From the summary of existing studies ( Table 1 ), it is evident that there is a close relationship between the commensal microbiome of the host and the occurrence and development of HCC. Most of the research has focused on fecal samples. Data on serum samples and liver tumor tissue samples also exist.

In a study comparing the gut microbiome of elderly HCC patients and healthy individuals using next-generation sequencing of fecal samples, both α diversity and β diversity were statistically different. At the genus level, several bacterial groups, including A.Blautia, Fusicatenibacter, Anaerostipes, Lachnospiraceae_ND3007_group, CAG-56, Eggerthella, Lachnospiraceae_FCS020_group, and Olsenella were significantly decreased in the HCC group compared to the control group. Conversely, the abundance of E. coli-Shigella, Fusobacterium, Megalococcus, Veillonella, Tyzzerella_4, Prevotella_2, and Cronobacter was significantly increased in the HCC group (31). Furthermore, the composition of bacteria in the serum of liver cancer patients differs from that of cirrhosis patients (32). Pseudomonas was found to be significantly reduced in HCC, whereas Staphylococcus, Acinetobacter, Klebsiella, and Trabulsiella were significantly enriched. The presence of intratumoral microbiome in HCC tumor tissues is of increasing concern (33), and significant differences in microbial composition have been observed between liver cancer tissues and para-cancerous tissues. Notably, the microbial diversity in liver tumor tissues is significantly higher than in para-cancerous tissues. The abundance of Enterobacteriaceae, Fusobacteria, Neisseria, and other microbiome in HCC tissues is considerably higher, while certain anti-tumor bacteria, such as Pseudomonas spp., is decreased.

Ninety percent of HCC cases arise from cirrhosis, during which liver cells undergo chronic cycles of necrosis and regeneration (42). Compared with patients with cirrhosis (34), those with early-stage liver cancer caused by cirrhosis showed increased intestinal microbiome diversity. Specifically, phylum Actinobacteria was significantly increased in early HCC. Correspondingly, 13 genera, including Gemmiger, Parabacteroides, and Paraprevotella, were enriched in early HCC compared to liver cirrhosis. On the other hand, phylum Verrucomicrobia was decreased in early HCC. At the genus level, 12 genera, including Alistipes, Phascolarctobacterium, and Ruminococcus, were significantly reduced, while six genera, including Klebsiella and Haemophilus, were increased in early HCC compared to controls. Notably, a reduction in butyrate-producing bacteria (Ruminococcus, Oscillibacter, Faecalibacterium, Clostridium IV, and Coprococcus) and an increase in lipopolysaccharide-producing bacteria (Klebsiella and Haemophilus) were observed. Butyrate is the primary energy source of the intestinal mucosa, playing a pivotal role in bacterial energy metabolism and intestinal health (43). Therefore, reducing butyrate-producing bacteria may contribute to intestinal mucosal destruction and the development of liver cancer (44). Increased LPS levels can activate the NF-κB pathway and produce pro-inflammatory cytokines, leading to liver inflammation and oxidative damage, promoting the development of HCC (45). Furthermore, research in Argentina found specific changes in the Firmicutes members and identified potential biomarkers of HCC, including Odoribacter, Butyricomonas, and Lachnospiraceae family genus Dorea (35).

In another study comparing primary liver cancer patients with cirrhosis patients and healthy individuals, the diversity of Firmicutes showed a downward trend from the healthy group to the liver cirrhosis group to the primary liver cancer group. Enterobacter ludwigii increased in the primary liver cancer group. Further analysis revealed correlations between clinical indicators and intestinal microbiome, such as a positive correlation between Veillonella and AFP in the primary liver cancer group and a negative correlation between Subdolicapsulum and AFP (36).

