, we see great potential in this strategy, and suggest that randomized controlled trials are needed to find the answer, and explore how deeply FMT enhances outcomes of patients similar to those in the trial published by Zeiser et al.3 We see a real chance that the additive effect of FMT will show significantly better results. Furthermore, modifications of protocols for combined treatment with FMT and ruxolitinib merit consideration. First clinical trial, besides of this report, was registered (NCT04269850).6 BHM is the recipient of an NIHR Academic Clinical Lectureship (CL-2019-21-002). The Division of Digestive Diseases at Imperial College London receive financial and infrastructure support from the NIHR Imperial Biomedical Research Centre (BRC) based at Imperial College Healthcare NHS Trust and Imperial College London. JB and GWB are the founders of the fecal microbiota bank Human Biome Institute. BHM has received consultancy fees from Finch Therapeutics Group, MA, USA. Other authors declare no conflicts of interest. JB, GWB, and AT planned the treatment modality; JB, MJ, PK, LCh, KL, EKP gathered the data; JB, MJ, and AT verified the underlying data; JB, MJ, and BHM wrote the publication. Figure S1 Timeline of the procedures performed in patients with steroid-refractory/dependent, acute, gastrointestinal graft-versus-host disease. Created with BioRender.com Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts