Host and gut microbiota crosstalk: A new paradigm for colorectal cancer immunotherapy

Abstract

Although immunotherapy for colorectal cancer (CRC) has recently gained widespread attention, many patients continue to exhibit inherent or acquired resistance due to a lack of tumor-infiltrating lymphocytes and the poor immunogenicity of cancer cells within an immunosuppressive tumor microenvironment. Advances in high-throughput sequencing and bioinformatics have increasingly highlighted the role of the gut microbiome (GM) in modulating the quantity and phenotypes of innate and/or adaptive immune cells, thereby influencing CRC pathogenesis and the clinical response to immunotherapy. The GM maintains a symbiotic relationship with the host, contributes to protection against opportunistic pathogens, and supports intestinal homeostasis. Dysbiosis in GM composition and metabolite profiles drive uncontrolled inflammatory cascades, induces oxidative DNA damage, and promotes neoplastic progression. Furthermore, targeting GM populations, such as next-generation probiotics or dietary interventions, can enhance the prevalence of effector T cells in a “hot” immunogenic microenvironment, thereby ameliorating the efficacy of CRC immunotherapy, improving survival and potentially reducing toxicities in patients. This review briefly summarizes current findings and molecular mechanisms underlying host-GM mutualism under physiological and cancerous conditions, which may inform the development of novel strategies for CRC diagnosis, treatment, and prevention.

Key Words: Colorectal cancer; Gut microbiota; Immunotherapy; Immune cell; Tumor microenvironment

Core Tip: From birth onward, host health is closely linked to the delicate balance of ecological microbiota composition. The microbiota utilizes distinct enzymatic capabilities to metabolize nutrients and produce immunomodulatory byproducts that influence organ function and behavior. A strong and rapid association has been identified between gut microbiome dysbiosis, defined as a disruption in microbial load or diversity, colorectal cancer pathogenesis, and responsiveness to immunotherapy. Therefore, clarify the mechanism of gut microbiome dysbiosis and develop microbe-based intervention through health-oriented and adaptable mechanisms, such as oral probiotic administration, will become an indispensable component of colorectal cancer immunotherapy in the future.

INTRODUCTION

Colorectal cancer (CRC) is the third most prevalent malignancy and the second leading cause of cancer-related mortality worldwide[1]. The rising incidence of CRC at a younger age has been linked to Westernized dietary patterns, sedentary lifestyles, delayed medical diagnosis and treatment, and exposure to colibactin, a genotoxin produced by certain bacterial strains[2,3]. Early-life exposure to colibactin plays a significant mutagenic role, contributing to the increased incidence of CRC in individuals under 50 years of age. CRC is a highly heterogeneous malignant epithelial neoplasm of the intestine, driven by the accumulation of genetic and epigenetic alterations within an immunosuppressive tumor microenvironment (iTME)[4]. Despite substantial progress, the molecular complexity of CRC continues to result in poor clinical outcomes and frequent recurrence following treatment modalities such as surgical resection, neoadjuvant chemoradiotherapy, and immunotherapy. Consequently, there is an urgent need to elucidate the mechanisms underlying CRC’s development of therapeutic resistance, and to develop more effective preventive and therapeutic strategies aimed at improving patient survival and quality of life.

With the rapid advancement of next-generation sequencing (NGS), research on the human microbiome has emerged over the past decade as a promising interdisciplinary field bridging basic and clinical medicine[5,6]. Humans have co-evolved with diverse commensal microflora, including bacteria, fungi, viruses, protozoa, and archaea, forming unique mutualistic holobiont ecosystems within individual hosts. These microbial communities play essential roles in maintaining physiological functions such as metabolism, immune regulation, and inflammation[7-9]. From birth onward, host health is closely linked to the delicate balance of ecological microbiota composition[10]. The microbiota utilizes distinct enzymatic capabilities to metabolize nutrients and produce immunomodulatory byproducts that influence organ function and behavior. The probiotic-fermented milk drink kefir has been shown to reverse low-grade inflammation and reduce CRC susceptibility induced by neonatal overfeeding, through modulation of the gut microbiome (GM) and the mucosal immune system[11]. These findings offer new insights into GM alterations associated with CRC pathogenesis and suggest potential therapeutic avenues.

