The gastrointestinal microbiome, small bowel bacterial overgrowth, and microbiome modulators in cystic fibrosis
Abstract
People with cystic fibrosis (pwCF) have an altered gastrointestinal microbiome. These individuals also demonstrate propensity toward developing small intestinal bacterial overgrowth (SIBO). The dysbiosis present has intestinal and extraintestinal implications, including potential links with the higher rates of gastrointestinal malignancies described in CF. Given these implications, there is growing interest in therapeutic options for microbiome modulation. Alternative therapies, including probiotics and prebiotics, and current CF transmembrane conductance regulator gene modulators are promising interventions for ameliorating gut microbiome dysfunction in pwCF. This article will characterize and discuss the current state of knowledge and expert opinions on gut dysbiosis and SIBO in the context of CF, before reviewing the current evidence supporting gut microbial modulating therapies in CF.
1 INTRODUCTION
Cystic fibrosis (CF) is a heritable disease caused by mutations in the gene encoding the CF transmembrane conductance regulator (CFTR) protein.1 CFTR is expressed throughout the pancreatic duct, and hepatobiliary and gastrointestinal (GI) tracts, where dysfunction contributes to the altered intestinal environment and GI symptoms that represent a significant source of morbidity for people with CF (pwCF). CFTR is expressed on the apical membrane of enterocytes where it is responsible for regulation of ion and water content, via secretion of chloride and bicarbonate across the intestinal epithelium. CFTR dysfunction results in reduced bicarbonate secretion and low intestinal pH as well as dehydrated and inspissated mucus.2 Below we explore the impact of CFTR dysfunction on the intestinal microbial environment and potential therapy options.
2 GASTROINTESTINAL MICROBIOME IN CF
A state of intestinal dysbiosis (alterations in the microorganism composition of the gut) is well described in pwCF. Various factors, some specific to CF, are thought to contribute to an altered intestinal milieu and resultant dysbiosis.3 These include CFTR dysfunction itself;4, 5 acidic and dehydrated intestinal mucus;6, 7 increased luminal viscosity, decreased intestinal transit;8-10 lack of endogenous pancreatic enzymes and resultant fat malabsorption;11, 12 increased intestinal permeability; and impaired innate immunity.13 In addition to CFTR-related mechanisms, iatrogenic and lifestyle factors also contribute. These include a high fat/high calorie diet;12, 14 early, repeated, and/or chronic antibiotic exposure; acid suppression; and, in the case of organ transplant recipients, immunosuppressive medications.3, 15-18
The GI microbiome refers to the collective genomes of microorganisms (bacteria, viruses, protozoa, and fungi) in the gut. The majority of bacteria that colonize the GI tract reside in the colon. However, smaller numbers of microorganisms also reside in the stomach and small intestine. The fecal microbiome is commonly used as a surrogate, albeit imperfect, for the colonic microbiome. Dysbiosis, or alterations in microbiome homeostasis, is a recognized feature of various disease processes and is thought to be a major contributor to intestinal inflammation in CF.16, 19 In the healthy state, the GI microbiome serves important functions including maintenance of gut-epithelial barrier function; resistance against colonization by enteric pathogens; synthesis of essential vitamins and amino acids; regulation of fat metabolism; immune system modulation; and breakdown of otherwise indigestible complex carbohydrates into short chain fatty acids (SCFAs).19 SCFAs are themselves important for GI epithelial health by serving as an energy source for enterocytes, ameliorating intestinal inflammation, modulating the production of several inflammatory cytokines, reinforcing the epithelial defense barrier and regulating intestinal motility.3, 12, 19-25 Inflammation and the altered intestinal environment described in CF and other inflammatory conditions will in turn exert selective pressure on intestinal microbes as organisms with the ability to perform more efficiently in such conditions have a growth advantage, further contributing to dysbiosis.12, 16, 26
