Gut microbiota changes associated with Clostridioides difficile infection and its various treatment strategies

ABSTRACT

Human gut microbiota are critical to both the development of and recovery from Clostridioides difficile infection (CDI). Antibiotics are the mainstay of CDI treatment, yet inherently cause further imbalances in the gut microbiota, termed dysbiosis, complicating recovery. A variety of microbiota-based therapeutic approaches are in use or in development to limit disease- and treatment-associated dysbiosis and improve rates of sustained cure. These include the recently FDA-approved fecal microbiota, live-jslm (formerly RBX2660) and fecal microbiota spores, live-brpk (formerly SER−109), which represent a new class of live biotherapeutic products (LBPs), traditional fecal microbiota transplantation (FMT), and ultra-narrow-spectrum antibiotics. Here, we aim to review the microbiome changes associated with CDI as well as a variety of microbiota-based treatment approaches.

KEYWORDS:

• Microbiome
• microbiota
• Clostridium difficile
• Clostridioides difficile
• antibiotic-associated dysbiosis
• live biotherapeutic product
• Firmicutes
• metronidazole, vancomycin, fidaxomicin

Introduction

Clostridioides difficile causes infectious colitis and is the most common organism causing healthcare-associated infections (HAI) in the United States.¹ Healthcare-associated C. difficile infection (CDI) represents 15% of all HAI and is equally prevalent in the community.^1,2 Furthermore, CDI recurrence occurs in up to 25% of patients treated with standard-of-care antibiotics, representing a disproportionately large disease burden compared to other acute infectious diseases.^3–5

The primary risk factor for CDI is exposure to antibiotics, which disrupt the normal composition and function of the gut microbiota.^6–9 This microbiota imbalance, or dysbiosis, allows for C. difficile to colonize and proliferate in the intestine to cause disease. C. difficile exists in two forms: a dormant, antibiotic-resistant spore, and a metabolically active, vegetative cell. As C. difficile spores are hardy, ubiquitous, and able to survive in the presence of oxygen, they are considered the primary driver of transmission, which occurs via the fecal-oral route.^10–12 However, disease development is dependent on the ability of a spore to colonize the intestine, germinate into a vegetative cell, and produce one or more toxins, including toxin A (TcdA), toxin B (TcdB), and more rarely binary toxin (CDT).^12–14 These toxins initiate a cascade of cellular events resulting in fluid secretion, inflammation, and tissue damage to produce the symptoms of CDI, such as diarrhea and abdominal pain (Figure 1).^14,15

Figure 1. CDI pathogenesis.

Created with BioRender.com.

Disruptions in the gut microbiota are critical to CDI development while the restoration of homeostatic bacterial diversity and abundance is essential for recovery.¹⁶ Interest in the microbiota and the microbiome, the latter of which includes the microorganisms and their collection of genomes and gene products, has therefore increased in recent decades as a promising area for CDI therapeutics.¹⁷ The increased availability of targeted sequencing methods, most commonly 16S ribosomal RNA (rRNA) gene sequencing for bacteria, has allowed for progress in our understanding of CDI pathogenesis.^18,19 This knowledge is being leveraged to develop narrow-spectrum antibiotics and more refined microbiome-based approaches to manage CDI.^20,21 One of these approaches encompasses a new class of microbiota-based, Food and Drug Administration (FDA)-approved drugs, called live biotherapeutic products (LBPs). The first LBP indicated for the prevention of CDI recurrence was approved by the FDA in November 2022, and a second was approved in April 2023.^22,23 Here, we aim to summarize the benefits and potential risks of these approaches through a discussion of the microbiome changes associated with active CDI and the therapeutics used for treatment.

Gut microbiota in healthy adults and in those with CDI

A primer: microbiome science in humans

A basic appreciation of the composition of gut microbiota in healthy humans is essential to understanding the causes and impacts of dysbiosis. The healthy gut microbiota encompasses an incredibly diverse and complex community of bacteria (Table 1), which are commonly characterized using alpha- or beta-diversity indices.^24,25 Alpha-diversity, or the diversity of microbes within samples, is expressed as the number and abundance distribution of distinct types of organisms. In contrast, beta-diversity is used to measure the diversity between samples and can be applied to compare diversity between individuals at a single time point or in one individual over time. Despite these standardized metrics, microbiome science remains a rapidly evolving and unstandardized field. For example, measuring and comparing results across studies is challenged by heterogeneous and progressing technologies, analysis pipelines and corresponding taxonomic level reported, and taxonomic reclassifications, amongst others.^26,27 Regardless, some patterns have emerged from the literature among two populations: healthy adults and those diagnosed with CDI.

