Prevention of Dysbiosis Complications with Autologous Fecal Microbiota Transplantation (auto-FMT) in Acute Myeloid Leukemia (AML) Patients Undergoing Intensive Treatment (ODYSSEE study): First Results of a Prospective Multicenter Trial

Introduction. There is a growing body of evidence highlighting the possible interactions between the microbiota and cancer. However, little is known about the consequence on the microbiota of the intensive chemotherapy in patients with AML. Indeed, intensive treatments of AML are known to negatively impact gut microbiota composition; it is measured by a deep shutdown of diversity indices (eg. Simpson), also called dysbiosis. Decrease in microbiota diversity may impact the immune system and may have consequences on the outcome of AML patients. Therefore, development of strategies to manipulate the gut microbiota may minimize treatment-related complications and potentially improve outcomes. With this background, we developed a single arm prospective phase I/II multicenter trial (ClinicalTrials.gov Identifier: NCT02928523) aiming to evaluate use of auto-FMT in AML patients treated with intensive chemotherapy and antibiotics.

Patients and methods. For this purpose, at the time of admission and AML diagnosis, patients were requested to donate stools that were screened according to standard recommendations. If deemed appropriate according to protocol criteria, stools were conditioned, processed with a patented diluent, and stored frozen until transplantation performed after recovery from aplasia and before the second course of chemotherapy. The auto-FMT inoculum was administered as an enema, and was prepared for each individual patient according to standardized and reproducible processes developed by the sponsor of the trial. The primary endpoint was the recovery of microbiota diversity and correction of dysbiosis after auto-FMT. The microbiome diversity and richness were measured at baseline (prior to induction chemotherapy), after induction chemotherapy, and around 10 days post-auto-FMT. Microbiome restoration was assessed at the genus and species level, on the basis of metagenomic results obtained for the 3 sequential samples. Illumina TruSeq Nano Prep kit was used for sequence-library preparation and samples were sequenced on HiSeq 2500 2x125bp, 40 million reads/sample. Secondary objectives included monitoring of the immune system, safety and feasibility of auto-FMT, and a set of parameters exploring the impact of auto-FMT on different patient-related outcomes (clinical status and recovery).

Results. At time of this planned intermediate analysis, 35 AML patients eligible for intensive induction chemotherapy (intermediate and unfavorable karyotype) with a median age of 53 (range, 23-67) years were screened. The current interim results concern the 9 patients who fulfilled all criteria for inclusion and were able to proceed to auto-FMT after intensive induction therapy ("3+7" induction or equivalent). Prior to the start of induction chemotherapy, the mean baseline Simpson index (measured at genus level) was 0.836 (95%CI, 0.755-0.987). At the end of induction chemotherapy, a highly significant decrease of -0.519 ([-0.677, -0.361], p<.001) was observed, highlighting the deleterious impact on the gut microbiome of intensive chemotherapy and the multiple antibiotics received during induction. Per protocol, these patients were able to receive an auto-FMT prior to receiving their subsequent consolidation chemotherapy. Auto-FMT was well tolerated, only one serious adverse event was reported: a case of E. Coli infection which resolved after appropriate antibiotherapy. After auto-FMT, the Simpson index returned back to baseline level (difference=0.023 ([-0.135, 0.181], p=0.776), and virtually identical results were obtained when Simpson was measured at species level. The ongoing immunomonitoring of these 9 patients suggested data correlated with the dysbiosis correction.

Conclusions. This first multicenter prospective trial reached its primary endpoint in all analyzed cases thus far, and established the capacity of auto-FMT to correct dysbiosis and restore a normal microbiota in AML patients receiving intensive induction chemotherapy and wide-spectrum antibiotics. Correlations with clinical and immune parameters that may influence AML outcome are currently being analyzed and will be presented. The apparent modulation of the immune system through microbiota restoration is a promising venture in the treatment of AML patients warranting further investigations. (The trial was funded by MaaT Pharma whose product was tested in this protocol).