To administrate the best-fit treatment, it is indispensable for cancer stages. Usually, cancer staging is based on the size of the tumor, whether it has metastasized to lymph nodes, and whether it has metastasized to other organs. Different cancers have their unique staging system. Various staging systems used in HCC include the European systems [French staging system, Barcelona Clinic Liver Cancer (BCLC) staging system, and the Cancer of the Liver Italian Program (CLIP)] and Asian systems [Okuda staging system, Japan integrated Staging (JIS), Tokyo score and Chinese University Prognostic Index (CUPI)] (46). There were significant differences in the intestinal microbiome of patients at different stages of HCC (37). In 86 stool samples, 604 bacterial genera were identified. Compared with healthy controls, patients with stage I HCC showed enhancement of Actinomyces, Atopobium, Desulfococcus, Enterobacter, Paraprevotella, Planctomycetes, Prevotella, Veillonella, and many unidentified genera. Patients with stage II HCC exhibited enrichment of Desulfococcus, Enterobacter, Lactococcus, Leptotrichia, Paraprevotella, Planctomycetes, Prevotella, Veillonella, and many unknown genera. Patients with stage III HCC demonstrated multiplication of Actinomyces, Atopobium, Desulfococcus, Enterobacter, Haemophilus, Lactococcus, Leptotrichia, Neisseria, Oribacterium, Prevotella, Rothia, Selenomonas, Veillonella, and many unidentified genera. Moreover, Desulfococcus, Enterobacter, Prevotella, and Veillonella were increased in all stages of HCC. However, patients with stage I HCC showed a reduction in Acidaminococcus, Cetobacterium, Coprobacillus, Pyramidobacter, Turicibacter, and two unidentified genera; patients with stage II HCC exhibited a decrease in Anaerotruncus, Cetobacterium, and an unknown genus. Additionally, patients with stage III HCC showed a reduction in Acidaminococcus, Anaerostipes, Anaerotruncus, Butyricimonas, Cetobacterium, Cloacibacillus, Coprobacillus, Holdemania, Methanobrevibacter, Odoribacter, Pyramidobacter, Turicibacter, and four unidentified genera. Cetobacterium was reduced in all stages of primary HCC.

Another study tested 45 patients’ feces with early, intermediate, and late stages and found that the intestines of the early-stage liver cancer group had a higher abundance of Clostridium, Firmicutes, and Streptococcus (38). The mid-stage liver cancer group presented more intestinal Ruminococcaceae, Pasteurellaceae, Tanticharoenia, and Vagococcus genera. The advanced liver cancer group displayed more Bifidobacteria, Actinobacteria, Barnella, Porphyromonas, and Pseudomonas. In summary, during the development of liver cancer, there are significant differences in the composition of the intestinal flora of patients, in addition to changes in relevant clinical indicators.

HBV is one of the most critical factors causing HCC, accounting for approximately 50% of cases (47). Studies have compared the differences in intestinal microbiome between HBV-HCC and non-HBV non-HCV related HCC (NBNC-HCC) patients (39). Fecal samples from 90 patients were analyzed, revealing significant differences in the diversity and composition of the intestinal microbiome. Specifically, HBV-HCC patients had higher species richness, while NBNC-HCC patients had decreased Firmicutes and increased Proteobacteria at the phylum level. NBNC-HCC patients had fewer potential anti-inflammatory and pro-inflammatory bacteria, whereas HBV-HCC patients had more potentially anti-inflammatory bacteria. Overall, HBV-HCC and NBNC-HCC patients exhibited variations in the abundance of bacteria involved in different functions or biological pathways, suggesting that alterations in specific gut microbiome could have therapeutic benefits for both types of HCC. Another study compared gut microbiome differences between HBV-HCC patients and healthy individuals (40). The researchers found that HCC patients with high tumor burden had an enrichment of Bacteroides, Lachnospiracea incertae sedis, and Clostridium XIVa. Furthermore, a strong correlation was observed between the expression of these three genera of bacteria and the host tumor microenvironment genes. Based on further clinical characterization and database analysis, serum bile acids were identified as potential communication mediators between these bacteria and the host transcriptome. This study suggests that changes in the tumor immune microenvironment associated with intestinal microbiome, mediated by serum bile acids, might be crucial in tumor burden and adverse clinical outcomes in HBV-related HCC.

Numerous recent studies have investigated the contribution of fungi to the onset and progression of various tumors, such as those found in colorectal (48–55), pancreatic (56), skin (57), bladder (58), lung (59) cancers, and more. Meanwhile, Fungal colonization in cancer patients manifests in different tissue and cell type niches. One of the most recent pan-cancer study has detected fungi in 35 types of tumors, indicating that they may exist in tumor cells or immune cells, similar to intracellular bacteria (24). However, the mechanisms by which fungi enter these cells in tumor models and how they survive within them are still unknown.

As opportunistic pathogens, fungi pose a severe threat to people’s health, particularly those with compromised immune function, such as AIDS patients, tissue transplant recipients, or cancer patients (60), additionally, individuals with mutated antifungal immunity-related genes are prone to frequent mycoses (61). To evade the innate immune surveillance, fungi employ similar strategies in different diseases (62). Such as Cryptococcus neoformans has been reported to cross the blood-brain barrier by concealing itself within macrophages, leading to meningitis (63). And another example is Candida albicans, which can survive within macrophages by undergoing morphological transformation, forming hyphae, and compromising the integrity of the phagosomal membrane (64).