A key feature of the CF gut microbiome is decreased species diversity (reduced richness and abundance) as well as delayed maturation and altered functionality.11, 27-34 Decreased microbial diversity is well described in many inflammatory, metabolic, autoimmune, and systemic diseases.3, 30, 35 In addition to decreased diversity, there is altered composition of the GI microbiome in CF compared to healthy controls.12, 17, 28, 30, 32, 36, 37 A major feature is that CF patients have fewer SCFA-producing bacteria (e.g., Prevotella, Parabacteroides, Alistipes) and an increased abundance of bacteria associated with inflammation and colorectal cancer (CRC) (e.g., Enterococcus, Clostridium, Ruminococcus) when compared to non-CF controls.34 An increased relative abundance of Firmicutes and reduced Bacteroidetes, as well as an increase in γ-Proteobacteria concentration is consistently reported in CF cohorts.29, 34, 38, 39 It is possible that dysbiosis is most extreme in early life and there is a shift toward more typical abundances as children with CF age. A longitudinal analysis of 39 children with CF from birth through age 4 years demonstrated that several taxa which have been associated with dysbiosis in CF change with age toward a more “healthy-like” composition (becoming less dysbiotic). One notable exception is that Bacteroides, which may play an important role in immune programming, are reduced in pwCF and do not significantly increase with age.18 Proinflammatory and CRC-related microbiota reported in increased abundance in CF include Enterobacteriaceae, Streptococcus, Veillonella, and Escherichia coli.3, 29, 36, 40 Furthermore, the relative abundance of E. coli is not only increased in the CF gut but E. coli isolated from fecal samples of children with CF demonstrated an increased growth rate in the presence of glycerol, likely a result of selection pressure from increased intestinal glycerol phospholipids from dietary fat.26 Antiinflammatory organisms, such as Faecalibacterium prausnitzii, which are known producers of SCFAs (butyrate), are reduced.16, 27, 32, 41 Bacteria of the genus Bifidobacterium demonstrated decreased abundance in CF. Bifidobacterium have been shown to inhibit pathogenic bacteria, improve intestinal barrier function, and suppress proinflammatory cytokines.27, 42 Additional metagenomic findings, likely reflecting selective pressures of the CF gut, include an increase in the species Enterococcus faecalis and Enterococcus faecium, frequently known to exhibit antibiotic resistance; as well as increased relative abundance of Ruminococcus spp. (e.g., R. gnavus). These Ruminococcus spp. are mucin degraders and thus their presence may reflect preference for increased mucus accumulation in the CF intestine.34, 41, 43 Another mucin-degrading and antiinflammatory genera noted to be reduced in the stool of children with CF is Akkermansia, which further decreases with age compared to healthy controls.3, 18, 44, 45 McKay et al. demonstrated that Akkermansia negatively correlated with stool calprotectin (a marker of intestinal inflammation) and positively correlated with FEV1% predicted.45 Overall, from a functional perspective, the CF intestinal microbiome demonstrates an increased capacity to metabolize nutrients, antioxidants, and SCFAs, and a decreased propensity to synthesize fatty acids.12, 30, 33
Gut dysbiosis has implications for various manifestations of CF. Dysbiosis is thought to contribute to the increased inflammatory state and impaired barrier function in the CF intestine.3 As described above, individuals with CF have increased abundance of proinflammatory pathogens (Staphylococcus, Streptococcus, Escherichia, Shigella, Enterobacter, Morganella, and Veillonella dispar) and decreased antiinflammatory microbes (Baceteroides, Bifidobacterium adolescentis, and F. prausnitzii).3, 4, 26, 30, 36, 46 The link between dysbiosis and inflammation may be largely due to reductions in SCFAs, which have important implications for intestinal health by maintaining epithelial integrity, regulating intestinal pH and ameliorating intestinal inflammation.15 SCFA-producing organisms, as well as levels of butyrate, propionate, and acetate (the main SCFAs produced by gut bacteria), are decreased in children with CF compared to controls.30, 32, 47 In addition, the CF microbiome is enriched for genes responsible for SCFA catabolism and there has been a demonstrated positive correlation with these genes and fecal calprotectin.11, 30, 46 Antibiotic usage, high-fat diet, and prolonged intestinal transit, all of which are factors in CF, are also associated with a reduction in SCFA levels.15, 48-50 Conversely, the inflamed enteric environment selects for organisms capable of anaerobic respiration which are better adapted for such conditions. These include members of the Enterobacteriaceae family that are noted to be increased in inflammatory bowel disease and CF.15
