Gastrointestinal microbiome and Helicobacter pylori: Eradicate, leave it as it is, or take a personalized benefit–risk approach?

Abstract

Helicobacter pylori (H. pylori) is generally regarded as a human pathogen and a class 1 carcinogen, etiologically related to gastric and duodenal ulcers, gastric cancer, and mucosa-associated lymphoid tissue lymphoma. However, H. pylori can also be regarded as a commensal symbiont. Unlike other pathogenic/ opportunistic bacteria, H. pylori colonization in infancy is facilitated by T helper type 2 immunity and leads to the development of immune tolerance. Fucosylated gastric mucin glycans, which are an important part of the innate and adaptive immune system, mediate the adhesion of H. pylori to the surface of the gastric epithelium, contributing to successful colonization. H. pylori may have beneficial effects on the host by regulating gastrointestinal (GI) microbiota and protecting against some allergic and autoimmune disorders and inflammatory bowel disease. The potential protective role against inflammatory bowel disease may be related to both modulation of the gut microbiota and the immunomodulatory properties of H. pylori. The inverse association between H. pylori and some potentially proinflammatory and/or procarcinogenic bacteria may suggest it regulates the GI microbiota. Eradication of H. pylori can cause various adverse effects and alter the GI microbiota, leading to short-term or long-term dysbiosis. Overall, studies have shown that gastric Actinobacteria decrease after H. pylori eradication, Proteobacteria increase during short-term follow-up and then return to baseline levels, and Enterobacteriaceae and Enterococcus increase in the short-term and interim follow-up. Various gastric mucosal bacteria (Actinomyces, Granulicatella, Parvimonas, Peptostreptococcus, Prevotella, Rothia, Streptococcus, Rhodococcus, and Lactobacillus) may contribute to precancerous gastric lesions and cancer itself after H. pylori eradication. H. pylori eradication can also lead to dysbiosis of the gut microbiota, with increased Proteobacteria and decreased Bacteroidetes and Actinobacteria. The increase in gut Proteobacteria may contribute to adverse effects during and after eradication. The decrease in Actinobacteria, which are pivotal in the maintenance of gut homeostasis, can persist for > 6 mo after H. pylori eradication. Furthermore, H. pylori eradication can alter the metabolism of gastric and intestinal bacteria. Given the available data, eradication cannot be an unconditional recommendation in every case of H. pylori infection, and the decision to eradicate H. pylori should be based on an assessment of the benefit–risk ratio for the individual patient. Thus, the current guidelines based on the unconditional “test-and-treat” strategy should be revised. The most cautious and careful approach should be taken in elderly patients with multiple eradication failures since repeated eradication can cause antibiotic-associated diarrhea, including severe Clostridioides difficile-associated diarrhea and colitis and antibiotic-associated hemorrhagic colitis due to Klebsiella oxytoca. Furthermore, since eradication therapy with antibiotics and proton pump inhibitors can lead to serious adverse effects and/or dysbiosis of the GI microbiota, supplementation of probiotics, prebiotics, and microbial metabolites (e.g., butyrate + inulin) should be considered to decrease the negative effects of eradication.

Key Words: Helicobacter pylori; Eradication; Gastrointestinal microbiota; Dysbiosis; Fucosylated glycan; Inflammatory bowel disease

Core Tip: Helicobacter pylori (H. pylori) is generally regarded as a human pathogen, but it can act as a commensal symbiont. H. pylori colonization may have beneficial effects on the host by regulating gastrointestinal microbiota and protecting against some allergic and autoimmune disorders and inflammatory bowel disease. H. pylori eradication can cause various adverse effects and alter the gastrointestinal microbiota, leading to dysbiosis. Therefore, eradication cannot be an unconditional recommendation in every case of H. pylori infection, and the therapeutic decision should be based on a personalized assessment of the benefit vs risk.

