Faecal microbiota transplant for chronic norovirus infection: a case report of donor microbiota engraftment without clinical success

Chronic norovirus (NoV) in immunocompromised individuals is associated with substantial morbidity and mortality, yet effective treatment options are limited [1]. Emerging evidence suggests that the intestinal microbiome plays a role in NoV pathogenesis and susceptibility to infection [2–4]. Given its proven efficacy in treating recurrent Clostridioides difficile infection, faecal microbiota transplant (FMT) is being evaluated for an expanding range of applications, including chronic NoV, with studies demonstrating safety in immunocompromised individuals [5]. However, outcomes have been variable. A 2017 case series reported no benefit from FMT for chronic NoV in a 32-year-old female with common variable immunodeficiency and a 65-year-old female with chronic lymphocytic leukaemia (CLL) [6]. Conversely, the first documented successful use of FMT for chronic NoV was reported in 2020 in a 68-year-old female organ-transplant recipient, who experienced complete symptom resolution and viral clearance following colonoscopic FMT [7]. More recently, a retrospective case series reported four additional successful cases of FMT for chronic NoV in immunocompromised patients, including a patient who achieved symptomatic response despite persistent viral shedding [8]. Here, we describe the case of an immunocompromised patient with debilitating chronic NoV infection who, despite successful donor microbiota engraftment after FMT, failed to achieve symptomatic improvement or viral clearance. A 72-year-old female presented with chronic gastrointestinal distress and progressive malnutrition. Her medical history included a matched related donor allogeneic stem-cell transplant in 2007 for CLL, with subsequent Richter’s transformation to diffuse large B-cell lymphoma. She underwent chimeric antigen receptor T-cell therapy in 2020 for aggressive relapse. From mid-2019 onwards, she experienced frequent nausea and diarrhoea, accompanied by anorexia, progressive weight loss, serum micronutrient deficiencies, and functional decline. Diarrhoea often exceeded 12 episodes per day and her weight reached a nadir of 31 kg. Extensive gastrointestinal workup was unrevealing except for persistently positive NoV PCR from March 2019 to July 2022. Intravenous immunoglobulin, rifaximin, nitazoxanide, and a multitude of medication, diet, supplement, and probiotic interventions were largely unsuccessful in reducing symptom burden or reversing malnutrition. Given the lack of viable treatment options, our multidisciplinary team recommended compassionate use of FMT. Single-patient expanded-access approval was obtained from the US Food and Drug Administration, along with Institutional Review Board approval from the University of Maryland Baltimore and informed written consent from the patient. In July 2022, FMT was performed via colonoscopy using 250 mL of previously frozen, unrelated donor stool that had been screened for pathogens (OpenBiome; Cambridge, MA). Patient stool samples were collected at several time points prior to and up to 4 months following FMT; an aliquot of donor stool was reserved at the time of the procedure. All samples were stored in OMNIgene® GUT solution and 16S rRNA gene amplicon sequencing of the V3–V4 regions was performed on an Illumina NextSeq 1000 instrument at the University of Maryland School of Medicine, Maryland Genomics core facility. Following FMT, the patient’s diarrhoea, nausea, and malnutrition persisted, and serial NoV tests remained positive, consistently with clinical treatment failure. The composition of the microbiota in stool samples collected before and after FMT was compared with the donor sample (timeline in Fig. 1A). The most frequently observed bacterial taxa across all samples were Akkermansia muciniphila, Faecalibacterium prausnitzii, Bacteroides uniformis, and Bacteroides vulgatus, together accounting for a median of 31.2% (range 17.7%–48.5%) of relative abundance. Late post-FMT samples showed a compositional profile resembling that of the donor, suggesting donor microbiome engraftment (Fig. 1A). Shifts in faecal microbiota composition and diversity following FMT. (A) Stacked bar plot of relative abundance of the donor sample at time of faecal microbiota transplant (FMT) and multiple pre-FMT (6, 3 days), early post-FMT (1, 4, 6, 14, 30 days), and late post-FMT (52, 60, 91, 122 days) stool samples from the patient recipient, with late post-FMT patient samples more similar to those of the donor than pre- and early post-FMT samples. (B) Principal component analysis graph based on the first two principal components (PCs), with 69.8% of the variance accounted for by the first PC and 10.1% of the variance accounted for by the second PC, and all donor and patient samples from 30 days post-FMT onwards clustering closely on PC1 and other earlier samples distributed along PC1. (C) Box and whisker plots of multiple alpha diversity indices for the donor and patient at pre-FMT and post-FMT time points. Principal component analysis was used to assess differences in microbiota composition between samples across time points and revealed distinct clustering of pre-FMT and early post-FMT samples from late post-FMT