Intestinal dysbiosis and inflammation are also thought to contribute to CF patients developing CRC more often and earlier than the general population. There is a 5–10 times increased prevalence of CRC in the CF population, and it occurs two to three decades earlier.51, 52 The risk of gastrointestinal cancer is additionally increased (between two and five times higher) in pwCF who have undergone a lung transplant compared to nontransplanted individuals.52 In a large systematic review of studies involving adult CF patients and non-CF controls, CF patients were found to have a higher abundance of bacteria associated with CRC as well as decreased SCFA-producing bacteria.34, 39, 47 Fusobacterium spp., widely associated with CRC, are notably more abundant in both stool and colonic mucosal samples in pwCF.3, 30, 34, 39, 53, 54 Other organisms which have been implicated in CRC tumorigenesis and found to be elevated in CF include Bacteroides fragilis, E. faecalis, Clostridium hathewayi, Clostridium symbiosum, Flavonifractor plautii, and Dialister spp.34 Dayama et al. demonstrated associations between gut dysbiosis, such as reduction in butyrate-producing organisms and abundance of CRC-associated bacteria, with host CRC-related gene expression (LCN 2 and DUOX2) and enrichment of oncogenic pathways.39 However, most of these studies are not adequately powered to separate out the effects of other likely contributors to dysbiosis and inflammation, such as diet, antibiotic use, CFTR mutation, and immunosuppression, on CRC risk.
In addition, GI dysbiosis in CF has implications outside of the intestinal tract and may contribute to CF-related liver and lung diseases. CF liver disease (CFLD) represents the fourth leading cause of death in pwCF.55 Several forms of CFLD, cirrhotic and noncirrhotic, are now recognized. The pathophysiology behind CFLD is not well established. CFTR is expressed on bile duct epithelium and thus a prevalent theory, particularly for the cirrhotic form of CFLD, is that dysfunctional CFTR results in inspissated bile and plugging of small bile ducts, leading to intrahepatic accumulation of toxic bile acids with increased free radicals and subsequent hepatocyte injury and indirect periportal fibrosis as hepatic stellate cells are activated to produce collagen.56, 57 However, as all patients with the same CFTR mutation do not have the same rates of CFLD, other genetic or environmental factors must modify the risk of developing advanced CFLD in pwCF. One proposed model involves the gut–liver axis, whereby intestinal inflammation, bacterial dysbiosis, and increased small intestinal permeability result in bacterial translocation into the portal circulation, activation of hepatic inflammatory pathways, and resulting fibrogenesis and portal fibrosis.58, 59 Alterations in the intestinal microbiome have been described in other liver pathologies, such as nonalcoholic steatohepatitis, primary sclerosing cholangitis, cholestasis, and cirrhosis.59-65 In a murine study involving CFTR knockout mice, the authors demonstrated that in the absence of CFTR, which results in increased intestinal permeability, diet-induced dysbiosis (from a high medium chain triglyceride diet) works in conjunction with an ill-adapted immune response to promote cholangiopathy.66 A pilot study comparing CF patients with and without cirrhosis demonstrated a trend toward higher relative abundances of Firmicutes (Clostridium) and lower levels of Bacteroidetes in CF patients with cirrhosis compared to those without liver disease. Furthermore, fecal microbiome alterations appeared to correlate with macroscopic, advanced intestinal lesions (ulcerations) found on video capsule endoscopy.58 These studies lend support to the hypothesis that intestinal pathology in pwCF, and associated dysbiosis, may influence the development of advanced cirrhotic CFLD.