Table 1. Notable taxonomic classification and names of the four most dominant bacterial phyla in the human gut.

Phylum (2021 ICSP Nomenclature)	Class	Order	Family	Genus
Actinobacteria (Actinomycetota)	Actinobacteria	Actinomycetales	Corynebacteriaceae	Corynebacterium
		Bifidobacteriales	Bifidobacteriaceae	Bifidobacterium
	Coriobacteriia	Coriobacteriales	Atopobiaceae	Atopobium
			Coriobacteriaceae	Collinsella
Bacteroidetes (Bacteroidota)	Bacteroidia	Bacteroidales	Bacteroidaceae	Bacteroides
			Porphyromonadaceae	Porphyromonas
			Prevotellaceae	Prevotella
Firmicutes (Bacillota)	Bacilli	Bacillales	Staphylococcaceae	Staphylococcus
		Lactobacillales	Enterococcaceae	Enterococcus
			Lactobacillaceae	Lactobacillus
			Streptococcaceae	Streptococcus
	Clostridia	Eubacteriales	Clostridiaceae	Clostridium
			Eubacteriaceae	Eubacterium
			Lachnospiraceae	Blautia
				Roseburia
			Ruminococcaceae	Faecalibacterium
				Oscillospira
				Ruminococcus
			Peptospreptococcaceae	Clostridioides difficile
	Erysipelotrichia	Erysipelotrichales	Erysipelotrichaceae	Faecalicoccus
	Negativicutes	Veillonellales	Veillonellaceae	Veillonella
Proteobacteria (Pseudomonadota)	Epsilonproteobacteria	Campylobacterales	Helicobacteraceae	Helicobacter
	Gammaproteobacteria	Enterobacteriales	Enterobacteriaceae	Escherichia
				Shigella

Abbreviations: ICSP, International Committee on Systematics of Prokaryotes.

Gut microbiota composition in healthy adults

Microbiota composition is diverse across the global population based on age, diet, environment, and the presence or absence of comorbidities.^28–30 However, the composition considered “healthy” for the majority of CDI-related trials is largely derived from studies of socioeconomically developed societies in Europe and North America.³⁰ Two of these, the Metagenomics of the Human Intestinal Tract (MetaHIT) project and the Human Microbiome Project, analyzed fecal samples from adults in Europe and the United States, respectively, and both observed species from the Bacteroidetes and Firmicutes phyla to be of highest abundance.^24,25 Although estimates vary widely by study, these two phyla are commonly estimated to compromise ~ 90% of total gut microbiota, with around two-thirds of microbiota belonging to the Firmicutes and 20–25% belonging to the Bacteroidetes.^24,31,32 The remainder is composed of less common phyla, namely Actinobacteria, Fusobacteria, and Proteobacteria, with proportional estimates differing depending on which methods were used for quantification. The gut microbiome is also unique to each human. Each individual harbors thousands of bacterial species, but far fewer (~160 bacterial species) represent a core human microbiome with even fewer (57 bacterial species) conserved across >90% of individuals.²⁴ The Human Gut Microbiome project also included a second fecal sample from 131 individuals, obtained a mean of 219 days after first sampling.²⁵ Among these, beta-diversity was more similar between repeated samples from the same person when compared to the beta-diversity between individuals. These observations suggest both alpha- and beta-diversity significantly differ between individuals, even in the absence of CDI, yet the Firmicutes and Bacteroidetes phyla universally predominate.

Gut microbiota composition in patients with CDI prior to CDI-directed antibiotic therapy

Gut microbiota dysbiosis is a prerequisite to CDI and thus, the microbiota composition of patients with CDI differs from that of healthy patients.^33–41 The variation in microbiota diversity is large in patients with CDI, as it is in non-diseased individuals. As antibiotics are the primary cause of dysbiosis in many of these patients, patterns of dysbiosis are likely reflective of differences in the effect of specific antibiotics on host microbiota.⁴² Furthermore, patients with primary CDI often have a lower degree of dysbiosis when compared to patients with recurrent CDI, who appear more dysbiotic.^34,43,44 Specifically, those with recurrent disease appear to have a lower relative abundance of Firmicutes and Bacteroidetes as well as a lower alpha-diversity (i.e., species richness) when compared to primary CDI cases.