GM changes for CRC immunogenicity and therapy

The gastrointestinal tract hosts the largest and most diverse microbial community, containing more than 1000 species and approximately 10¹⁴ microorganisms, organized in a spatially structured manner[9,12]. Components of the GM exhibit distinct interindividual variability and readily adapt to dynamic environmental and host-derived cues. Although the core GM is 95% conserved across individuals and comprises fewer than 20 genera or species, its collective genetic material generates critical signals that mediate host-microbiota interactions, particularly in the development and regulation of innate and adaptive immune systems[13]. Under homeostasis, the GM prevents colonization by invasive opportunistic pathogens through the production of antagonistic metabolites or enzymes, stimulation of inflammatory cytokine secretion, and promotion of lymphocyte differentiation[14,15]. In response, the host immune system eliminates pathogenic microbes while preserving beneficial microflora to support overall physiological function. In the iTME of CRC, studies have reported a marked increase in the abundance of Fusobacterium and a notable decrease in Ruminococcus, accompanied by protumor neutrophil enrichment, disruption of mucosal immunity, and tumor metastasis[16]. Microbiota-targeted interventions aim to reprogram immune pathways, such as antigen processing and presentation (APP), T-cell receptor signaling, and T-cell activation, to enhance antitumor immune responses[17]. Therefore, it is plausible that manipulation of the GM could serve as a viable clinical adjuvant to conventional CRC immunotherapy. Such interventions may fine-tune the complex crosstalk between host immune components and microbial communities, providing cell-type-specific protection against infection and contributing to overall health improvement.

A strong and rapid association has been identified between GM dysbiosis (GMD), defined as a disruption in microbial load or diversity, CRC pathogenesis, and responsiveness to immunotherapy, due in part to the close anatomical proximity of the intestinal microbiota to the tumor sites[18-20]. The GM operates through distinct biochemical cycles, shaped by its taxonomic composition. In a pathologically hypoxic luminal environment, this ecosystem favors the overgrowth of harmful bacteria and a reduction in probiotic populations, ultimately resulting in an abnormal nutritional supply.

Several core pathogens associated with CRC, including Fusobacterium nucleatum (Fn), colibactin-producing Escherichia coli (CoPEC), and enterotoxigenic Bacteroides fragilis, increase the genotoxicity of intestinal epithelial cells (IECs) and impair barrier integrity, while concurrently inhibiting the growth of antagonistic microbes[21-23]. Following translocation through a compromised intestinal barrier, the conditional pathogen Alcaligenes faecalis has been shown to exacerbate inflammation during colitis and suppress intestinal immune surveillance, ultimately contributing to the transition from inflammation to cancer[24]. Notably, fecal microbiota transplantation (FMT) from CRC patients has been demonstrated to induce intestinal dysplasia, abnormal cellular proliferation, and colonic neoplasia in murine models[25,26]. These findings support a strong association between intestinal dysbiosis and CRC initiation and progression. Various oncogenic stimuli derived from GMD can trigger irreversible cell cycle arrest, referred to as cellular senescence, and dysregulated immune cell activation, both of which contribute to chronic inflammation and tumor development. A novel diagnostic technique for assessing individual intestinal dysbiosis has been developed, offering superior precision and sensitivity for the prevention, diagnosis, personalized immunotherapy, and clinical management of CRC[27].

As a representative approach in cancer immunotherapy, immune checkpoint inhibitors (ICIs) promote the recruitment and expansion of previously inactive or naive T-cell clones within the iTME, thereby restoring their capacity to target and eliminate cancer cells[28]. Despite the therapeutic potential of ICIs as a first-line treatment, the majority of CRC patients exhibit drug resistance due to a deficiency of tumor-infiltrating lymphocytes and the inherently low immunogenicity of cancer cells. Notably, Chinese yams have been shown to remodel GMD in CRC, thereby enhancing ICI efficacy by suppressing tumor-associated macrophage type 2 (TAM2) populations and promoting the infiltration of cytotoxic CD8+ T cells (CTLs)[29]. When the dynamic mutualism between host and microbiota is disrupted, oncomicrobial passengers gain a competitive growth advantage within poorly regulated inflammatory niches that provide a more permissive environment. These accumulated pathogens recruit tumor-infiltrating myeloid cells, enabling evasion of host immune responses. Therefore, identifying strategies to manipulate GMD to enhance antitumor activity and improve responses to ICIs is critical[30,31]. For example, following programmed death-1 (PD-1) blockade, the keystone bacterial species Akkermansia muciniphila (AKK) has been shown to increase the population of CCR9+ CXCR3+ CD4+ central memory T helper cells (Th), positively impacting progression-free survival in cancer patients[32]. There is an urgent need to understand the coordinated roles of specific gut bacteria and their metabolites in CRC immunotherapy, as revealed through both animal and human microbiome studies.

In this review, we summarize current evidence regarding the role of the GM in the occurrence, progression, and treatment of CRC. We highlight key principles of host-microbiota crosstalk that mediate immune regulation of malignant cells, with implications for the development of novel and promising clinical translation strategies.