The gut–lung axis generally refers to the ability of the gut microbiota to influence the course or outcome of underlying lung disease and vice versa.67 Both lung and gut dysbiosis start early in the CF population, and likely mutually impact one another.18, 68-70 The mechanism through which intestinal microbiota affects lung health is not fully elucidated but might include immunological cross-talk between the similarly structured mucosal environments of these two organs as well as direct transmission of microbes between the lung and intestinal tract.67 A longitudinal study involving infants and young children with CF demonstrated that certain lung pathogens colonize the gut before colonization of the respiratory tract and there was also a concordance between bacteria that were increasing or decreasing over time in both compartments.69 In addition, studies have described a relationship between infant breastfeeding and respiratory microbial diversity.69, 70 Multiple studies have demonstrated associations between gut microbiota and respiratory clinical measures in CF. Hoen et al. demonstrated that gut microbiota composition during the first 6 months of life differs significantly between infants with CF who experience a respiratory exacerbation versus those who do not. This study also suggested a correlation between gut alpha diversity and time to initial respiratory exacerbation.70 Intestinal microbiota alpha diversity is also reduced in CF patients with severe lung dysfunction and is positively correlated with time to initial exacerbation in infants, and higher ppFEV1 in adults.29, 70 In terms of compositional associations, alterations in gut microbiota may predict or precede airway events. In a cohort of infants and young children with CF, Parabacteroides was significantly decreased in intestinal samples before initial colonization with Pseudomonas aeruginosa.70 Other studies have demonstrated associations between specific intestinal genera and ppFEV1 in children and adults.29, 30, 45 The interplay between gut microbiome and respiratory health is intriguing and in need of further investigation. Furthermore, the neonatal period might afford a critical window for intervention as it is a time of rapid acquisition of gut microbiomes before stabilizing between 3 and 5 years of age.67, 71
In addition to colonic dysbiosis, pwCF are also thought to be susceptible to developing small intestinal bacterial overgrowth (SIBO).72 From a clinical perspective, SIBO is a clinical entity that is poorly defined and lacks diagnostic consistency and precision. A more recent definition, albeit nonspecific, refers to the “presence of excessive numbers of bacteria in the small bowel, causing gastrointestinal symptoms.”73, 74 The propensity to SIBO in pwCF, similar to dysbiosis in the large intestine, has been attributed to impaired, or altered, motility and accumulation of thickened viscous mucous that acts as an anchor for bacteria and impairs bacterial defenses provided by intestinal Paneth cells.72, 75 Estimates of the prevalence of SIBO in CF depend somewhat on the diagnostic criteria used with some studies indicating a frequency of up to 40%.72, 76 Symptoms typically associated with SIBO include bloating, flatulence, abdominal discomfort, and diarrhea, which are nonspecific and can also be explained by other GI issues seen in CF. Steatorrhea can be seen in more severe cases of SIBO, which can confuse the picture in pwCF and exocrine pancreatic insufficiency.73 In more advanced cases, or if left untreated, SIBO can result in malabsorption of essential nutrients. Deconjugation of bile acids and consequent depletion of the bile acid pool can result in fat malabsorption and fat-soluble vitamin deficiency. Vitamin B12 deficiency can result from consumption of cobalamin by anaerobes, competitive binding with cobalamin from bacterially generated metabolites of cobalamin at the ileal receptor, and, in severe cases, mucosal injury involving the binding site. Bacterial synthesis of folic acid can lead to high folate levels and thus the somewhat unusual combination of high folate and low vitamin B12 levels should raise suspicion for SIBO.73 Diagnosis of SIBO in CF presents some unique challenges. One diagnostic tool involves small intestinal aspiration with endoscopy. This is invasive and requires sedation or, in pediatrics, general anesthesia. Furthermore, in CF, bacteria tend to reside in a poorly soluble and sticky mucus layer that strongly adheres to the intestinal wall, which likely affects yield from luminal aspiration and may require intestinal biopsies to assess bacterial load.75 Obtaining a clean uncontaminated sample can also be difficult as oral flora may be picked up as the scope passes through the mouth. To complicate, the challenges and difficulties in sampling the small intestine have also meant there is paucity of understanding and correlation between SIBO and evaluation of the small intestinal microbiome using next-generation nonculture microbiology techniques. The use of breath testing in the diagnosis of SIBO lacks universal acceptance and has never been fully validated.73 Breath testing measures exhaled gases produced by bacterial fermentation of an ingested substrate (typically lactulose or glucose). However, it should be noted that metabolism of lactulose may reflect bacterial load in the colon rather than the small intestine. Breath testing should measure methane in addition to hydrogen and it has been demonstrated that pwCF produce methane more often in SIBO as compared to non-CF controls, possibly due to increased mucins in the CF intestine.77 Other considerations in CF are that the frequent use of antibiotics may predispose to colonization with predominantly methane-producing bacteria. In addition, the association between SIBO and delayed intestinal transit may make interpretation of breath testing difficult. Finally, in CF patients with advanced lung disease, breath tests may be of limited value due to gas retention in mucus plugged airways.72 Given the above limitations, empiric treatment and observation for resolution of symptoms is a widely practiced approach to diagnosis and management of SIBO. Common antibiotic regimens for treatment of SIBO include metronidazole, ciprofloxacin, or rifaximin with a 7–10-day course, improving symptoms for up to several months in 46%–90% of patients.73 However, evidence supporting selection or duration of antibiotics for empiric treatment is sparse at best. As such, empiric treatments should be discontinued if no clear benefit is observed, and alternative diagnoses considered.