TO THE EDITOR

We read with great interest the article by Niu et al[1], which showed that the effectiveness of quadruple Helicobacter pylori (H. pylori) eradication therapy containing bismuth depended on the gastric microbiota, and a high rate of H. pylori eradication was associated with the presence of Rhodococcus, Lactobacillus, and Sphingomonas, which were significantly enriched in the gastric mucosa in the successful eradication group[1]. The role of lactobacilli, mainly beneficial bacteria, in H. pylori infection, including eradication, has been well studied[2]. However, the authors showed for the first time the importance of Rhodococcus and Sphingomonas, which are more likely to be opportunistic or pathobiont species with unclear functions in the human gastrointestinal (GI) tract[3,4], in successful eradication of H. pylori[1]. It is noteworthy that in gastric cancer (GC), when the abundance of H. pylori decreased, several taxa (including Rhodococcus and Lactobacillus, discussed by Niu et al[1]) in the gastric mucosa significantly increased, which may indicate their potential involvement in GC after H. pylori infection[5]. In addition, H. pylori was negatively correlated with some opportunistic bacteria/pathobionts such as Haemophilus, Streptococcus, Neisseria, and Fusobacterium in the success group[1]. The results obtained by Niu et al[1] may suggest that H. pylori competes not so much with beneficial bacteria as with pathobionts, and eradication may ultimately worsen the gastric microbiota.

Indeed, not only does the composition of the gastric microbiota affect H. pylori eradication, but eradication significantly affects the microbiota of both the stomach and intestine, which can lead to marked and long-term dysbiotic changes. Dysbiosis of the microbiota after H. pylori eradication can be caused by many factors: by the action of antibiotics and proton pump inhibitors; by the loss of H. pylori leading to changes in the immune response of the GI mucosa; and by changes in the microenvironment of the GI tract, including those in microbial metabolic pathways and changes in gastric acidity associated with both pharmacotherapy and loss of H. pylori[6,7]. Features of the dysbiotic changes, their duration, and the rate of restoration of the disturbed microbiota vary greatly in different studies.

Diverse effects of H. pylori eradication on gastric and intestinal microbiota

Recent studies have demonstrated not only short-term but also long-term (≥ 6 mo) changes in the gastric microbiota after H. pylori eradication. In about half of cases (52.3%), eradication led to the predominance of proinflammatory Acinetobacter in gastric corpus mucosa with a decrease in microbial diversity in patients with endoscopic follow-up for > 1 year[8]. An earlier study showed that Acinetobacter was enriched in patients with persistent gastric inflammation 1 year after H. pylori eradication[9]. Moreover, some bacteria in the gastric mucosa (Actinomyces, Granulicatella, Parvimonas, Peptostreptococcus, Prevotella, Rothia, and Streptococcus), which are predominantly of oral origin, were associated with precancerous gastric lesions (atrophy and/or intestinal metaplasia) 1 year after H. pylori eradication[9]. Actinomyces, whose abundance can increase in the absence of H. pylori, might significantly increase the risk of GC[10]. Thus, some studies demonstrated a contribution of various gastric bacteria to precancerous gastric lesions after H. pylori eradication[9]. In general, Actinobacteria decreased after H. pylori eradication, Proteobacteria increased during short-term follow-up and then returned to baseline levels, and Enterobacteriaceae and Enterococcus increased in the short-term and interim follow-up[11]. Alternatively, it has been shown that in regions with high GC risk, H. pylori is one of the main factors in gastric dysbiosis and successful eradication can lead to the restoration of gastric microbiota[12].

H. pylori eradication also affects the gut microbiota. Bismuth quadruple therapy leads to short-term dysbiosis of the gut microbiota with an increased abundance of Proteobacteria and decreased abundances of Bacteroidetes and Actinobacteria. The increase in gut Proteobacteria may contribute to adverse effects during eradication therapy[13]. In another study, H. pylori eradication was associated with significant alterations in the gut microbiota that did not completely recover 6 wk after treatment[7]. In general, there was a decrease in Actinobacteria, which are pivotal in the maintenance of gut homeostasis, compared with baseline throughout the follow-up (> 6 mo) after eradication[11]. Furthermore, eradication therapy alters microbial functional pathways and the metabolism of gastric and gut bacteria[9,14].

Conversely, other studies showed that successful H. pylori eradication exerts beneficial effects on gut microbiota, including increased probiotic Bifidobacterium and downregulation of drug-resistance mechanisms[12]. Liou et al[15] generally confirmed the long-term safety of H. pylori eradication therapy but reported incomplete restoration of microbial diversity after 1 year and clinically irrelevant but significant increases in body mass index (BMI) and body weight at that time.

Interestingly, an increase in body weight/body mass index after H. pylori eradication had been identified earlier[16]. Suggested mechanisms of this effect range from an improvement in the symptoms of postprandial dyspepsia[16] to changes in the regulation of leptin and ghrelin[17] mediated by antibiotic-associated changes in the microbiota (especially by the imbalance between bacterial producers of lactate and acetate)[18]. In general, however, the data in various studies are contradictory and indicate weight gain, weight loss, or the absence of an effect of H. pylori eradication on body weight; this may be due to differences in the characteristics of the studied populations, such as age, nosology, and composition of the GI microbiota[19]. Further in-depth study of the microbiome-mediated effects of H. pylori and eradication therapy on human host metabolism, including nutrient uptake, energy homeostasis, bodyweight, hormone secretion, lipid profile, and glucose homeostasis/glycemic control, will provide clinically important findings for the management of H. pylori infection.

