Invited Commentary 

Fecal Microbiota Transplant for Multidrug Resistance—No Benefit Without Disruption?

Brendan J. Kelly, Michael H. Woodworth, Jennie H. Kwon

JAMA Intern Med Published Online: April 20, 2026

doi: 10.1001/jamainternmed.2026.0668

Fecal microbiota transplant (FMT) has transformed the medical treatment of recurrent Clostridium difficile infection (R-CDI). An explosion of randomized clinical trials during the 2010s demonstrated the effectiveness of FMT in treating patients with R-CDI, which has become the archetypal disorder of dysbiosis that is characterized by a collapsed microbiome with low bacterial community diversity and depleted microbial metabolites. FMT, delivered enterally across multiple formulations, cured greater than 90% of patients with R-CDI in some trials,¹ prompting medical societies to recommend FMT for R-CDI and spurring efforts to design more scalable alternatives to FMT that were manufactured with either bacterial consortia from homogenized human gut communities or a consortia of cultured bacteria. There are currently 2 fecal microbiota products (FMPs) approved by the US Food and Drug Administration for preventing R-CDI, and a consortium of cultured gut bacteria, termed a live biotherapeutic product (LBP),² is currently in phase 3 trials. FMPs and LBPs, which work by shaping gut microbial ecology, represent an entirely new approach to treating infectious diseases.³

As FMT was applied to treating R-CDI, researchers recognized that FMT might also exclude other multidrug-resistant organisms (MDROs) from gut microbial communities. The potential benefit of FMT to achieve MDRO decolonization or pathogen burden reduction prompted excitement given the established association between persistent MDRO gut colonization and MDRO infection risk. Early trials of FMT for MDRO decolonization sustained this enthusiasm and suggested mechanisms by which FMT might be broadly effective.^4,5 In this issue of JAMA Internal Medicine, Narang and colleagues⁶ present evidence from a new randomized clinical trial of FMT for MDRO decolonization, which provides an opportunity to recalibrate our expectations for FMT.

In a randomized, double-blind trial, 114 patients with gastrointestinal illnesses and persistent MDRO colonization defined by serial rectal swabs were randomized 1:1 to receive FMT vs a sham treatment. The authors⁶ found that there was no significant difference in the percentage decolonized across the FMT (31.0%) and sham (30.4%) groups. Antimicrobial resistance genetic determinants, measured via a multiplex polymerase chain reaction panel, also failed to demonstrate a significant difference between FMT and sham interventions, although both groups demonstrated a reduction in detected antimicrobial resistance genes. These instructive and valuable negative results recall long-standing challenges to studying FMT. The first challenge is the substantial heterogeneity across different FMT product composition and protocols for FMT administration. R-CDI trials consistently showed FMT to be an effective intervention despite a wide variety of products, with additional variation across routes of administration and recipient microbiota conditioning with pre-FMT antibiotics and laxatives. MDRO decolonization may require microbes or metabolites present in only a subset of FMT products, and the variation observed across FMT outcomes in early MDRO decolonization trials may reflect the need to tailor microbiome-based therapeutics to target specific MDROs.

Another major challenge of FMT clinical trials is FMT recipient heterogeneity and the lack of qualified biomarkers to identify patients who may most benefit. While R-CDI is thought to necessarily be a manifestation of dysbiosis, Narang et al⁶ provided additional evidence that the microbial ecology of MDRO colonization is more complex. The low gut microbial community diversity observed in almost all patients with R-CDI was not observed in participants with MDRO colonization in this trial. Heterogeneity in colonization resistance at the time of FMT may affect FMT outcomes; in previously reported trials, MDRO clearance was increased in patients with lower pre-FMT gut microbial community diversity.

Host gut function may also interact with FMT and influence its efficacy. Narang and colleagues⁶ included patients with pancreatitis, decompensated cirrhosis, cholangitis, inflammatory bowel disease, and other conditions, each with distinct gut physiology and typical antibiotic exposures. Heterogeneity of hepatobiliary function and heterogeneity of exposure to antibiotics may also have contributed to the absence of an effect observed from FMT on the primary outcome, clearance of MDRO at 4 weeks postprocedure. While this diverse population reflects clinical gastroenterology practice, it may have diluted the potential for benefit for patients with more impaired gut microbiota colonization resistance.

The trial by Narang et al⁶ adds to a growing body of literature that extends FMT beyond its success for preventing R-CDI into the broader battle against MDROs. The trial’s negative results are valuable in light of the study’s rigor, which points toward the need to explore alternative hypotheses. Future microbiota trials should consider enriching for patients with evidence of microbiota disruption in addition to MDRO colonization. Treating colonization without dysbiosis may be asking FMT to solve a problem it was never able to address.