Biomarkers Evaluated as Predictors of Success or Failure of Fecal Microbiota Transplantation (FMT) in Patients with Multiple (≥3) Bouts of Clostridium difficile Infection (CDI)

Student: Jude Fahoum
Mentor: Herbert DuPont ’61 (Texas School of Public Health)

The unnecessary overuse of antibiotics has set obstacles in the progressive pathway such wonder drugs paved for medicine. One example can be seen through the increase in healthcare-associated bacterial infections mainly caused by Clostridium difficile. The most effective treatment for Clostridium difficile infections (CDI) is Fecal Microbiota Transplantation (FMT) in which stool from healthy donors is administered to patients with CDI through enema or pills. This study aimed to find a biomarker that can serve as a predictor for success/failure of FMT in the tens of thousands of patients seen each year with this complication.

Clostridium difficile infection (CDI) is the most common hospital-associated infection resulting in 70% of all fatalities from diarrhea-causing pathogens in the U.S. It develops in people with depleted gastrointestinal microbiota and recurs in 25% of patients (RCDI). The treatment of choice for RCDI is microbial restoration through a process called fecal transplantation using healthy donor bacteria which cures ~90% of patients after a single transplantation. There is currently no biomarker to see which patients with CDI should receive FMT and nor is there one to predict the success/failure of FMT in the hundreds of thousands of patients seen each year in the United States with this complication.

The goal of this study was to find biomarkers predictive of success/failure of FMT. Fifty-one subjects with ≥ 3 CDI bouts undergoing FMT at University of Texas Program for Restoration of Intestinal Microbiota and providing stool samples at days 0 (pre-FMT) and days 7, 14 and 30 (post FMT) were enrolled. The stool samples were assayed for inflammation biomarkers, calprotectin and lactoferrin, and markers of C. difficile (CD), CD antigen (GDH) and CD toxins A/B by commercial immunoassay, using positive and negative controls and standard curves. FMT recipients were grouped into successful or unsuccessful based on cure or recurrent CDI during the 2 months after FMT. Subjects becoming ill requiring anti-CD therapy before they could provide the four stools were excluded from the study.

51 subjects undergoing FMT for RCDI provided four stool samples. The small quantity of stool available did not allow all studies to be performed at all time points for some subjects, explaining different assay numbers. Median fecal lactoferrin concentrations at four respective time points showed non-significant increases in the groups failing treatment (n=16): 2,114 ng/ml; 5,882 ng/ml; 4,764 ng/ml and 2,025 ng/ml versus those cured (n=35): 735 ng/ml; 1,812 ng/ml; 1,072 ng/ml; and 882 ng/ml (P=0.1738). The median calprotectin concentrations in the respective stools from 16 subjects failing FMT were: 0 µg/g; 7,493 µg/g; 1,670 µg/g and 0 µg/g versus 0 µg/g for each of the four samples provided by the 34 subjects cured (P = 0.001). Presence of fecal GDH in the four respective stools in the 15 subjects failing FMT were more often positive: 1 (6.7%); 11 (73.3%); 8 (53.3%); and 8 (53.3%) versus the samples provided by the 34 subjects with cures: 1 (2.9%); 4 (11.8%); 7 (20.6%) and 7 (20.6%) (P=0.001). Presence of fecal CD toxins occurred in a greater frequency in the 16 subjects studied failing FMT: 2 (13%); 12 (75%); 8 (50%); 4 (25%) versus the 35 cured subjects: 2 (5.7%); 0 (0%); 0 (0%) and 0 (0%) (P<0.0001).

The study identified fecal markers predictive of success or failure of FMT in the management of multiple recurrent CDI. All values were similar comparing pre-FMT samples in the subjects who failed treatment or were cured. On day 7 after FMT a divergence became obvious. Presence of fecal calprotectin, CD antigen (GDH) and CD toxin(s) at that point after FMT were highly predictive of clinical failure of the microbial restoration treatment. Limitations of the study include a subset of subjects failing FMT before 30 days being excluded from the study, minimizing the number of failures in the study. Also, fecal samples were insufficient in quantity to perform all assays in given subjects. This study identified three strong fecal marker predictors of success in FMT: calprotectin, GDH and CD toxins. The findings have therapeutic and prognostic implications for treatment of this important chronic disorder seen in subjects receiving antibiotics or being confined in hospitals.