HOST-GM MUTUALISM IN HEALTHY PEOPLE AND CRC PATIENTS

The mammalian gut functions not only as an organ for nutrient absorption but also as the largest site of peripheral immune defense. In healthy individuals, the symbiotic GM educates the host immune system on tolerance, reactivity, and defense, conferring mutual benefits that support eubiosis in exchange for an anoxic environment and a rich energy supply provided by the host[33]. AKK utilizes intestinal mucin as a unique source of carbon and nitrogen and promotes the proliferation of mucin-secreting goblet cells, potentially contributing to the maintenance of intestinal barrier integrity[34]. The GM promptly expresses antimicrobial peptides, primes innate immune cells through the recognition of microbe-associated molecular patterns (MAMPs), and initiates a pattern recognition receptor-mediated immune response to prevent the penetration, survival, and replication of pathobionts within host cells[34]. Site-specific clustering of gut fungi mediates host-protective immunity and promotes intestinal epithelial repair through CD4+ Th-dependent interleukin (IL)-22 signalling[35]. Consequently, the GM is highly sensitive to environmental factors within the host. As the second-largest gene pool in the human body, GMD is likely to contribute to increased intestinal health risks.

Moreover, the extensive diversity of microbial metabolites profoundly influences host physiology by establishing a complex network within the microbe-host interaction. Microbe-derived short-chain fatty acids (SCFAs), including acetate, propionate, formate, butyrate, lactate, and succinate, are key mediators that directly modulate intestinal barrier function through interactions with both immune and nonimmune cells. Fibrolytic microbes degrade indigestible carbohydrates in the colon, producing SCFAs through fermentation. These small molecules diffuse into immune-associated regions, promoting the generation of anti-inflammatory regulatory T cells (Tregs), a process accompanied by colonic tuft cell expansion and compartmentalized accumulation that supports the maturation and function of the mucosal epithelium[36]. When intestinal and circulating metabolites are altered by inflammatory bowel disease, protective immunity becomes compromised, leading to increased susceptibility to infection due to the dysregulation of dendritic cells (DCs), macrophages (Mφs), and CD4+ T cells[37]. In CRC patients, reduced levels of SCFAs and probiotics, including Roseburia spp., Lactobacillus spp., and Bifidobacterium spp., are commonly observed. Supplementation with SCFAs enhances the immune system’s capacity to cooperate with the efficacy of PD-1 therapy, as SCFAs may initiate programmed cell death signaling and inhibit CRC cell proliferation[38]. These findings indicate that GMD disrupts the metabolic output of gut commensals. Increased awareness of disease-associated metabolomic profiles advances the understanding of interconnected cellular signaling pathways triggered by environmental risk factors and offers promising directions for improving health outcomes and managing complex diseases.

It is well established that the disorganized outer mucus layer of the intestinal barrier provides a permanent habitat for the commensal GM. Mucins, the primary components of mucus, contain a variety of receptors that facilitate bacterial adhesion through glycan interactions. These structures selectively retain beneficial microbiota while inhibiting pathogen attachment[39]. In response to microbial trafficking challenges, the gut initiates robust immune responses at specialized microanatomical sites, such as gut-associated lymphoid tissues. Microfold cells with the gut-associated lymphoid tissues promote microbial antigen transcytosis, facilitate the maturation of secretory immunoglobulin A+ B cells, and support the activation of antigen-specific effector T-cells (Teffs) to maintain mucosal immunity[40]. Segmented filamentous bacteria, the most abundant commensal species in the ileal mucosa, exploit suppressed entry points of microfold cells to mount a strong defense against the survival of invading pathogens[41]. As a regulator of intestinal immunity, the signalling lymphocyte activation molecule family member 4 is imprinted by the GM on T cells, B cells, professional antigen-presenting cells, natural killer cells (NKs), and innate lymphoid cells (ILCs), highlighting the critical role of the GM in immune system maturation and development[42]. In addition, the GM promotes immune tolerance and helps re-establish immune homeostasis by enhancing the cross-differentiation of CD8+ T cells into major histocompatibility complex I+ (MHC-I+) CD4+ T cells, a mechanism that may be relevant in the context of human immunodeficiency virus infection, cancer, and autoimmune disorders[43].

Next-generation probiotics for overall health

Currently, next-generation probiotics (NGPs), identified through NGS and bioinformatics approaches, exhibit distinct biological properties with potential health benefits and clinical applications. Faecalibacterium prausnitzii, one of the most prevalent anaerobic NGPs in the human colon, plays a protective role by preventing IEC damage, improving paracellular hyperpermeability, and inhibiting the activity of pro-inflammatory effectors[44]. Faecalibacterium prausnitzii also suppresses the enrichment of resident Mφs and supports the self-renewal and differentiation of epithelial progenitor and stem cells through Wnt/β-catenin signaling. Moreover, its supernatant inhibits CRC cell proliferation in a time- and dose-dependent manner[45].