Despite these nuances, all patients with CDI generally have a significant loss of microbiota diversity, characterized mainly by a lower relative abundance of species belonging to the Bacteroidetes and Firmicutes phyla, including species from the Bacteroidaceae, Clostridiaceae, Eubacteriaceae, Lachnospiraceae, Prevotellaceae, and Ruminococcaceae families.^{33,35–38,45,46} This often occurs concomitantly with an expansion of species belonging to the Proteobacteria phyla, and the Enterobacteriaceae family in particular.^39–41 These changes are present in patients prior to CDI-directed antibiotic therapy, yet further dysbiosis occurs as a result of antibiotic treatment.

CDI antibiotic therapies

CDI antibiotic treatment overview

Although antibiotics are a necessity in CDI treatment to eradicate C. difficile bacteria, no antibiotics are C. difficile-specific, and each causes some further degree of dysbiosis or collateral damage to the host microbiota. For decades, two relatively broad-spectrum antibiotics, metronidazole and vancomycin, served as the standards-of-care for the treatment of CDI.^47–50 However, both agents are associated with unacceptably high rates of CDI recurrence, likely due to their effect on the existing gut microbiota during treatment.^3,4,51–54 Fidaxomicin, which was FDA-approved in 2011, is more narrow-spectrum and has significantly lower rates of recurrence (4.0–19.5%) compared to vancomycin (16.8–26.9%).^3,4,53,54 Although rates of initial clinical cure are similar to those seen with vancomycin, the majority of current clinical practice guidelines recommend fidaxomicin over vancomycin based on this benefit.^55,56 Regardless of agent, the duration of therapy is recommended to last 10–14 days, with contemporary clinical trials utilizing a 10-day course of therapy.^55,56

Effect of standard-of-care CDI antibiotic treatment on gut microbiota

Investigations into the effect of non-CDI antibiotics on the gut microbiota have been published periodically since the 1980s; however, data regarding CDI antibiotic therapy (i.e., vancomycin and fidaxomicin) are relatively sparse and have largely focused on vancomycin.^{39–41,57–62} Much of these data are derived from CDI clinical trials in which vancomycin was used as the standard-of-care comparator.^{39,40,59–61} Notably, all but two of these studies use oral vancomycin dosed as 125 mg four times daily for 10 days.^57,58 These studies generally suggest that oral vancomycin decreases the relative abundance of species from the Actinobacteria phylum, including Bifidobacteriaceae and Choriobacteriaceae families, from the Bacteroidetes phylum, including Bacteroidaceae and Prevotellaceae families, and certain species from the Firmicutes phylum, including Clostridiaceae, Eubacteriaceae, Lachnospiraceae, and Ruminococcaceae families.^{39,40,57,58,60,61} Increases in the relative abundance of other species from the Firmicutes phylum, such as Lactobacillaceae, are also observed, as is a notable expansion of the Proteobacteria phylum. One study found that oral vancomycin significantly reduced the total bacterial biomass in patients with CDI from a median of 5.74 × 10⁹ copies per gram of stool to a median of 1.81 × 10⁶ copies per gram of stool, which corresponds to a median decrease of 0.53 log₁₀ copies per gram of stool.⁴¹

Although there are fewer studies reporting the effects of fidaxomicin on the gut microbiota, several patterns have emerged.^39,40,61 First, the decreases in the relative abundance of Bacteroidaceae, Bifidobacteriaceae, Clostridiaceae, and Prevotellaceae observed in vancomycin-treated patients are not seen or are less pronounced in fidaxomicin-treated patients.^39,40,61 In addition, there does not appear to be any increases in the relative abundance of species from the Proteobacteria phylum. To our knowledge, there are no data describing fidaxomicin’s effect on total bacterial biomass. Together, these data suggest that fidaxomicin causes relatively less dysbiosis when compared to vancomycin, which may contribute to the lower rates of CDI recurrence observed in fidaxomicin-treated patients in clinical trials.^3,4,53,54

Fecal Microbiota Transplantation (FMT)

