Defined Fecal Microbiota Transplantation for Clostridium Difficile Diarrhea

The study is on indefinite HOLD due to the loss of funding that occurred during the pandemic emergency. Subsequently, a key collaborator left our institution, and as a near-term result, the protocol awaits reactivation. Three patient subjects were enrolled, all 3 patients/subjects were cured of the infection, and there were no adverse events or sequelae observed or reported. The aim of the study continues to confirm and extend the work of Trede and Rask-Madsen (Lancet 1989;1:1156-1160) that administration of a defined fecal microbiota will lead to rapid and sustained resolution of C. difficile-associated chronic relapsing diarrhea. FDA required 4 non-geriatric qualified patients to be studied before including the elderly. However, C. difficile-associated chronic relapsing diarrheal illness is predominantly a disease of the elderly, so this requirement GREATLY impeded timely enrollment. No protocol deviations have occurred. The current rationale behind FMT for CDI is that the introduction of microbes from a healthy donor should allow for the restoration of a normal microbial community in the diseased host with consequent suppression of C. difficile colonization and disease pathogenesis. The first modern use of FMT was reported in a 1958 case series of 4 patients with pseudomembranous enterocolitis. The first case of confirmed CDI treated with FMT was reported in 1983; treatment was curative. Until 1989, retention enemas were the most common technique for FMT. Alternative methods for delivering FMT have included fecal infusion via duodenal tube (1991), rectal tube (1994), and colonoscopy (1998). FMT for recurrent CDI has been used successfully, whether administered by nasogastric tube, rectal administration by colonoscopy, or rectal tube, including self-administration at home by enema. FMT has proven to be remarkably effective and remarkably safe without any significant problems (see below and attached reviews and meta-analyses). Increasing interest is emerging regarding the changes in the intestinal microbiota associated with CDI. In 2008 Chang et al. constructed small (< 200 sequences per subject) 16S rRNA gene libraries from the stools of 4 patients with first-time CDI and 3 patients with recurrent CDI. Based on 16S rRNA gene classification, they found that the fecal microbiomes of patients with an initial episode of CDI were similar at the phylum level to healthy subjects (i.e., the majority of sequences belonged to dominant fecal phyla Bacteroidetes and Firmicutes), while a major reduction or loss of Bacteroidetes was observed in patients with recurrent CDI. The loss of the Bacteroidetes was accompanied by the expansion of other phyla, including Proteobacteria and Verrucomicrobia, which are normally minor constituents of the fecal microbiota. Khoruts et al. (2010) compared the microbiota of a patient with recurrent CDI before and after FMT by using terminal-restriction fragment length polymorphism and clone-based 16S rRNA gene sequencing. Before transplantation, the patient's microbiota was deficient in members of Bacteroides and instead was composed of atypical fecal genera such as Veillonella, Clostridium, Lactobacillus, Streptococcus, and unclassified bacteria similar to Erysipelothrix. Two weeks after the infusion of donor fecal suspension, the bacterial composition of her feces approached normal and was dominated by Bacteroides sp. strains. In 1989, Tvede and Rask-Madsen used a combination of nine normal fecal organisms to treat 6 patients with chronic relapsing C. difficile diarrhea. These investigators cultivated 10 strains of bacteria, including Enterococcus (Streptococcus) faecalis (1108-2), Clostridium inoculum (A27-24), Clostridium ramosum (A3I-3), Bacteroides ovatus (A40-4), Bacteroides vulgatus (A33-14), Bacteroides thetaiotaomicron (A33-12), Escherichia coli (1109), E. coli (1108-1), Clostridium bifermentans (A27-6), and Blautia producta (Peptostreptococcus productus) (1108-2) in broth for 48 h to a concentration of approximately 10 to the 9th power bacteria/mL. Two mL from each bacterial culture were admixed with 180 mL saline that had been pretreated in an anaerobic chamber for 24 h; the bacterial suspension was then instilled rectally. This procedure was followed promptly by a decline of C. difficile to undetectable levels by culture and the loss of detectable toxin from the stools. Normal bowel function was restored within 24 hours, and abdominal symptoms disappeared. Stool cultures and toxin assays for C. difficile remained negative during a year of follow-up. It is especially important to note that feces from none of the 6 patients contained Bacteroides sp.