3 MICROBIOME MODULATION
In light of the aforementioned issues and importance of the gut microbiome in CF, the capacity to modulate the gut microbiome, with intentions to improve gut health, has gained greater traction in recent years.78, 79 Administration of a single or combination commensal strains of bacteria (probiotics), indigestible carbohydrates which promote proliferation of commensals (prebiotics), inanimate microorganisms and/or their components that confer a health benefit to the host (postbiotics), or a combination of all three (synbiotics) have the potential to produce significant changes in the microbiome conducive to the improvement of gut health.16 CFTR modulation may also reduce gastrointestinal inflammation and correct the dysfunctional microbiome characteristic of CF.80, 81 Given the marked gut dysbiosis4, 28, 30 and intestinal inflammation30, 82, 83 established as a hallmark of the CF gut, these therapies present exciting opportunities to ameliorate the adverse gut changes that occur in pwCF.
3.1 Microbiome modulation and probiotics
Probiotics describe a range of live organisms that, in adequate amounts, confer a health benefit to the host.84 Probiotics are proposed to confer potential beneficial effects within the gastrointestinal system through several mechanisms, including interacting directly with the microbiome and immune system, and facilitating the production of organic acids and small molecules.16, 85 Existing probiotics are largely based on bacterial strains from genus Lactobacillus, Bifidobacterium, and other lactic-acid-producing bacteria,86, 87 while further development of improved probiotics derived from established “beneficial” species such as Akkermansia,88 Eubacterium, Propionibacterium, Faecalibacterium, and Roseburia is ongoing.86, 88
Probiotics have previously demonstrated varying efficacy in other diseases with links to GI dysbiosis. For instance, a meta-analysis into the influence of probiotic supplementation for Parkinson's-disease-related gut dysbiosis demonstrated improved frequency of bowel movement and stool consistency.89 Further, the reduction of necrotizing enterocolitis incidence90, 91 has also been observed with probiotics supplementation. However, consensus recommendations from the American Gastroenterological Association on the role of probiotics in management of gastrointestinal disorders have highlighted significant knowledge gaps and heterogeneity between studies, and variability in the probiotic strains studied.92
Systematic reviews have synthesized growing evidence for the use of probiotics in pwCF.92-95
The most comprehensive review identified 17 clinical trials to date and was based off data from the 12 identified randomized controlled trials (RCTs),96 five of which studied or noted a gastrointestinal alteration due to probiotic intervention. Of these, trials either utilized a single Lactobacillus strain L. rhamnosus GG (LGG) or L. reuteri,93, 97-99 or a multistrain formulation with fructooligosaccharides100 at dosages ranging from 108 to 1011 CFU/day.
In the meta-analysis by Coffey et al., significant reductions in fecal calprotectin (MD −47.41 μg/g, 95% CI −93.28 to −1.54) (p = .04) over a period of 6 months were reported. However the authors acknowledged several limitations and suggested interpretation of their findings with caution.96 Selective reporting, incomplete outcome data, and poor generalizability increased the chance of biases influencing the meta-analysis, while concerns also exist regarding the external validity of probiotics, due to heterogeneity in probiotic supplementation.96 Thus, while promising signs certainly exist, further studies including longitudinal trials should be performed to better understand the therapeutic potential of probiotics.