Efficacy and safety

Although various versions of fecal microbiota transplantation (FMT) have been performed around the world for decades, its use only recently became more routine and widely available with the emergence of supportive infrastructure, including stool banks.⁶³ FMT, defined as the transfer of stool from a healthy donor into the gastrointestinal tract of a patient, is recommended by clinical practice guidelines for patients with multiply recurrent CDI.^55,56,64 These recommendations are based on several clinical trials, which suggest that FMT is both safe and effective when performed in patients with CDI.^65,66 In general, treatment success occurs in between 80% and 90% of patients, with higher rates of treatment success being achieved following a second or subsequent FMT.⁶⁵ Notably, the rates of treatment success reported in randomized controlled trials (68%) are significantly lower than the rates reported in single-arm open-label trials (83%),⁶⁵ suggesting that a selection bias and/or ascertainment bias may have contributed to an overestimation of FMT success. Although adverse effects happen in approximately one-quarter of FMT patients, these adverse effects are generally mild and gastrointestinal in nature.⁶⁶

Screening and standardization of FMT donor stool is unstandardized and has been heterogeneous over time. Safety concerns have therefore been noted surrounding the potential transmission of pathogenic bacteria, which presents a risk for infection, especially among those who are immunocompromised.^67–69 Case reports documenting these worst-case scenarios highlight the importance of rigorous sample screening procedures.⁶⁷ Unfortunately, the risk of pathogen transmission from donor to recipient has only become more relevant as highlighted by the COVID−19 pandemic and Mpox epidemic.⁶⁸

Microbiota changes and engraftment

Despite there being a large body of literature, including several RCTs, describing the clinical successes with FMT in various populations with CDI, relatively few have documented the gut microbiota changes in these patients post-FMT.^65,70 The majority of these studies are single-center case series that analyze small groups of patients.^70–78 When taken together, these studies confirm that alpha-diversity among pre-FMT patients is low, and suggest that the engraftment of donor species occurs in most cases.^70–78 Notably, one study suggests that engraftment efficacy differs by species.⁷⁸ Unsurprisingly, pre-FMT samples have a lower relative abundance of species belonging to the Bacteroidetes and Firmicutes phyla, including species from the Bacteroidaceae, Clostridiaceae, Eubacteriaceae, Lachnospiraceae, Prevotellaceae, and Ruminococcaceae families, and an increases in the relative abundance of the Proteobacteria phylum.^70–78 Furthermore, the gut microbiota in post-FMT samples resembles that of a healthy person: species from the Bacteroidaceae, Clostridiaceae, Eubacteriaceae, Lachnospiraceae, Prevotellaceae, and Ruminococcaceae families are restored, while species from the Proteobacteria phylum are diminished. These observations from human gut microbiota studies before and after-FMT have shed light on engraftment kinetics and have served as the basis for the development of LBPs.

Live Biotherapeutic Products (LBPs)

Efficacy and safety

Although the emergence of LBPs, a new class of microbiota-based drugs, was marked by the FDA-approval of fecal microbiota, live-jslm (FMBL-jslm; brand name Rebyota) in November 2022, a second agent, fecal microbiota spores, live-brpk (FMSL-brpk; brand name Vowst), was approved soon after in April 2023.^22,23 Several others, such as VE303 and MET−2, remain in development for CDI.^17,79,80 LBPs are defined as non-vaccine, biological products that contain live organisms and are applicable to the prevention, treatment, or cure of a disease or condition in human beings.⁸¹ In the context of CDI, they are intended to reduce disease recurrence following antibiotic treatment in individuals with one or more recurrent episodes. Although compositional differences between historical microbiota-based therapies, such as FMT, and LBPs are varied, the formal creation of LBPs as a drug class that is regulated by the FDA represents a shift in the framework for how these therapies will be viewed and regulated in the future, regardless of whether they were initially derived from donor stool. FMT usually refers to the administration of minimally processed donor stool under an investigational new drug application and has not been associated with any pathway to an FDA-approved indication.⁸² With the creation of the LBP class, a new approval pathway has been established for microbiota-based therapies that requires product standardization, production using good manufacturing practices, and rigorous indication-seeking clinical trials meeting FDA standards and demonstrating both safety and efficacy.⁸¹

