Clostridioides difficile infection (CDI) is the leading cause of nosocomial diarrhea in Europe, with over 120,000 cases and almost 3,700 deaths per year. This infection is characterized by a high risk of recurrence after cure, ranging from almost 20% after a first episode to over 60% after 2 recurrences, or in the case of specific risk factors. Currently, first-line treatment of CDI is based on oral antibiotics such as fidaxomicin or vancomycin. These antibiotic treatments, which are effective in 89% and 86% of first-episode cases respectively, do not correct the microbiological imbalance underlying the onset of CDI and may, on the contrary, encourage recurrence by contributing to the maintenance of a deleterious change in the microbiota (dysbiosis) through the elimination of bacteria other than C. difficile, due to their spectrum of activity. In a number of patients, this ecological imbalance can no longer be restored after antibiotic treatment, leading to multiple recurrences of CDI. In this context, fecal microbiota transplantation (FMT) has been validated for over 10 years for the prevention of recurrence in multi-recurrent CDI. The principle of FMT is based on the use of a pharmaceutical preparation made from the stool of a healthy donor, administered within the digestive tract of a patient for therapeutic purposes. Currently, in the case of multiple recurrences, it is the recommended first-line treatment (from 2 recurrences) and the most effective, with a clinical efficacy preventing recurrence of CDI in 69% to 89% of cases at 8 weeks post-treatment, with a good safety profile. Among the microbial factors promoting CDI, the loss of the bacterial species Faecalibacterium prausnitzii constitutes a specific therapeutic target. F. prausnitzii is a commensal bacterium of the human gut, making up nearly 5% of the fecal microbiota, and has been shown to be associated with an individual's state of health. A drop in its relative abundance is associated with an increased risk of numerous diseases, such as Crohn's disease and colorectal cancer. In CDI, F prausnitzii is greatly diminished. Moreover, low abundance of F. prausnitzii is predictive of C. difficile recurrence. Its abundance in stools is increased after FMT and is also predictive of response to treatment. From a pathophysiological point of view, one of the preventive effects of F. prausnitzii on recurrence would be mediated by its ability to hydrolyze the bile acids involved in the germination of C. difficile spores. The aim of this Phase I/II trial is to assess the efficacy and safety of oral administration of EXL01, a single isolated unmodified strain of F. prausnitzii, in preventing CDI recurrence in high-risk patients at W8. The study will be conducted in 2 parts. The phase I (Part A) is planned to include 6 patients. The phase II (Part B) will include 50 patients in two arms (25 patients respectively in the placebo and EXL01 arm).