Findings from an RCT published following the Coffey et al.96 Cochrane review have provided further support for therapeutic potential of probiotics in pwCF. In children treated with LGG, Ray and colleagues99 demonstrated statistically significant differences in the distribution of the dominant bacterial genus compared to placebo patients. Within the 12-month supplementation period, the majority of LGG-treated patients transitioned to a Bifidobacterium-dominated microbiota, while placebos adopted a Bacteroides-dominated gut microbiota. The transitioned gut microbiome in LGG-treated patients, now dominated by a bacterial species widely established as beneficial to the gut, was also linked to better clinical outcomes. These patients experienced significantly lower concentrations of fecal calprotectin compared to Bacteroides-dominated gut microbiomes.
Despite the evident limitations in existing research, preliminary results from individual trials and systematic reviews demonstrate the clear potential of probiotics in reducing gastrointestinal inflammation and dysbiosis in CF. Further exploration of this field is propitious, and merits long-term RCT studies examining the impact of probiotics on the gut milieu. The specific investigation of very young children, when there is still capacity to modulate the microbiota before maturation, is also crucial to determining the ability of probiotics to modify the natural history of dysbiosis. One approved, multicenter, double-blind, randomized, placebo-controlled trial (PEARL-CF trial) aims to further characterize the changes in CF intestinal microbial profile that occur with long-term probiotics supplementation (ACTRN12616000797471).
3.2 Microbiome modulation and prebiotics/diet
There is increasing recognition that diet is a significant and direct determinant of the gut microbiome. Different components of dietary intake, such as carbohydrates, proteins, fats, and fibers, may serve as substrates for specific microbes. As these microbes break down and ferment these dietary components, they produce metabolites that can influence both the overall structure of the microbial community and the biochemical environment within the gut. SCFAs, as described earlier, are examples of potential beneficial metabolites derived from dietary fibers and resistant starch.101 In obesity, prebiotics have been reported to cause specific metabolic effects, including reducing the abundance of key biomarkers such as C-reactive protein, plasma cholesterol, triglycerides, and fasting plasma insulin.102-104 Breads, cereals, onions, and garlics naturally harbor prebiotics; however, dietary supplements such as inulin, lactulose, oligofructose, fructooligosaccharides, and galactooligosaccharides are also prebiotics with proven efficacy in improving gut health.101, 105
While studies have investigated prebiotics in CF as part of a synbiotic preparation method, only one existing study focuses on the isolated use of prebiotics in CF to target GI microbiome modulation. Since prebiotic efficacy assumes the successful uptake and utilization of the supplement by the host microbiome, Wang et al.106 utilized a combined method of metagenomic sequencing, in vitro fermentation, amplicon sequencing, and metabolomics to identify if CF adults could utilize high amylose maize starch (HAMS) as a prebiotic.106 They identified that the CF gut microbiome harbors the ability to ferment HAMS, despite low abundances of common taxa. While reduced levels of acetate remained, production of butyrate and propionate were also comparable with healthy control levels. Interestingly, while the commensal genus Faecalibacterium demonstrated strong associations with SCFA production in healthy controls, its role was instead performed by Clostridium sensu stricto 1 in the microbiota of CF patients. Therefore, taxa commonly associated with the CF gut may be able to exploit prebiotic supplementation to produce beneficial metabolites, despite reduced abundances of commonly commensal bacteria, an exciting revelation with therapeutic importance.
It is well established that the high-energy, high-fat diet critical in CF nutrition management may often be achieved through high energy, nutrient-poor foods, leading to poor gastrointestinal health.14 McKay and colleagues identified positive correlations between fecal calprotectin and takeaway food and noncore food, and negative correlations with percentage grains and wholegrains in patients with CF.45 However, while studies support associations between dietary fiber supplementation and improved gut health in both CF and healthy patients,45, 107 its influence on the progression of the CF microbiome is not yet understood.
Furthermore, the efficacy of dietary fiber is mixed in other chronic conditions which demonstrate microbiome dysfunction and increased inflammation. Low fat, high-fiber diets (LFD) have demonstrated potential in decreasing markers of inflammation and intestinal dysbiosis in ulcerative colitis. Fritsch and colleagues identified significant decreases in Actinobacteria abundance, while demonstrating increases in F. prausnitzii when comparing LFD to the standard American diet.108 However, an alternative study identified no significant clinical benefit in F. prausnitzii abundance when supplementing Crohn's disease patients.109 A meta-analysis appraising RCTs which investigated dietary fiber interventions in inflammatory bowel disease also identified some clinical potential in using germinated barley foodstuffs and fructooligosaccharides to reduce inflammation and disease activity, but evidence was extremely sparse and results regarding other investigated forms of fiber supplementation were inconclusive.110 Thus, fiber supplementation in CF gut health lacks concrete evidence of clinical benefit, and yet the evident variability in response between different gastrointestinal conditions merits future detailed, CF-specific investigations into the therapeutic potential it possesses.