Nontoxigenic Bacteroides fragilis strains, currently regarded as leading NGP candidates, prevent the adherence of opportunistic pathobionts to colonic epithelial cells and enhance mutualistic bacterial diversity[46]. These strains modulate the epithelial CD4+ T-cell compartment by promoting IL-10+ CD4+ T cells and suppressing pro-inflammatory IL-17+ CD4+ T cells in response to physiological stress or damage[47]. Microbe-derived mechanisms have evolved to support host-commensal symbiosis by inhibiting IEC apoptosis and attenuating pathogen-induced Th17 responses.

Recent evidence indicates that AKK abundance is associated with a healthier intestinal environment. AKK reduces circulating levels of pro-inflammatory bacterial lipopolysaccharides linked to increased gut permeability and stimulates the expression of tight junction proteins in IECs[48]. AKK recruits pro-inflammatory Mφs and CTLs into tumor beds, thereby inhibiting CRC progression through tumor necrosis factor-α upregulation and reduced PD-1 expression[49]. Additionally, Lactobacillus and Bifidobacterium species have been extensively studied in chronic inflammation-related disorders and are associated with favorable gastrointestinal tolerance and minimal clinical adverse effects. Lactobacillus and its metabolites enhance DC-mediated IL-10 production, thereby balancing pro-inflammatory and tolerogenic immune responses[50]. A combination of Lactobacilli and Bifidobacteria strains reduces tumor volume by fostering a healthy microbial environment rather than inducing immune-mediated or metabolomic disturbances[51]. Certain druggable bioactive molecules produced by probiotics, such as exopolysaccharides, mobilize CCR6+ CD8+ Teffs from Peyer’s patches to tumor sites, thereby enhancing the tumor-killing efficacy of ICIs[52]. These findings support the development of NGP-based therapeutic strategies, highlighting their potential as medicinal agents rather than mere dietary supplements. Such approaches aim to promote a more stable and interconnected microbial community, restore microbial homeostasis in the host, and alleviate symptoms associated with gastrointestinal diseases.

HOST-GM MUTUALISM IN CRC DEVELOPMENT

Using mathematical modeling techniques, approximately 70%-90% of CRC incidence is attributed to environmental factors, including diets low in fiber and high in red meat[53]. Abnormal lipid metabolism is increasingly recognized as a pathological hallmark of CRC. Notably, a high-fat diet (HFD) fails to support healthy regeneration of the intestinal epithelium. Consequently, changes in the relative composition of the GM occur, presenting a causal dilemma for lipid-degrading microbial populations. This microbial shift induces genetic alterations and disturbs the homeostasis of the gut lumen, ultimately promoting pathogenic infection and potentially carcinogenesis[54,55]. Translocation of specific pathobionts, induced by an HFD, is considered an initiating event that compromises immune protection and damages the intestinal epithelial barrier. Dysregulation of innate immune components is observed between DCs and ILC3s, with ILC3s producing IL-22 to mitigate intestinal inflammation associated with HFD-induced gut dysbiosis[56]. Following exposure to an HFD, Mφs are polarized toward the TAM2 phenotype, accompanied by a reduction in both the number and functionality of intratumoral CD8+ T cells. This observation suggests that CRC-associated GMD contributes to immune perturbations in the inflamed gut, potentially playing a key role in facilitating the immune system’s endogenous tumor escape behavior[57].

Mice fed an HFD exhibit an increased abundance of potentially harmful bacterial genera and a decreased presence of beneficial genera, along with associated oncogenic metabolomic abnormalities. These changes result in decreased expression of tight junction proteins, increased intestinal permeability, and enhanced proliferation and transformation of IECs[58]. In addition, expression of farnesoid X receptor, the principal regulator of bile acid synthesis, is reduced in patients with CRC[59]. Bile acid synthesis represents a key metabolic pathway for lipids, glucose, and energy utilization by the GM. When secondary bile acids (sBAs) become excessively enriched due to impaired negative feedback regulation, they can induce uncontrolled inflammation, elevate mitochondrial oxidative stress, and cause solubilization and disruption of cell membranes, ultimately leading to cellular toxicity[60].

Deoxycholic acid, one of the most abundant sBAs in humans, inhibits CD8+ T-cell activation, proliferation, effector molecule production, and cytotoxic function, thereby promoting CRC progression via the calcium signaling pathway[61]. sBAs also downregulate CXCL16 expression on endothelial cells, a chemokine essential for the recruitment of natural killer T cell to sites of inflammation[62]. Furthermore, sBAs interact with secretory immunoglobulin A, whose regulation depends on sBA levels, to enhance the adhesion and biofilm formation of enterotoxigenic Bacteroides fragilis. This process initiates a multistep protumorigenic inflammatory cascade, accelerating myeloid cell-dependent colonic neoplasia[59,63].

