Background:

Acute graft-versus-host disease (aGVHD), especially steroids-refractory aGVHD (SR-aGVHD), is still one of the most severe complications after haploidentical hematopoietic stem cell transplantation (haplo-HSCT). Previously, we reported the high overall response rate of 71.4% after the combined treatment of fecal microbiota transplantation (FMT) with ruxolitinib as a salvage treatment in intestinal SR-aGVHD. Also trials have been applied in prophylaxis in order to prevent SR-aGVHD, but substantial challenges remain. Autologous FMT (auto-FMT), using the individual's own stool, may overcome the limits of suitable FMT donors and mitigate the risks of disease transmission from a donor stool. Here we report the results of a prospective, randomized trial (ClinicalTrials.gov ID: NCT03148743) that was conducted to compare the efficacy and safety of auto-FMT as the GVHD prophylaxis with non-FMT group in preventing aGVHD.

Methods:

Patients with malignant hematological diseases who met the inclusion criteria, aged from 18 to 60 years old, had autocoprobiotic collection and retention before haplo-HSCT. Myeloablative conditioning regimen based on BUCY was performed according to the disease type. After the hematopoietic reconstruction and 20 days after haplo-HSCT, patients received auto-FMT in a form of capsules,while control group received empty capsules three times daily for swallowing. Overall survival (OS), aGVHD relapse rate and malignancy relapse rate were assessed. The follow-up period was started after the first enrolled case, and ended in 1 year after the last.

Results:

By the last follow-up, 76 patients underwent haplo-HSCT were enrolled in the study, including 36 in auto-FMT group,with a median age of 34 (range 18-54) years, and 40 in control group, with average age of 32 years old (range: 18-64). The primary diseases included acute myeloid leukemia in 52.6% patients, acute lymphoblastic leukemia in 35.5%, and other hematological malignancies in 11.9%. Baseline characteristics were balanced between the two groups. 78.9% (60/76) enrolled patients also received GVHD prophylaxis with cyclosporine A, MMF, short-term MTX and rabbit anti-thymocyte globulin (ATG), while others received tacrolimus instead of cyclosporine A to reduce the organ toxicity.

Cumincidence of aGVHD was 22.2% (8/36) in auto-FMT group, while 42.5% (17/40) in non-FMT group (P = 0.04). Intestine-involved case was observed in the 1/8 aGVHD patients in auto-FMT group. Others manifested skin rash, undergoing grade 1 aGVHD, and soon be controlled by the use of steroids or the adjustment of immunosuppression. 8/17 aGVHD patients in non-FMT group underwent sever enteritis with diarrhea, among whom 2 patients died at last follow-up. Three patients in auto-FMT group experienced chronic GVHD (cGVHD, 3/36, 8.3%), while ten patients(10/40, 25%) in non-FMT group (P = 0.06).

Viral reactivations (50% vs 52.5%), hemorrhagic cystitis (22.2% vs 30%), and pneumonia (22.2% vs 37.5%) were the most frequent adverse events observed in our study. On the other hand, the relapse of the malignancy occurred in 13.8% (5/36) and 27.5% (11/40) in respective group. With a median follow-up of 14.5 (range 0.5-35.2) months for all 76 patients, 62 patients survived and 14 died. In auto-FMT group, three patients (3/36, 8.3%) died, including 2 of relapse and 1 of infection; in non-FMT group, 11 patients (11/40, 27.5%) died, including 8 of relapse, 2 of aGVHD and 1 of infection. The 2-year overall survival (OS) was 90.2% vs 64.5%, respectively. Auto-FMT prophylaxis showed significantly longer OS compared with non-FMT group (P = 0.03).

Conclusion:

Our study underlined the effective GVHD prophylaxis of auto-FMT after haplo-HSCT, significantly reducing the cumincidence of aGVHD and cGVHD, as well as improving the survival.

Disclosures

No relevant conflicts of interest to declare.

This content is only available as a PDF.
2024
Sign in via your Institution