H. pylori status and the human gut microbiome

The presence or absence of gastric H. pylori can significantly affect the gut microbiota. For example, Nitrospirae were found exclusively in H. pylori-negative patients. The role of this phylum, containing nitrite-oxidizing bacteria, in the human microbiome is unclear. In a study by Wang et al[20], Nitrospirae were found in the gastric mucosa in all patients with GC but not in patients with chronic gastritis. The authors suggested that these bacteria may be involved in carcinogenesis through enhanced production of N-nitroso compounds. A recent study demonstrated a possible pathogenetic link between enriched colonic Nitrospirae and drug-resistant epilepsy, implying that Nitrospirae can increase nitrite toxicity and cause blood-brain barrier dysfunction[21].

Proinflammatory Bacteroides ovatus and Fusobacterium varium, associated with ulcerative colitis and adenomatous polyps[22,23] as well as trimethylamine-producing Clostridium sp. AT5[24] were enriched in H. pylori-negative samples, while Bacteroides plebeius, characteristic of the healthy groups (vs patients with adenomatous polyps)[23] and butyrate-producing Eubacterium ramulus were enriched in H. pylori-positive samples[7].

Conversely, Butyricimonas spp., including Butyricimonas virosa, associated with bacteremia in patients with GI cancer (colon and duodenal adenocarcinomas) and diverticulitis[25] as well as Bacteroides coprophilus, specifically enriched in ankylosing spondylitis[26], were enriched in H. pylori-positive individuals[7]. Proinflammatory Prevotella copri, associated with rheumatoid arthritis and microscopic colitis as well as Enterobacter cloacae and Klebsiella pneumoniae, pathogens commonly associated with hospital infections, were also enriched in H. pylori-positive individuals[27]. Moreover, gut microbial vitamin B₁₂ biosynthesis was significantly lower in H. pylori-positive individuals compared with H. pylori-negative individuals[27]. Dash et al[28] showed that the gut microbiota of H. pylori-infected individuals was characterized by a significantly increased abundance of Succinivibrio, Coriobacteriaceae, Enterococcaceae and Rikenellaceae as well as Candida glabrata and other unclassified fungi. The authors suggested a possible role for these H. pylori-associated changes in the gut microbiota in intestinal barrier disruption and development of colorectal carcinoma[28].

Potential protective and regulatory role of commensal H. pylori

Currently, H. pylori is generally regarded as a human pathogen and a class 1 carcinogen[29], responsible for 15% of the total cancer burden globally and up to 89% of all GC cases[30]. H. pylori is etiologically related to gastric and duodenal ulcers and mucosa-associated lymphoid tissue lymphoma. According to current guidelines, it is almost always subject to unconditional eradication based on the “test-and-treat” strategy[31]. However, although H. pylori is present in > 50% of the world’s population, sequelae of infection occur in only 20% of infected individuals, and malignant complications, such as GC, occur in < 3% of infected people[7]. Therefore, there is also an alternative point of view that H. pylori is a commensal symbiont[32,33]. Back in 1998, Blaser[32] wrote that, “H. pylori can thus be regarded as indigenous or ‘normal’ flora, which most humans acquire within the first few years of childhood and then carry for life.”

Unlike other pathogenic/opportunistic bacteria, H. pylori colonization of newborns/infants is facilitated by T helper type 2 immunity and leads to the development of immune tolerance[34]. Most likely, the co-evolution of H. pylori and the human host over millennia has led to the fact that this bacterium is considered a commensal symbiont, not a pathogen, by the host’s immune system. This is indirectly confirmed by the fact that α1,2-fucosylated glycans of the GI epithelium, which are an important part of the innate and adaptive immune system (they create a symbiotic environment for the host and microbiota and protect against pathogens), mediate the adhesion of H. pylori to the surface of the gastric epithelium, contributing to successful colonization[35]. As a result, early colonization of H. pylori can have a positive effect on the host, for example on the regulation of the hormones leptin and ghrelin as well as on protection against some allergic (e.g., asthma) and autoimmune diseases and against inflammatory bowel disease (IBD)[36,34]. The inverse association between H. pylori and some potentially proinflammatory and/or procarcinogenic bacteria found in various studies[20,22-24] may suggest a regulatory function of H. pylori toward the GI microbiota. The available evidence for beneficial effects of H. pylori toward the GI microbiota, as well as potential protective effects against certain diseases, should not be ignored by the gastroenterological community.