IECs integrate signals derived from tumorigenic pathobionts and contribute to host-microbiota mutualism that supports malignant progression. TLRs on the surface of IECs recognize bacterial toxins via MAMPs, leading to increased production of mitochondrial reactive oxygen species, which ultimately accelerates colitis-associated dysplasia and tumorigenesis[64]. IECs infected with Fn exhibit aberrant activation of the extracellular signal-regulated kinase/signal transducer and activator of transcription 3 signaling pathway, which is associated with a somatic mutation in a host RNA helicase and implicated in colorectal tumorigenesis[65]. When GMD disrupts intestinal immune homeostasis, impeding defense against virulent pathobionts and failing to suppress colonic inflammation, IECs promote polarization toward TAM2 and induce Th17 cell responses, thereby initiating a tumorigenic signaling cascade[66]. Additionally, IEC-recruited Tregs enhance the Th1/Th17 ratio by driving the differentiation of naive CD4+ T cells, contributing to protumoral IL-17 production[67].

Recent studies have demonstrated that intestinal Fusobacterium exert lineage-specific effects on oncogenesis in CRC[68]. The probiotic Streptococcus thermophilus is notably absent in stool samples from CRC patients, and its metabolite, β-galactosidase, has been shown to promote cellular senescence by impairing energy homeostasis in neoplastic cells[69]. Dysregulation of host-oncomicrobiota interactions, characterized by a competitive growth advantage of pathogenic species, exacerbates mucosal disruption and fosters the development of a unique oncogenic driver iTME. This dysbiosis induces pathogenic mechanisms similar to those driven by oxidative stress, contributing to the transformation of the gut epithelium from a normal state to hyperplasia and, ultimately, to malignancy (Figure 1). The GM primes Mφ bystander potency as an intrinsic mechanism to activate nuclear factor-κB, induce DNA damage, and inhibit DNA repair processes[70]. GMD also promotes IEC tumorigenesis, accompanied by functional exhaustion of DCs and CTLs in CRC[71]. However, aberrant host cell subtypes fail to fully initiate antitumor immune surveillance and eradication due to elevated levels of myeloid-derived suppressor cells (MDSCs), TAM2, and tumor-associated neutrophils with the iTME[72,73]. Therefore, targeted restoration of GMD and associated metabolic reprogramming, such as through fiber-rich diets or NGP treatments, alongside activation of tumor-targeting downstream immune effector cells, may represent promising therapeutic strategies.

Figure 1 Schematic diagram of gut microbiome dysbiosis attribute to colorectal cancer initiation and development. In healthy individuals, the symbiotic gut microbiome educates the host immune system to induce tolerance and re-establish immune homeostasis for survival and replication in host cells. In colorectal cancer development and progression, various oncogenic stimuli derived from gut microbiome dysbiosis increase the genotoxicity of intestinal epithelial cells and damage barrier permeability, which subsequently triggers a multistep protumoral inflammatory cascade, marked by regulatory cell recruitment (e.g., myeloid-derived suppressor cell, tumor-associated macrophage type 2) and effector cell loss (e.g., CD8+ effector T-cell, B cell), allowing tumor immune evasion. SCFAs: Short-chain fatty acids; AMPs: Antimicrobial peptides; IEC: Intestinal epithelial cell; Treg: Regulatory T cell; IL: Interleukin; Th17: T helper 17 cell; Mφ: Macrophage; DC: Dendritic cell; LPS: Lipopolysaccharide; ROS: Reactive oxygen species; CRC: Colorectal cancer; Teff: Effector T-cell; TAM2: Tumor-associated macrophage type 2; MDSC: Myeloid-derived suppressor cell.

Common CRC-associated microbial species

Although an increasing number of studies have reported GMD in CRC, identifying specific “oncogenic” microorganisms responsible for CRC initiation and/or progression remains challenging due to individual variability in GM composition. Nonetheless, metagenomic technologies and related approaches allow for the characterization of taxonomic and functional diversity, metabolomic profiles, and molecular mechanisms within perturbed microbial ecosystems (Table 1).

Table 1 Effect of pathogenic bacteria on colorectal cancer tumorigenesis.

CRC-associated bacteria	Virulence factors	Proposed carcinogenic mechanisms	Ref.
Fn	Fap2	Inhibited natural killer cell cytotoxicity and tumor-infiltrating lymphocyte activities	[76]
	FadA, Fap2, RadD, and FomA	Increased bacterial biomass and acute inflammation with pro-tumorigenic potential, tumor-associated macrophage type 2 and regulatory T cell infiltration in the tumor	[75]
	Formate	T helper 17 cell-favored proinflammatory profiles and metabolic shift with CRC stemness, invasion and metastasis	[77]
CoPEC	Colibactin	Impaired antitumor T-cell response and procarcinogenic TME leading to immunotherapy resistance	[79]
		Oncogenic-driven lipid reprogramming for lower tumor immunogenicity and acquired chemoresistance	[80]
ETBF	BFT	A pro-carcinogenic multi-step inflammatory cascade and nuclear factor-κB-activated myeloid-cell-dependent neoplasia	[63]
NTS	AvrA, SopE, SopE2, SopB, and SptP	Oncogenic transformation of pre-transformed cells upon targeting the host’s β-catenin, MAPK, and AKT signaling pathways	[81]
	Bacterial invasion and intracellular replication	Host cell metabolic transformation associated with high mammalian target of rapamycin activation	[82]