3.3 CFTR modulators and the microbiome
Therapeutic modalities in CF have advanced significantly in the last decade with the availability of CFTR modulators—personalized, small molecule therapies targeting the basic defect(s) within the CFTR protein.81, 111 Whereas previous CF therapy largely involved symptomatic relief or was targeted toward treatment of complications of CF, the targeting of the CFTR protein itself has proved an effective and tolerable intervention which greatly reduces pulmonary exacerbations and improves lung function.112, 113 Excitingly, there is evidence suggesting that these therapies reverse or improve the “pathognomonic” CFTR-related defects in the gastrointestinal tract. More specifically, significant improvements in gastrointestinal pH, and the normalization of gastrointestinal histopathology with resolution of inspissated material in the intestinal glands of CF patients have been reported.114, 115
Despite the above, there is relative paucity of literature regarding the “downstream” effects of CFTR modulation on gut microbiome and inflammation. Ooi and colleagues performed a prospective observational study, collecting stool samples from 16 CF patients (eight adults and eight children) who had commenced ivacaftor.81 They identified a significant reduction in stool calprotectin following the commencement of ivacaftor (median interquartile range [IQR]: 154.4 [102.1–284.2] vs. 87.5 [19.5–190.2] mg/kg, p = .03). The reduction in calprotectin following ivacaftor treatment is hypothesized to be a result of the aforementioned physicochemical changes in the gut, with a downstream effect of a less proinflammatory microbiota community. Nevertheless, there is now evidence that intestinal inflammation could be reduced or reversed in patients via CFTR modulation.81 Investigation of the bacterial microbiome via 16S rRNA gene region amplicon sequencing also identified significant increases in genera Akkermansia, and decreases in the family Enterobacteriaceae. Interestingly, linear regression analysis revealed that these decreased relative abundances of Enterobacteriaceae were associated with reduced levels of stool calprotectin. This is empirical evidence that suggests ivacaftor treatment may have contributed to the selective reduction of pathogenic Enterobacteriaceae, and supports an associative link between CFTR-gene related gut dysbiosis and intestinal inflammation.
However, there have been conflicting findings from two separate studies. A similarly designed study by Ronan et al. of adults with the G551D CFTR mutation on ivacaftor failed to identify any significant changes in fecal calprotectin and microbial diversity.116 The rationale for this is not entirely unclear. Perhaps the exclusively adult cohort with an older median age accounted for this difference, in line with overall observation in other affected CF organs that the most notable benefits of CFTR modulation is observed when modulators are commenced at a younger age. Another study by Pope and colleagues sought to identify whether CFTR modulators ivacaftor (in a cohort of 12 pancreatic sufficient [PS] pwCF and an R117H CFTR variant) or lumacaftor/ivacaftor (in a cohort of eight pancreatic insufficient [PI] pwCF and an F508del variant) altered fecal measures of malabsorption or fecal microbiomes. Increased fecal fat content has been postulated by the authors as a driver for fecal dysbiosis in pwCF, given the high-fat, energy-dense, and nutrient-poor diet utilized in CF treatment.12, 28 Further, CF patients with exocrine PI often present with increased dietary lipids in the GI tract due to malabsorption.80 In this study, Pope et al. identified trends in the PI cohort toward decreased fecal fat content and fecal microbiota akin to those who were PS with CF. However, unlike Ooi and colleagues, no discernible differences in the fecal microbiota were identified. While the latter's study cohort was largely comprised of PI patients harboring the G551D CFTR mutation and treated with ivacaftor, the former utilized one cohort of PS patients, and another cohort of PI patients treated with lumacaftor/ivacaftor, an intervention which has demonstrated limited efficacy in restoring CFTR function among patients with the F508del mutation.117 The median followup period for the Pope study was also shorter by 2 months than for the Ooi study, which may be an additional reason for the contrasting findings.
