Microbiome Therapy Achieves High ORR and Survival in Acute GI GVHD

MaaT013, a pooled allogeneic fecal microbiotherapy, significantly outperformed historical benchmarks for response and survival in patients with refractory acute graft-vs-host disease (aGVHD) affecting the gastrointestinal (GI) tract (GI-aGVHD), according to final results from the phase 3 ARES trial (NCT04769895) presented by Florent Malard, MD, PhD, at the 52nd Annual Meeting of the European Blood and Marrow Transplantation (EBMT) Society in Madrid, Spain.¹

“We, using the microbiotherapy MaaT013, achieved some deep and durable response rates that translate into good overall survival [OS] in a population that is known for dismal clinical outcomes,” Malard, professor of hematology at Sorbonne University in Paris, France, said in his presentation.

An Area of Critical Unmet Need

GI-aGVHD is a severe complication of allogeneic hematopoietic cell transplantation driven by donor T-cell attack on host tissues. Although corticosteroids represent the standard first-line treatment and ruxolitinib (Jakafi) is approved for steroid-refractory disease, no approved therapies exist beyond these 2 options. Off-label best available therapies have shown limited benefit, and third-line patients face a median OS of just 86 days, with a 12-month survival rate of approximately 29%.²

MaaT013, developed by MaaT Pharma, is a high-richness, high-diversity pooled allogeneic fecal microbiota product containing a minimum of 1.35 × 10¹¹ viable bacteria per dose. It is administered rectally as a 150-mL suspension across 3 doses on days 1, 5, and 10, with an optional fourth dose permitted in cases of relapse after day 28 response.¹

Trial Design and Patient Population

ARES was a multicenter, open-label phase 3 study enrolling adult patients with grade 2 to 4 GI-aGVHD who were refractory to corticosteroids and either refractory or intolerant to ruxolitinib. Patients with cytomegalovirus colitis, overlap chronic GVHD, hyperacute GVHD, or active uncontrolled infection were excluded. Prior to receiving MaaT013, all enrolled patients underwent vancomycin pretreatment, an important procedural step in preparing the gut environment for microbiome restoration.

Of 90 patients assessed for eligibility, 66 received at least 1 dose of MaaT013 and comprised the full analysis set. The median age was 55.5 years, and 53% of patients were male. Of the enrolled patients, 58% had grade 3 and 33% had grade 4 aGvHD, all were ruxolitinib refractory, and 86% were steroid refractory. Most patients (77%) had exclusive lower GI involvement, and 88% had classical aGVHD. Treatment delivery was generally well maintained, with 92% of patients receiving at least 2 doses and 80% receiving the full 3-dose course. Among those who received the enema, the median retention time was 124 minutes, with a mean of 151.6 minutes.

Meeting Key Objectives

The study met its primary end point with a day 28 GI overall response rate (GI-ORR), assessed by an independent review committee (IRC), of 62.1%, significantly exceeding the preestablished historical control threshold of 22% (P <.0001). Notably, IRC and investigator assessments were highly concordant, with agreement rates of 96.4% for day 28 GI-ORR and 94.5% for day 28 all-organ ORR. Responses were deep, with 38% of patients achieving complete response and 20% achieving very good partial response at day 28. The all-organ overall response rate at day 28 was similarly high at 63.6%.

Responses proved durable: GI-ORR was maintained in 47.0% of patients at day 56 and 44% at month 3, with complete responses predominating at both timepoints. The estimated cumulative incidence of loss of GI response at 12 months was 20%, whereas loss of all-organ response was 26% at month 6.

Responses Translate to Survival

OS results were particularly striking given the historically poor outcomes in this population. At a median follow-up of 344.5 days, the median OS was not reached, with a rate of 59% at 6 months and 54% at 12 months, far exceeding previously reported 12-month survival rates of around 15% in comparable third-line populations.

The survival benefit was strongly linked to treatment response. Day 28 GI responders had 6- and 12-month survival rates of 76% and 68%, respectively, compared to just 28% at both timepoints among nonresponders (log-rank P <.0001). Among the 29 patients (44%) who died during the study, the leading causes were severe infections (n= 13), disease progression (n = 5), general health deterioration (n = 4), rectal hemorrhage (n = 2), underlying malignancy relapse (n = 2), and cerebral hemorrhage or cardiorespiratory arrest (n = 2 and n = 1, respectively).

Acceptable Safety in a Vulnerable Population

Treatment-related adverse events (AEs) occurred in 29% of patients, with the most common being constipation (7.6%), abdominal pain, and septic shock (4.5% each). Bacterial bloodstream infections were observed in 51% of patients overall, reflective of the severe immunocompromise and damaged intestinal mucosa characteristic of this population rather than a direct consequence of treatment.

“It can be concerning to have this kind of infectious events, but we must remember that we’re speaking about patients with GI GVHD that had steroids, that had ruxolitinib; we know this is a kind of patient [who] have a lot of sepsis and that have a lot of [bacterial] translocation, even if you don’t give those patients microbiotherapy,” Malard said.

Serious AEs were reported in 80% of patients overall, although the majority reflected the severe underlying GVHD. Only 8 serious AEs in 7 patients were considered treatment related by investigators, and 1 death was deemed related to MaaT013. The investigators characterized the overall safety profile as consistent with what would be expected in an immunocompromised, heavily pretreated, often cytopenic population with damaged intestinal mucosa at heightened risk of infection.

The investigators concluded that MaaT013's response rates and survival outcomes support a favorable benefit-risk profile and position it as a potential first microbiome-based therapy for refractory acute GvHD.

REFERENCES

1. Malard F, Sanz J, Mediavilla et al. MaaT013 for ruxolitinib-refractory acute graft-versus-host disease with gastrointestinal involvement: final results from the ARES phase III trial. Presented at: 2026 EBMT Annual Meeting; March 22-25, 2026; Madrid, Spain. Abstract GS2-8.

2. Clausen J, Perez Simon JA, Carre M, et al. Key results from the CHRONOS multicentre retrospective cohort study describing patient outcomes in third-line acute gastrointestinal GVHD. Presented at: 2026 EBMT Annual Meeting; March 22-25, 2026; Madrid, Spain. Poster B005.