While these heightened requirements differentiate LBPs from other microbiota-based approaches to CDI management, such as probiotics and FMT, the rationale behind their use remains the same. LBPs do not exert antimicrobial activity against C. difficile and thus do not expedite time to or increase overall rates of initial clinical cure. Instead, they are intended to restore a healthy balance of microbiota, thereby decreasing the likelihood of recurrence. Randomized controlled trials of LBPs have therefore focused on a primary efficacy endpoint of 8-week treatment success, generally defined by an absence of CDI recurrence.^83–85 Phase III trials have exclusively included patients with at least one prior CDI episode and have demonstrated the efficacy of LBPs: observed rates of 8-week treatment success ranged between 60% and 88% following LBP receipt compared to 43–60% in placebo groups.^83,84 Furthermore, given the potential for long-term changes associated with gut microbiota alterations, safety and efficacy have uniquely been assessed for as long as 24 months post dose.⁸⁶ Results from extended follow-up times have echoed the success seen at 8 weeks: 79% of FMSL-brpk recipients had not developed CDI recurrence by 6 months post dose (vs. 53% of placebo recipients; p < .001), while 64% of FMBL-jslm recipients remained CDI-free at 24 months post dose in a phase II trial.^86,87 Importantly, no safety concerns have been noted in any LBP trials and most adverse effects have been mild, self-limiting, and gastrointestinal in nature.^{83,84,86–90}

Microbiota changes and engraftment

LBPs may assist in the restoration of a healthy microbiome by either transiently restoring beneficial bacteria, thereby allowing the host to equilibrate back to their baseline through diet and lifestyle, or by direct engraftment of product-supplied microorganisms.^75,91 Although both FDA-approved LBPs are obtained from donor-derived stool, they have unique compositions and approaches to aiding in microbiota restoration. FMBL-jslm is a whole-stool product that contains a variety of beneficial microbiota mimicking that of a healthy host. Following donor and product screening for pathogenic organisms, product batches are checked for minimal total and live counts of pre-identified bacterial taxa to ensure standardization.⁹² In contrast, FMSL-brpk contains purified spores from roughly 50 species from the Firmicutes phylum.^83,90,91 Following similar donor and product screening procedures, FMSL-brpk donor stool is then processed via an ethanol-based inactivation step to kill non-spore organisms, and the spore concentration is measured to ensure standardization.⁹⁰ Regardless of mechanism, LBPs have almost universally demonstrated beneficial shifts in microbiota compositions.

In general, all patients who experience treatment success, whether they received a LBP or placebo, have improvements in microbiota diversity. However, those receiving LBPs often experience larger and more rapid shifts that trend toward the profile specifically provided by each product.^{83,86,90,93–95} As previously described, patients with CDI generally have reductions in the relative abundance of species from the Firmicutes and Bacteroidetes phyla and an increase in the relative abundance of bacteria from the Proteobacteria phylum.^{33,35–41,45,46,70,74,76} Accordingly, the majority of patients that do not experience CDI recurrence have beneficial changes in the composition of these taxa.⁷⁵ However, the microbiome of patients receiving LBPs appears different from those of placebo recipients, even when comparing subgroups of patients experiencing treatment success.⁹³ This is characterized by the increased presence of product-specific strains in addition to higher mean relative abundances and higher overdispersion.^83,93 Notably, increases in alpha- and beta-diversity observed with FMBL-jslm have persisted for up to 24 months post-treatment.^86,93 Although these long-term follow-up data are largely absent from the FMT literature, they are likely to become the standard for LBPs, which is promising.

Furthermore, engraftment magnitude and timeliness of LBP-provided material has been associated with treatment success.^83,90,95 Although microbiota shifts persist for up to 60 days post-dose, engraftment has been shown to occur within the first 7 days following product administration for both FMBL-jslm and FMSL-brbk.^83,86,90 Engraftment kinetics at 1 week post-dose appear to be of particular importance for reducing recurrence and engraftment magnitude appears dose-dependent.⁹⁰ Critically, the failure of FMSL-brpk in one of its phase II trials was attributed to underdosing of the product, thereby not permitting the level of engraftment associated with treatment success through 8 weeks.⁹⁰ These data highlight the importance of dose-finding studies and engraftment studies in the development of LBPs.