and donor samples along principal component 1, which explained 56.4% of the variance (Fig. 1B). A higher abundance of A. muciniphila was inversely correlated with F. prausnitzii abundance, and increasing F. prausnitzii abundance was associated with post-FMT time points. Donor samples displayed similarly high F. prausnitzii and low A. muciniphila abundance. These PCoA patterns further support donor microbiome engraftment. Alpha diversity was assessed by using Shannon, abundance-based coverage estimator (ACE), and Simpson indices (Fig. 1C). Pre-FMT samples demonstrated a significantly lower diversity than both post-FMT and donor samples by ACE, with trends towards significance observed when using Shannon and Simpson indices. Norovirus is a leading cause of acute gastroenteritis and infection is typically self-limited. In immunocompromised patients, however, viral shedding and symptoms may persist for weeks to years [9]. Prolonged symptomatic illness in this population is associated with malabsorption, weight loss, and increased mortality [1]. In the absence of an approved vaccine or definitive therapy, experimental treatments for chronic NoV have included immunomodulators such as interferon and immunoglobulin, the antiviral ribavirin, and the antiprotozoal agent nitazoxanide. Growing evidence suggests that the intestinal microbiome influences NoV susceptibility and pathogenesis. Proposed mechanisms include microbial alterations to interferon pathways or bile acids, the binding of NoV to bacterial histo-blood group antigens, and direct interaction of bacteria with noroviruses [2, 3]. However, the specific mechanisms and bacterial taxa involved in the host response to NoV remain largely unknown. In one study of 35 healthy adults, a higher abundance of Ruminococcaceae and Faecalibacterium spp. was associated with reduced susceptibility to NoV infection [3]. Similarly, in a human NoV challenge experiment, asymptomatic individuals demonstrated smaller microbiome compositional shifts between pre-infection and post-infection compared with symptomatic individuals, suggesting that microbiome resilience upon infection may be protective or, conversely, that the diarrhoea severity or duration of symptoms may drive microbiome disruption [4]. In our case, microbial composition and diversity analyses suggested successful donor microbiome engraftment after FMT and, despite a lack of adverse effects, which is in line with prior literature describing the safety of FMT in immunocompromised patients, no clinical improvement occurred [5]. Furthermore, strains of many of the predominant taxa observed pre- and post-FMT (e.g. F. prausnitzii, A. muciniphila) share similar known functional capacity as short-chain fatty acid producers, which are known to beneficially modulate mucous production and promote gut-barrier integrity [10, 11]. Possible explanations for the treatment failure may include the patient’s age, host or viral genetics, baseline microbiome composition, intestinal mucosal integrity, prior antibiotic exposure, and underlying medical conditions. Our patient’s medical history included intensive lymphodepletion therapies and poor immune reconstitution despite durable oncologic remission and the cessation of immunosuppressive therapy. The FMT was performed after >4 years of chronic NoV infection, similarly to the case reported by Lahtinen et al. in which a woman with common variable immunodeficiency and 2.5 years of chronic NoV infection also failed to respond to FMT [6]. Prolonged infection may allow viral adaptations in an already genetically diverse pathogen [12]. The FMT was delivered via colonoscopy, following the approach used in the first reported successful case by Barberio et al. [7]. However, because NoV primarily affects the small intestine, upper-tract administration may be superior, as supported by subsequent successful cases in which gastric or jejunal delivery was used [8]. A limitation of our study is that we characterized only the faecal microbiome, which may be distinct from that of the small intestine, where NoV replication predominates; additionally, we did not quantify changes in NoV levels via viral titres. The complex interplay between the host immunity, mucosal barrier, intestinal microbiome, and NoV pathogenesis remains incompletely understood. Longitudinal studies capturing microbiome dynamics throughout the course of infection and following FMT treatment could further contribute to our understanding of the therapeutic potential of microbiome-based interventions for chronic NoV. K.S. and E.v.R. conceived and designed the project, collected samples, and contributed to the manuscript. E.v.R. performed the FMT. E.v.R. and A.R. obtained regulatory approvals. M.A. and J.R. performed the DNA sequencing, analysed and interpreted the sequencing data, and contributed to the manuscript. All authors read and approved the final manuscript. This work was supported by the authors’ departmental funds. The authors would like to thank the patient for participating in this study. J.R. is co-founder of LUCA Biologics, a biotechnology company focusing on translating microbiome research into live biotherapeutics drugs for women’s health, and is a consultant for Ancilia Bio. All other authors declare that they have no competing interests.