This study will initially enroll 12 subjects. Based on the relatively large experience with FMT reported to date, we anticipate a prompt and sustained response as determined by a cessation of fever, leukocytosis and diarrhea and a loss of abdominal discomfort in >8 (80%) of the subjects. If such a response is not observed in >3 of the first 6 patients treated, we will reevaluate the composition of the FMT mixture and/or consider whether more than one inoculum should be given. If the 80% or greater success rate is achieved, clinical follow-up and studies of the microbiome will continue for a total of 12 months. During that time, we would plan to obtain new approvals to test other methods of inoculum delivery (e.g., via NG tube or oral administration in a buffered solution). In contrast to the use of raw stool of unknown and variable composition, the inoculum is defined and grown under anerobic conditions with defined media. Pre-administration oft therapy the inoculum will be examined by gram strain and then will be cultured to ensure that it contains the orgamisms that were used. The chance of a contaminant is thought to be very unlikely and contamination with an unknown pathogen essentially impossible. As an additional safety measure the inoculum is given into the gastrointestinal tract which is designed for food and other non-sterile materials. We anticipate a prompt and sustained response, as determined by a cessation of fever, leukocytosis and diarrhea and a loss of abdominal discomfort in >8 (80%) of the subjects. If such a response is not observed in >3 of the first 6 patients treated, we will reevaluate the composition of the FMT mixture and/or consider whether more than one inoculum should be given. If the 80% success rate is achieved, clinical follow-up and studies of the microbiome will continue for a total of 12 months. Inclusion Criteria: Only VA patients will be eligible for the study if they have had a confirmed diagnosis of CDI that has been treated for 10-14 days with recommended doses of metronidazole or vancomycin and has either failed to respond, or has responded and relapsed within 4 weeks of the end of treatment. The diagnosis will be regarded as confirmed by the presence of diarrhea (>3 unformed stools in a 24-hour period for 2 successive days) and abdominal discomfort. The presence of fever, leukocytosis, and a serum albumin <3 gm/dL will be recorded but will not be necessary for the diagnosis. Patients will be included after they have given informed consent and signed the appropriate consent form that has been approved by the Baylor IRB. Procedure: Patients with proven recurrent CDI at the MEDVAMC will be asked to participate. The health records of all VA patients entered will be flagged that they are enrolled in a reseach study. We presume that the strains described by Tvede and Rask-Madsen were clinical isolates as there is no record of these strains in later publications and they are not available from any of the public strain repositories such as ATCC, DSMZ, or CCUG. In place of these specific strains, we will use the following fully sequenced type strains, originally isolated from human feces, and purchased from ATCC: Bacteroides ovatus ATCC 8483, Bacteroides vulgatus ATCC 8482, and Bacteroides thetaiotaomicron ATCC 29148. Primary cultures will be grown according to GLP procedures (SOP attached) on pre-reduced anaerobic blood agar ANABA (Remel, Lenexa, KS), then broth cultures of each strain will be anaerobically cultivated at 37°C for 48 hours in pre-reduced chopped meat medium that will be made in house. Bacteria will be collected by centrifugation at 1,600 x g for 10 minutes and resuspended in 5 mL sterile pre-reduced 0.9% saline, using Macfarlane standards to estimate bacterial concentrations to target a final concentration of 10 to the power 8 to 9 cfu/mL. Aliquots (0.1 mL) will be removed for serial dilution and plating and anaerobic incubation at 37°C on pre-reduced ANABA plates to determine actual bacterial concentrations. Two mL of each of the three saline suspensions will be combined with 194 mL sterile pre-reduced 0.9% saline to form the infusion cocktail. The 200 mL bacterial suspension will then be administered into the small intestine using an Olympus ultra-slim gastro-duodenoscope. The small bowel route is chosen to avoid the potentially deleterious effects of gastric acid on survival of the inoculum and to prevent or reduce reflux of the volume administered into the stomach. Subjects will be then followed for up to 12 months. Transplant procedure: Patients will be pretreated with 4 days of oral vancomycin (125 mg q 6 hr) to reduce the C. difficile load, with the final dose being given at 6 PM on the evening before the procedure (approximately 14-16 hours before FMT). On the evening before FMT and again on the morning of the procedure, patients will receive 20 mg of omeprazole by mouth. On the morning of the procedure, a "super-slim" 5.5 mm diameter gastroscope will be passed per mouth into the patient's small intestine with an attempt to place the tip at or beyond the ligament of Treitz. Two hundred mL of the bacterial suspension will be instilled into the small intestine via a catheter introduced through the biopsy channel of the endoscope and the flushed with 25 mL of sterile pre-reduced 0.9% saline. After removal of the endoscope, after recovery, patients will be allowed to resume a normal diet and physical activities. Subjects will be seen daily when in the hospital (and contacted daily by phone after they are discharged) for 14 days. Subjects will then be contacted at 30 days and monthly for 3 months, then every 3 months for one year. Symptoms will be recorded at each contact using a standard questionnaire. Laboratory examinations including CBC with differential, basic metabolic panel and albumin will be obtained at 3 days, at 10 days and at 30 plus or minus 7 days. The results of clinical outcome will be combined with evaluation of the effect of the therapy on the fecal microbiome. Stools sample will be collected regularly up to 1 year or until proven relapse. Stools samples and rectal swabs will be otained daily during the initial study period until discharge from hospital or 14 days and then stool samples at 30 days and at 3 month intervals for one year. After discharge from hospital samples will be sent to the laboratory using cold packs by Federal Express. This should allow us to examine in detail the effect of the therapy on the C. difficile as well as on the fate of the administered bacteria (for which the entire genome has been sequenced and published). for details see attachments).