Presence of H. pylori infection and reduced risk of IBD: Is there a causal relationship? The potential protective role of H. pylori in the development of IBD, shown in some studies[37], may be related to both modulation of the gut microbiota and the immunomodulatory properties of H. pylori. An inverse association between H. pylori and potentially proinflammatory microbes (Bacteroides ovatus, Fusobacterium varium, Rhodococcus, Sphingomonas) supports a microbiome-modulating mechanism, while an immunomodulatory mechanism of protection against IBD may involve the activation of colonic mucus production by H. pylori via the NLRP3/caspase-1/interleukin-18 axis[38].

We suggest another possibility for the association between H. pylori infection and IBD, a non-causal relationship, which relates to the fucosylation status of host mucin glycans in the GIT. Individuals with a non-functional α(1,2)-fucosyltransferase 2 (FUT2) gene (they are termed non-secretors), who have loss-of-function mutations, cannot express α(1,2)-fucosylated glycans in the GI mucosa. Non-secretors (about 20% of the population) are more susceptible to infection by some pathogens (Escherichia coli, Neisseria meningitidis, Streptococcus pneumoniae, Haemophilus influenzae, Candida albicans) and have aberrant gut microbiota, with a reduction of beneficial Bifidobacterium spp. The FUT2 non-secretor phenotype increases susceptibility to Crohn’s disease, ulcerative colitis, primary sclerosing cholangitis, celiac disease, psoriasis, Behçet’s disease, type 1 diabetes, and so on but at the same time protects against H. pylori, which requires fucosylated glycans in the gastric mucosa for adhesion[35,39,40].

A recent study showed that patients with ulcerative colitis and Crohn’s disease had decreased FUT2 expression and α1,2-fucosylation in the colon[41]. In addition, Fut2 deficiency in the intestinal epithelium exacerbated colitis in epithelium-specific Fut2 knockout (Fut2^△IEC) mice, promoted the release of proinflammatory cytokines, and aggravated epithelial barrier damage[41]. The authors demonstrated for the first time that epithelium-specific Fut2 deficiency increased susceptibility to IBD through modulation of the gut microbiome and microbiota-mediated lysophosphatidylcholine generation. Lysophosphatidylcholine may have deleterious effects on the colon by promoting the release of proinflammatory cytokines, damaging the tight junctions and epithelial barrier in the colon epithelium, and exacerbating colonic inflammation in Fut2^ΔIEC mice[41].

In turn, upregulation of FUT2-mediated fucosylation in the intestinal epithelium, for example by exogenous L-fucose, can protect the intestinal barrier, enhance tight junctions, and alleviate intestinal inflammation[42]. Similar to the colon, increased expression of the fucosyltransferase genes FUT2 and FUT1 in the gastric epithelium promotes H. pylori adhesion and ultimately infection[43]. Thus, we suggest that increased expression of FUT2 in the gastrointestinal mucosa can simultaneously mediate both H. pylori infection and protection against IBD via an H. pylori-independent mechanism. In this case, a causal relationship between the presence of H. pylori and a reduced risk of IBD is unlikely. However, it cannot be ruled out that H. pylori mediates the protective effect of fucosylation against IBD by modulation of the gut microbiota or through an immunomodulatory mechanism[38].

Concluding remarks

Given the available data, eradication cannot be an unconditional recommendation in every case of H. pylori infection, as in the vast majority of people H. pylori is most likely a commensal[32,33], possibly beneficial in mutualistic interaction with the host (for example protecting against some allergic and autoimmune diseases[36]). We join the opinion of Chen et al[7] that the decision to eradicate H. pylori should be based on an assessment of the benefit–risk ratio for the individual patient. We also support the position of Miller and Williams[44] that “universal eradication” of H. pylori may cause more harm than good for the infected persons. Thus, the current guidelines based on the unconditional “test-and-treat” strategy[31] should be revised, including to reduce the excessive number of indications for eradication and to avoid empirical eradication therapy without a previous diagnostic test for H. pylori infection.