CRC: Colorectal cancer; TME: Tumor microenvironment; MAPK: Mitogen-activated protein kinase; AKT: Protein kinase B; Fn: Fusobacterium nucleatum; CoPEC: Colibactin-producing Escherichia coli; ETBF: Enterotoxigenic Bacteroides fragilis; NTS: Nontyphoidal Salmonella; Fap2: Fusobacterium autotransporter protein 2; FadA: Fusobacterium adhesin A; RadD: Radiation-sensitive DNA adhesins; FomA: Fusobacterial outer membrane protein A; BFT: Bacteroides fragilis toxin; AvrA: Anti virulence agent A; SopE: Salmonella outer protein E; SopB: Salmonella outer protein B; SptP: Salmonella protein tyrosine phosphatase.

Fn is one of the most extensively studied CRC-associated microbes, owing to its high abundance in tumor tissues and its association with poor clinical outcomes. Therapeutic strategies targeting Fn have been shown to enhance the immunogenicity of tumor cells and may help to overcome resistance to ICIs[74]. As an opportunistic inhabitant of the oral cavity, Fn adheres to and invades endothelial and epithelial cells, subsequently disseminating through the circulation and becoming enriched at oncogenic sites during infection. In CRC foci, Fn suppresses the activity of tumor-infiltrating lymphocytes and NKs, while promoting the immunosuppressive functions of TAM2, Tregs, and MDSCs[75,76]. Mature Fn-derived formate reprograms the metabolic activity of CRC cells via Th17, thereby enhancing cancer stemness, invasiveness, and resistance to therapy[77]. Although Th17 play a role in host defense against pathobionts, this function is compromised under disrupted immune equilibrium. In contrast, pathogen-driven Th17 responses can increase host susceptibility to adverse disease outcomes and neoplastic transformation[78].

CoPEC is enriched in inflamed and neoplastic lesions and is detected more frequently in human CRC biopsies compared with control samples[79]. Colibactin induces primary cell transformation and precedes the appearance of a distinct mutational signature, including chromosomal abnormalities characteristic of genomic instability. CoPEC contributes to CRC tumorigenesis and chemoresistance by promoting a metabolic energy trade-off under conditions of lipid overload, primarily through the upregulation of reactive oxygen species in cancer cells[80].

Nontyphoidal Salmonella (NTS) infections primarily occur in low- and middle-income countries and are commonly transmitted through contaminated food chains. Mild and repetitive exposure to NTS is considered an environmental risk factor that contributes to the acceleration of oncogenic cell transformation and tumor proliferation[81]. Upon invasion and intracellular replication, NTS hyperactivates the mammalian target of rapamycin pathway, the central regulator of cellular metabolism, thereby promoting cellular transformation and impairing effective host defense responses[82].

HOST-GM MUTUALISM IN CRC IMMUNOTHERAPY

Given that the GM is an essential and intrinsic component of the TME, it is increasingly recognized as a critical regulatory factor due to its interactions with host cells and its influence on interrelated gene expression involved in modulating clinical responses to immunotherapy (Figure 2)[83,84]. Upon physical attachment to the intestinal epithelium, specific GM members shape the T-cell receptor repertoire within intestinal T cells and stimulate diverse T-cell clones to exert either regulatory or pro-inflammatory activities in response to microbiota-derived cognate antigens. Subsequently, GM-produced metabolites penetrate beyond the mucosal barrier and enter host tissues and circulation, where they are sensed by immune cells.

Figure 2 Key mechanisms by which the gut microbiome reinforces immune checkpoint inhibitor’ antitumor efficacy. When responding to immune cells, friendly microbiota and metabolites facilitate host innate and adaptive immune responses, stimulating “cold-warm-hot” tumor microenvironment transition. They inhibit pathobionts and promote CD8+ effector T-cell infiltration. As “danger” signals, they are cross-presented by antigen-presenting cells (e.g., macrophage, dendritic cell) to activate natural killer cell, neutrophil, and CD8+ and CD4+ effector T-cell via the stimulator of interferon genes signaling and facilitate immune-mediated tumor clearance. AMP: Antimicrobial peptide; SCFA: Short-chain fatty acid; ICI: Immune checkpoint inhibitor; Teff: Effector T-cell; ILC3: Innate lymphoid cell 3; IEC: Intestinal epithelial cell; DC: Dendritic cell; Mφ: Macrophage; STING: Stimulator of interferon genes; NK: Natural killer cell; CRC: Colorectal cancer; TME: Tumor microenvironment.