Two hundred mL of the bacterial suspension (microbiota transplantation) will be instilled into the small intestine via a catheter introduced through the biopsy channel of the endoscope and the flushed with 25 mL of sterile pre-reduced 0.9% saline. After removal of the endoscope, after recovery, patients will be allowed to resume a normal diet and physical activities.

Binomial: success defined as resolution or no resolution of disease. Resolution is defined as either complete cessation of clinical symptoms or diagnostic confirmation of the absence of disease, during the period of follow-up after transplantation.

The outcome is binomial. Success defined as resolution or not. Resolution is defined as either complete cessation of symptoms or diagnostic confirmation of the absence of disease during the follow-up period. Resolution after 1 treatment is counted separately from resolution after retreatment due to failure. The scales used are failure or resolution with Type 1 Failure being defined as the continued occurrence of clinical illness during (1) the period of 8 weeks follow-up & before retreatment or (2) Type 2 Failure as failure during the period of 8 weeks of follow-up subsequent to retreatment. Relapse is defined as resolution, with subsequent return of signs & symptoms during the follow-up period confirmed by return of stool toxin and/or the typical endoscopic findings of C. difficile colitis.

Inclusion Criteria: Only VA patients will be eligible for the study if they have had a confirmed diagnosis of CDI that has been treated for 10-14 days with recommended doses of metronidazole or vancomycin and has either failed to respond, or has responded and relapsed within 4 weeks of the end of treatment. The diagnosis will be regarded as confirmed by the presence of diarrhea (>3 unformed stools in a 24-hour period for 2 successive days) and abdominal discomfort. The presence of fever, leukocytosis, and a serum albumin <3 gm/dL will be recorded but will not be necessary for the diagnosis. Patients will be included after they have given informed consent and signed the appropriate consent form that has been approved by the Baylor IRB. Exclusion Criteria: Exclusion criteria include: treatment with major immunosuppressive agents including prednisone >10 mg/day (or its equivalent), calcineurin inhibitors, mammalian target of rapamycin (mTOR) inhibitors, lymphocyte-depleting biological agents, anti-tumor necrosis factor agents, and others; chemotherapeutic antineoplastic agents; decompensated liver cirrhosis; serum creatinine >4 or need for hemodialysis; presence of an active malignancy other than superifical skin cancer (eg, basal cell); HIV/acquired immune deficiency syndrome; recent bone marrow transplant, or other cause of severe immunodeficiency; requirement for concurrent antimicrobial therapy; contraindication for ultra-slim endoscopy including severe chronic heart or lung disease; a chronic bedridden state; and any other condition suggesting that life span will not be >1 yr. -

Tvede M, Rask-Madsen J. Bacteriotherapy for chronic relapsing Clostridium difficile diarrhoea in six patients. Lancet. 1989 May 27;1(8648):1156-60. doi: 10.1016/s0140-6736(89)92749-9.

van Nood E, Vrieze A, Nieuwdorp M, Fuentes S, Zoetendal EG, de Vos WM, Visser CE, Kuijper EJ, Bartelsman JF, Tijssen JG, Speelman P, Dijkgraaf MG, Keller JJ. Duodenal infusion of donor feces for recurrent Clostridium difficile. N Engl J Med. 2013 Jan 31;368(5):407-15. doi: 10.1056/NEJMoa1205037. Epub 2013 Jan 16.