It may be worth recommending unconditional eradication only in patients with concomitant mucosa-associated lymphoid tissue lymphoma[45] and/or in individuals at high risk of GC, for example in groups of high familial (hereditary) risk[46] or in high-risk areas/populations where eradication effectively reduces the risk of GC[47]. In the latter case, the advisability of such an approach is unquestionable, if it is evidence-based. For example, a recent systematic review and meta-analysis provided moderate-certainty evidence that searching for and eradicating H. pylori can reduce the subsequent incidence of GC and death from GC in healthy asymptomatic infected people; the risk of GC decreased by 46% after eradication therapy[48]. However, the authors concluded that as all but one of the eligible trials were conducted in East Asian populations (in China, Japan, or South Korea), and the only trial conducted in a non-Asian population (in Colombia) did not demonstrate any benefit of such an approach, the results of the systematic review cannot be extrapolated to populations outside East Asia[48].

A well-known paradox, the low incidence of GC in some regions of Africa, Asia (e.g., in India), and Latin America with a high prevalence of H. pylori infection, also requires in-depth study; this is called the African[49] or Asian/Indian enigma[50]. Although the existence of this phenomenon is sometimes disputed[51], studies have shown that H. pylori alone is most likely not enough for the development of GC, even with a high prevalence of highly pathogenic strains. Therefore, it is necessary to take into account not only the virulence factors of H. pylori and the oncogenic potential of specific strains of H. pylori but also the genetics and ethnicity of the human host population, their dietary habits (including antioxidant and sodium levels), smoking, alcohol consumption, socioeconomic status, and coinfection (parasitoses/helminthiases) modulating the potentially protective T helper type 2 immune response[49,50].

An important factor influencing the serious consequences of H. pylori infection appears to be the co-evolution of H. pylori and the human host. A recent study demonstrated that the African human ancestry showed clear signs of co-evolution with H. pylori, while the European ancestry was maladapted. The Asian ancestry was intermediate but closer to the African ancestry[52]. This supports the hypothesis that H. pylori is a commensal symbiont rather than a human pathogen. Hopefully, a series of international prevalence surveys to investigate age-specific prevalence of H. pylori in areas of low and high GC risk, namely ENIGMA, recently launched under the auspices of the International Agency for Research on Cancer in Africa (Uganda), Asia (Iran), and Latin America (Chile, Costa Rica), will shed light on the regional characteristics of H. pylori infection and identify markers for GC risk stratification to offer reasonable preventive interventions for different populations[53].

In addition, H. pylori eradication is likely to be recommended in patients with cancer who are on therapy with immune checkpoint inhibitors or vaccine-based immunotherapy, for example in patients with non-small-cell lung cancer[54].

In patients with diseases negatively associated with H. pylori, such as IBD, microscopic colitis, celiac disease, asthma, multiple sclerosis, Barrett’s esophagus, esophageal adenocarcinoma, eosinophilic esophagitis, and so on, eradication should be carried out with caution, carefully weighing the risk-to-benefit in each case. Even though H. pylori eradication did not affect either the healing rate or the recurrence rate of pre-existing gastroesophageal reflux disease, the possibility of developing new erosive gastroesophageal reflux disease after eradication should always be kept in mind[55].

The most cautious and careful approach should be taken in elderly patients with multiple eradication failures since repeated eradication (second-/third-line therapies) can cause antibiotic-associated diarrhea, including severe Clostridioides difficile-associated diarrhea and colitis[56] and antibiotic-associated hemorrhagic colitis due to Klebsiella oxytoca[57]. In this regard, we support the recent conclusion of the American Gastroenterological Association experts that, “after multiple failed eradication attempts, the potential benefits of H. pylori eradication should be weighed carefully against the likelihood of adverse effects and inconvenience of repeated high-dose acid suppression and antibiotic exposure, particularly among individuals who are not at an identifiably higher risk of complications from persistent H. pylori infection (e.g., GC, peptic ulcer disease); in such scenarios, a shared decision-making approach should be seriously considered, especially in the elderly, those with frailty, and those with intolerance to antibiotics” (Best Practice Advice #9)[30].

Furthermore, since eradication therapy with antibiotics and proton pump inhibitors can lead to serious adverse effects and/or long-term dysbiotic changes in the GI microbiota, the supplementation of probiotics[58-61], prebiotics, and microbial metabolites (e.g., butyrate + inulin)[62] to reduce the negative effects of eradication should be considered[7]. In addition, alternative eradication regimens with limited or no antibiotic use, for example phage-based regimens[63], autoprobiotics[64], and natural agents and methods including traditional Chinese medicine[65], should be proposed, developed, and explored in future studies.

ACKNOWLEDGEMENTS

The authors are grateful to Dr. James Allen for editing a draft of this manuscript.