The evident discord between studies calls for further investigation into the therapeutic capacity of CFTR modulators. Fittingly, research investigating the impact of newer highly effective CFTR modulators such as elexacaftor/tezacaftor/ivacaftor (ETI) is ongoing. Schwarzenberg and colleagues identified decreases in fecal calprotectin following 6 months of ETI supplementation, as part of a 56-center prospective, observational study.118 Furthermore, preliminary abstracts published in the Journal of Cystic Fibrosis supplemental issue have identified a reduction in gastrointestinal symptoms over a 24-week period,116 and reductions in small bowel water retention with ETI supplementation.119 This growing evidence for the influence of CFTR modulators on gastrointestinal quality-of-life measures and intestinal manifestations in CF merits further exploration into their potential impact on microbiome composition.
3.4 Probiotics, prebiotics, and CFTR modulators in SIBO
Preliminary evidence surrounding the use of gut microbiome modulators like probiotics in SIBO supports the theoretical role of these alternative therapies. However, there remains a dearth of robust clinical trials into their therapeutic use. Probiotics are at present the most widely researched alternative therapy. Systematic review by Zhong and colleagues identified higher SIBO decontamination rate, decreased H2 concentration, and the relief of abdominal pain with probiotic supplementation.120 However, they could not conclusively determine that probiotics prevented SIBO from these observations. A later 2021 systematic review into alternative therapies echoed this sentiment, citing the small size and heterogeneity of treatment duration and strains used among existing studies as a key barrier to conclusive evidence.74, 121-123 The impact of diet in SIBO treatment is also restricted to analysis of specific dietary strategies, rather than supplementation. Dietary recommendations in SIBO are extensions of irritable bowel syndrome (IBS) treatment, where the avoidance of short-chain carbohydrates which are fermentable by small intestinal bacteria, like oligosaccharides, disaccharides, monosaccharides, and polyols, is preferred.74, 122 Higher hydrogen-breath test outputs in IBS patients have been positively correlated with diets high in these short-chain carbohydrates.124, 125 While CFTR dysfunction may promote SIBO, the only existing study investigating CFTR modulation and SIBO identified insignificant changes in hydrogen breath test results following lumacaftor/ivacaftor usage in pwCF.123 The current evidence, coupled with the reservations around the validity and variability of SIBO diagnosis, calls for larger-scale RCTs to further evaluate these therapies and their efficacy in the treatment of SIBO.
4 CONCLUSION
The intestinal microbiome in pwCF is characterized by reduced species diversity, delayed maturation, and altered functionality. Proposed contributors to dysbiosis include features intrinsic to CF as well as lifestyle and iatrogenic factors. Dysbiosis likely has both intestinal and extraintestinal implications as described above and there is growing interest in therapeutic options for microbiome modulation. Alternative therapies, such as probiotics and prebiotics, and current CFTR-gene modulators have demonstrated preliminary support for a therapeutic benefit in ameliorating gut microbiome dysfunction in pwCF, yet this evidence remains limited by certainty, due to small sample sizes, and heterogeneity of study methods and host factors. Further large scale, randomized, placebo-controlled trials are necessary to evaluate the true clinical potential of these therapies across the full gamut of CF and SIBO cases.
AUTHOR CONTRIBUTIONS
Nicole Green: Writing—original draft; writing—review and editing; investigation. Christopher Chan: Writing—original draft; writing—review and editing; investigation. Chee Y. Ooi: Conceptualization; project administration; writing—review and editing; supervision.
ACKNOWLEDGMENTS
Chee Y. Ooi is funded by the National Health and Medical Research Council (NHMRC, Australia) Investigator Grant (2020/GNT1194358). Open access publishing facilitated by University of New South Wales, as part of the Wiley - University of New South Wales agreement via the Council of Australian University Librarians.
CONFLICT OF INTEREST STATEMENT
Chee Y. Ooi has been a speaker, consultant, and on advisory boards for Vertex Pharmaceuticals. The remaining authors declare no conflict of interest.
Open Research
DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article as no new data were created or analyzed in this study.