The abundance of probiotics, such as Bifidobacterium and Lacticaseibacillus, is positively correlated with the antitumor effects of ICIs on CRC[85,86]. Probiotic-derived metabolites contribute to enhanced trafficking of CD8+ T cells into the TME, facilitate the transition from “cold” to “warm” and eventually “hot” immune phenotypes, and support intestinal integrity. As a clinically relevant adjuvant, the microbial tryptophan catabolite indole-3-carboxaldehyde not only enhances the therapeutic efficacy of ICIs but also alleviates ICI-induced intestinal toxicity. This process decreases epithelial permeability, suppresses low-grade inflammation, and corrects GMD through activation of the host IL-22 pathway[87].

How to promote cold-to-hot tumor transition through GM changes

Distinct immune landscapes have been characterized within the TME of CRC. In a “cold” non-T-cell-inflamed TME, Teff priming, trafficking, and function are inhibited, and APP are silenced. Conversely, in a “hot” T-cell-inflamed TME, there is extensive immune cell infiltration, including clonally expanded memory T-cell populations and exhausted Teffs, along with an abundance of immunosuppressive MDSCs and TAM2[88,89].

Therefore, understanding the immune landscape in CRC is critical for reshaping gut microbiota composition and stimulating the TME to unlock preexisting antitumor immunity and improve the efficacy of ICIs. In a clinical trial, Bifidobacterium longum subsp. longum BB536 enhanced both innate and adaptive immune responses by priming plasmacytoid DCs and sustaining interferon (IFN) production. Tonic IFN signaling subsequently promoted the cytotoxic activity of NK and T cells, as well as neutrophil-mediated phagocytosis of pathogens[90]. When favorable microbiota are phagocytosed by APP, such as immunostimulatory monocytes and DCs, these microbes release IFN through the cyclic GMP-AMP synthase/stimulator of interferon genes (STING) axis, increase APP, and generate a pro-inflammatory environment that facilitates T-cell priming and recruitment[91].

FMT involving a consortium of bacterial species from healthy donors into CRC mouse models collectively increased the expression of CXCL9 and CXCL10, along with other IFN-inducible genes in IECs. Additionally, FMT enhanced MHC-I expression in CD103+ plasmacytoid DCs and upregulated granzyme B in IFN-γ+-producing CD8+ CTLs, thereby augmenting ICI efficacy and improving survival[92,93].

These findings suggest that the enhanced antitumor effects of ICIs are driven by cumulative infiltration and proliferation of Teffs with tumors. They also define a microbial antigen-driven differentiation of immune cells maintained through an IFN-mediated feed-forward loop. Furthermore, enrichment of AKK by a high-fiber diet reprograms the antitumor immunogenic landscape within the iTME via IFN-dependent STING signaling, orchestrates IFN-NK-DC crosstalk, and promotes an optimal response to ICIs in cancer patients[94].

ILC3s are recently recognized innate immune cells that belong to the microbiota-sensing, tissue-resident lymphocyte lineage, characterized by a highly plastic ability to distinguish between pathogens and commensals[56]. Impairment of ILC3s in CRC patients is associated with tumor progression and resistance to ICIs[95]. In coordination with host feeding rhythms, segmented filamentous bacteria interact with the ILC3-IEC immunological circuit to induce high levels of epithelial AMP expression, suggesting that ILC3s are synchronized with host-microbe coexistence to suppress enteric infection[96]. Upon invasion by exogenous pathobionts, increased IL-22 production by ILC3s is induced through tuft cell inflammasome signaling, which plays a critical role in early host defense immunity[97]. The absence of gatekeeper MHCII+ ILC3 disrupts the microbiota-specific CD4+ Teff adaptive immune response, leading to a concomitant increase in Th17/Th1 and promoting pathological intestinal inflammation during GMD[98]. As lymphoid tissue inducers, CCR6+ ILC3s secrete CXCL10 and synergistically enhance the tumor-assault capability of ICIs by increasing the infiltration of CD4+ and CD8+ T cells into the TME[99]. Further investigation is needed to elucidate the significance of ILC3-gut microbiota mutualism in reshaping a T cell-inflamed TME and mediating protective antitumor immunity.