Another benefit observed following LBP administration is an overall reduction in the abundance of antibiotic-resistance genes (ARGs) and antibiotic-resistant organisms (AROs).^94–96 A novel transplantation index, used to quantify microbiome changes following FMBL-jslm receipt, demonstrated that FMBL-jslm more efficiently decolonized patients from AROs than placebo.⁹⁵ A separate analysis from FMBL-jslm phase II trial data demonstrated vancomycin-resistant Enterococcus (VRE) stool clearance in 72.7% (8/11) of patients who were colonized with VRE at the time of FMBL-jslm delivery.⁹⁶ However, as donor-derived products, FMBL-jslm and FMSL-brpk are both likely to contain ARGs and AROs that may not be present in recipients. Those with known pathogenic potential are screened for and removed during product processing; however, eradication of all ARGs and AROs is impossible. Indeed, a small number of ARGs originating from FMBL-jslm have been identified in patients following product receipt.⁹⁵ Long-term follow-up in patients who received LBPs and surveillance for these ARGs will be necessary to monitor for this potential consequence.

Bile acid profiling

A mechanistic hypothesis for the treatment success observed with LBPs may include the restoration of bile acid homeostasis following microbiome recovery, which has been measured in conjunction with several LBP clinical trials.^83,90,97 Although individual bile acids serve differing roles in CDI pathogenesis and there are exceptions to the rule, bile acids are often broadly grouped into primary versus secondary bile acids due to their roles in promoting or inhibiting C. difficile germination, respectively. C. difficile spore germination is induced by several primary bile acids.^11,98–101 A healthy, functioning gut microbiota metabolizes these primary bile acids into secondary bile acids, which generally inhibit C. difficile spore germination and outgrowth of vegetative cells to protect against disease development and/or recurrence. Therefore, a high ratio of secondary to primary bile acids is considered protective.^{98,99,101,102} Bile acid profiling in patients receiving LBPs demonstrated healthier bile acid compositions in LBP recipients versus placebo, including a decreased abundance of primary bile acids and an increased abundance of secondary bile acids.^83,97 These shifts appear to occur in parallel with product engraftment and positive correlations between the number of product-supplied species and the abundance of secondary bile acids have been noted.^83,90 It will be important for other LBPs in development to demonstrate functional changes using bile acid profiling or other targeted metabolomics to ensure the product-supplied species elicit their intended effects.

Conclusion

CDI is a microbiota-mediated disease. Dysbiosis, and frequently antibiotic-associated dysbiosis specifically, is a precursor to infection, and its persistence often leads to disease recurrence. Until recently, C. difficile-directed antibiotic treatment was the only management strategy supported by randomized controlled trials and recommended by clinical practice guidelines.¹⁰³ Although several new antibiotics and treatment strategies, including monoclonal antibodies, were developed in the past 15 years, none have sought to rectify the underlying host dysbiosis.^20,104,105 Instead, these antibiotics further contribute to host dysbiosis to varying degrees: fidaxomicin is more narrow-spectrum and preserves more of the remaining host microbiota, a benefit reflected in the lower rates of disease recurrence observed with fidaxomicin.^3,4,40

Despite a lack of FDA-approved products, a variety of microbiota-based approaches, including probiotics and FMT, have historically been adopted into clinical practice. These have demonstrated mixed success and have faced challenges related to product standardization and accompanying safety concerns.^67,106–108 LBPs represent a new class of CDI therapeutics that overcome many of these issues by undergoing various processing and purification steps and completing rigorous clinical trials.^81,109 Two products, fecal microbiota, live-jslm (formerly RBX2660) and fecal microbiota spores, live-brpk (formerly SER−109) are now FDA-approved.^83,84 Both have demonstrated efficacy in preventing CDI recurrence in adults following the completion of antibiotic therapy. The introduction of this class marks the beginning of a new paradigm in CDI management and offers promise for the continued development of microbiota-based CDI therapies.

Disclosure statement

TJC reports no competing interests to declare. AJGL reports research grant support paid to the University of Houston from Seres Health and consulting fees from Ferring Pharmaceuticals. KWG reports research grant support paid to the University of Houston from Acurx, Paratek, Summit Pharmaceuticals, and Seres Health and consulting fees from Ferring Pharmaceuticals.

Additional information

Funding

This manuscript was funded in part by Grant [P01AI152999] from the National Institute for Allergy and Infectious Diseases.