Gut microbiota restoration with oral pooled fecal microbiotherapy after intensive chemotherapy: the phase 1b CIMON trial

Abstract

Intensive induction chemotherapy (IC) combined with broad-spectrum antibiotics for acute myeloid leukemia (AML) leads to gut microbiota dysbiosis, promoting pathological conditions and an increased incidence of complications, possibly limiting eligibility to allogeneic hematopoietic cell transplantation (alloHCT). The purpose of this dose-ranging phase 1 study (CIMON) was to evaluate the first-in-man use of MaaT033, a pooled, allogeneic, lyophilized, and standardized fecal microbiotherapeutic product, formulated as a delayed-release capsule for oral administration. Primary objectives of the study were to evaluate the maximum tolerable dose of MaaT033 in 21 patients with AML having undergone IC and antibiotics. Secondary objectives were to assess MaaT033 safety, its efficacy in restoring the patients' gut microbiome using shotgun sequencing to evaluate the recommended dose regimen, and patient compliance. MaaT033 was shown to be safe and effective for gut microbiota restoration in patients with AML receiving IC and antibiotics, with an excellent gut microbiota reconstruction based on diversity indices at the species level and restoration of microbial communities close to the composition of the drug product. The maximum tolerable dose of MaaT033 was not determined because the interim results suggested adequate efficacy as measured by engraftment at lower doses (3 capsules per day). Moreover, inflammatory markers (C-reactive protein, interleukin-6) decrease with treatment, whereas short-chain fatty acids increase over time. A randomized, placebo-controlled phase 2b trial, in recipients of alloHCT patients is ongoing. This trial was registered at www.clinicaltrials.gov as #NCT04150393.

Graphical abstract

Figure 1.

Dose ranging and study flowcharts. (A) Details of dose regimen per cohort. Of note, no patient was treated in cohort 5. (B) Study flowchart. The dose-escalation process was monitored by the data and safety monitoring board. Visit 1: treatment start (D1) at the end of neutropenia. Feces and blood collection. Visit 2: safety interim visit (D7 ± 2), clinical assessment visit. Feces and blood collection. Visit 3: start of consolidation or other cycle of chemotherapy (D19 ± 5), clinical follow-up. Feces and blood collection. Visit 4: end of study (D44 ± 10) corresponding to the end of chemotherapy. Feces and blood collection. The recommended timing for MaaT033 administration was 1 capsule before breakfast (for cohorts 1 and 2), 3 capsules before breakfast (for cohorts 3 and 4), and 3 capsules before breakfast, 3 before lunch, and 3 before dinner (for cohort 5).

Figure 2.

MaaT033 impact on patients’ gut microbiota richness, engraftment, and Butycore levels relies on baseline microbiota (n = 21). For each patient’s cohort, for MaaT033, and at each visit (D1, D7, D19, and D44): (A) Richness index at species level. (B) Shannon index at species level. (C) Relative abundance of Butycore, a group of 15 SCFA (mainly butyrate)-producing bacterial genera (Blautia, Faecalibacterium, Alistipes, Eubacterium, Bifidobacterium, Ruminococcus, Clostridium, Coprococcus, Odoribacter, Roseburia, Anaerostipes, Oscillibacter, Subdoligranulum, Butyrivibrio, and Holdemanella). (D) Engraftment at species level. The definition of engraftment is linked with the detection of donor-specific bacterial species that were not present in the patient’s feces before MaaT033 administration. For panels A to D, medians with interquartile ranges are provided. (E) Engraftment of MaaT033 at each visit by richness measured at D1 (at species level). Statistical significance was performed using Pearson correlation test.

Figure 3.

List of species engrafted in patients. Each column corresponds to a patient (n = 17), sorted by cohort (1, 2, 3, 4) and time point (D7, D19, D44). Each line corresponds to a species that engrafted in at least 1 patient at a given time point. Blue indicates that the species has engrafted (ie, present in the product and found in the microbiota of the patient at a given time point when not present at baseline), whereas yellow indicates that the strain was found in the microbiota of the patient at a given time point but is not considered engrafted because it was present in both the patient at baseline and the product.

Figure 4.

Number of donors determined to have specifically contributed to the engraftment of MaaT033 species in patients’ gut (n = 17). For cohorts 1 and 2, MaaT033 batch comprised fecal material from 6 individual donors. For cohort 3, MaaT033 batch comprised fecal material from 5 individual donors. For cohort 4, MaaT033 batch comprised fecal material from 8 individual donors. The definition of specific engraftment (in green) is linked with the detection of donor-specific bacterial species and does not include the bacteria, which may be shared by at least 2 donors (ie, not specifically engrafted, in yellow).

Figure 5.

PCA performed with the host parameters data set (n = 21). Individual samples are represented by points colored and linked according to the corresponding visit. The arrows depict the variables, here the host parameters. The longer they are, the more they drive the samples’ projections; the closer they are, the more they vary together. Specific parameter groups are highlighted in color (SCFAs, chemokines, and inflammatory parameters). For a better visualization, variables with negligible impact on sample projections (with small arrows) were removed from the figure after computation. CRP, C-reactive protein; IL, interleukin; IP, Interferon gamma-induced protein 10; MCP-1, monocyte chemoattractant protein 1; PC1, principal component 1; PCA, principal component analysis; TNF, tumor necrosis factor.