In the context of ICI therapy, the capacity of CRC cells to prime and amplify antigen-presenting cell-dependent innate sensing and T cell-mediated tumor-killing responses is critical. Tumor cells acquire heterogeneous phenotypes through abnormal APP expression or by hijacking immune signaling pathways to evade dynamic immune pressures within the iTME. Due to their highly commensal genomes, members of the GM, such as Enterococcus hirae, may inadvertently mimic tumor antigens while escaping self-tolerance mechanisms[100]. Subsequently, a bacteriophage-encoded, MHC-restricted epitope from Enterococcus hirae is cross-presented by DCs, leading to activation of memory IFN-γ+ CD8+ and CD4+ T cells that specifically target and eliminate tumor cells. Microbe-specific T cells are frequently detected within tumor sites, where they exhibit unique functions by recognizing tumor antigens, mediating cross-reactivity, and attacking neoplastic tissues[101]. One gut microbial metabolite, methylglyoxal, promotes immunogenic cell death and facilitates the recruitment of CD8+ T cells and NKs via STING signaling, thereby enhancing the abscopal effect of cancer therapy[102]. Growing evidence indicates that the GM interacts with the intestinal immune system and modulates the pro- and antitumorigenic balance within the TME. Enterococcus secretes antigen A via MAMPs, initiating antitumor immunity through activation of the key inflammatory nuclear factor-κB signaling pathway in ICI-responsive patients[103]. If specific microorganisms and their metabolites amplify host immunosurveillance through the release of “danger” signals, they may serve as novel adjuvants to enhance tumor antigenicity, thereby contributing directly to neoantigens generation capable of refreshing the immune system and cold-to-hot tumor transition. This shift has the potential to increase immune vigilance, stimulate the release of immunostimulatory chemotactic cytokines, promote effector immune cell infiltration, and enhance the remedial activity of ICIs, ultimately triggering a cascade of favorable biological processes that support immune-mediated tumor clearance.

CONCLUSION

The widespread adoption of NGS has led to substantial progress in elucidating the functional and mechanistic roles of GMD in CRC progression and suboptimal responses to immunotherapy. While existing studies have characterized the biological effects of MAMP-pattern recognition receptor signaling triggered by CRC-associated oncomicrobes and metabolites, reshaping the microbial community represents a promising strategy to enhance the antitumor efficacy of ICIs through targeted adjuvant interventions. Nevertheless, microbiome-based technologies remain in the early stages of clinical translation. Human clinical trials must be designed with rigorous biological, methodological, and logistical frameworks (Figure 3). Furthermore, it is essential to establish personalized and standardized approaches for therapeutic microbial modulation to ensure consistent safety and efficacy, thereby facilitating the integration of microbiome-based therapies into multidisciplinary clinical practice.

Figure 3 History of core research achievements in the microbiota-colorectal cancer crosstalk. These key achievements, from a mere case report to detailed molecular and immunological mechanism explorations, mark significant advancements in the field of host-gut microbiome mutualism under physiological conditions and cancer development, tracing back from 1951 to the current era. With the deepening understanding that suggests remodeling the microbial community represents a potential method to reinforce the antitumor efficacy of immune checkpoint inhibitors, microbe-based intervention should translate into an indispensable component of clinical therapy in the future. CRC: Colorectal cancer; GM: Gut microbiome; ETBF: Enterotoxigenic Bacteroides fragilis; GMD: Gut microbiome dysbiosis; CoPEC: Colibactin-producing Escherichia coli; FMT: Fecal microbiota transplantation; ICI: Immune checkpoint inhibitor; Fn: Fusobacterium nucleatum.

Emerging clinical strategies to manipulate the gut microbiota, such as FMT, dietary interventions, and probiotics, suggest a higher probability of treatment response, improved survival, and potentially reduced toxicities in ICIs-treated patients. For example, the Mediterranean diet, riched in plant-based foods and low in processed foods and red meat, promotes the growth of beneficial SCFA-producing bacterial populations and enhances intestinal barrier integrity. Notably, microbiome-metabolome-host interactions facilitated by precision nutritional interventions may increase CD8+ T cell cytotoxicity and reduce infiltration of MDSCs, thereby alleviating ICI resistance[104].

Advances in multi-omics technologies have enabled the identification of dynamic host-GM mutualism and the establishment of tissue-specific protein-microbe associations that reflect pathophysiological signatures of CRC. These insights pave the way for precision, integrative, microbiota-directed therapies aimed at improving clinical outcomes and quality of life for cancer patients. Despite these developments, the capacity of the regulatory network between the gut microbiota and immune cells to effectively enhance CRC treatment remains incompletely understood[105,106]. In line with this concept, reciprocal beneficial or antagonistic interactions among microbial populations provide distinct causal insights that can be integrated into individualized GM profiles via host immune responses. Additionally, reliable single-copy protein-coding marker genes have been used to preliminarily characterize the diversity of microbial phylotypes[107]. Taken together, these findings support the notion that microbe-based interventions, operating through health-oriented and adaptable mechanisms, may become essential components of antitumor therapy in the future. particularly when an immunogenic, “hot” tumor microenvironment is successfully activated.