Stool samples from HCC and liver cirrhosis patients demonstrated decreased fungal diversity in the HCC group, with a higher abundance of Candida genus and Candida albicans (65). Similarly, patients with liver cancer showed reduced fungal diversity compared to the healthy group, with significantly higher abundances of Candida albicans and Malassezia in the feces of liver cancer patients. Studies on animals have verified that unusual growth of Candida albicans and Malassezia furfur in the gut can contribute to the onset of HCC (41).

Recent research that employs metagenomic sequencing and other techniques has given us a deeper understanding of the diversity of the human virome in various parts of the body, its connection to diseases, and how it forms during the early stages of life (66). This human virome is made up of bacteriophages, which are viruses that infect bacteria, as well as viruses that target other cellular microorganisms such as archaea, those that invade human cells, and those that are temporary components of food (67–73). Notably, the virome can elicit a broad range of immune responses (74).

Various types of cancer have been linked to viral infections ( Figure 1C ). For example, human papillomavirus (HPV) is recognized as a requisite component in the formation of both pre-invasive and invasive lower genital tract cancers, with the most common being cervical cancer (75), and Malignant transformation of HPV can also cause head and neck squamous cell carcinoma (HNSCC) (76). Likewise, the Epstein-Barr virus (EBV) is viewed as a primary causal agent in the occurrence of nasopharyngeal cancer (77). HBV is identified as a partially double-stranded DNA virus, in contrast to HCV, a positive-sense single-stranded RNA virus. These viruses are known as the primary causal factors of HCC. Furthermore, other liver viruses, like hepatitis G virus, are also possible causative factors for HCC, even though their correlation has not been conclusively proven (78). Adeno-associated virus 2 has been reported to cause HCC (79), and next-generation sequencing technologies have uncovered associations of other viruses with HCC, suggesting that a range of viruses could be involved in its pathogenesis (80).

HCC virome study mapped 93,904 viral epitopes and identified the most variable viruses in tumor patients versus controls, utilizing large-scale screening of thousands of serological samples and machine learning approaches (7). It elucidated a viral exposure signature to define the early onset of HCC through the large-scale screening of thousands of human samples and applying pan-viral-epitope screening of human virome in HCC. This viral exposure signature (VES) was defined by comparing population controls and HCC patients, comprising unique epitopes from 61 viral strains (18 high-risk associated strains and 43 low-risk associated strains). Notably, 11 of the 18 high-risk associated strains were HCV, a well-established trigger of HCC, and were more prevalent in cases than in controls. The VES also included previously unknown viruses associated with HCC, with an enrichment of cytomegalovirus (CMV), hepatitis delta virus (HDV), and influenza strains H1N1 and H3N2. Emerging infectious diseases significantly burden national public health and may substantially influence tumor progression and therapeutic efficacy. Therefore, the virome profile can be the key biomarker for clinical characteristics and outcomes.

Studies have focused on investigating the role of commensal microbiome in ICI treatment of HCC (97). Metagenomic sequencing has been used to detect the dynamic characteristics and specificity of the intestinal microbiome during anti-PD-1 immunotherapy in HCC. Stool samples from patients who responded to immunotherapy showed higher tax on richness and more gene counts than non-responders. The microbial composition of responders remained relatively stable at the phylum level and was enriched in Myxobacteria and Ruminococcus species, whereas Proteobacteria increased in non-responders (98).

Another study found that several specific taxa were significantly enriched in the clinical benefit response (CBR) group compared to the non-clinical benefit group (NCB). For example, patients with a higher abundance of Lachnospiraceae bacteria - GAM79 and Alistipes sp Marseille - P5997 had longer progression-free and overall survival times. On the other hand, patients with higher Veillonellaceae abundance had poorer progression-free and overall survival (99).

In a prospective analysis of fecal samples from unresectable HCC patients before immunotherapy, significant differences in fecal bacteria were observed between patients who responded to therapy and those who did not. Prevotella 9 was enriched in non-responders, while Clostridium, Lachnospiraceae, and Veillonella were dominant in responders. Furthermore, specific metabolites such as ursodeoxycholic acid and urscholic acid were significantly enriched in responders and closely related to Clostridium trichophyton’s abundance (100).

These findings suggest that the diversity and specific composition of intestinal microorganisms and changes in particular metabolites can be practical biomarkers to predict the clinical response and survival benefit of HCC immunotherapy. Moreover, certain microorganisms or metabolites enriched in non-responders or responders may become potential targets to improve the efficacy of immunotherapy for HCC. For example, combining acetate with a PD-1/PDL-1 blocker has significantly enhanced the anti-tumor effect (101).

In the intestines of healthy individuals, bacteria and fungi maintain homeostasis. Unfortunately, compared with bacteria, there is a lack of research on the role of commensal fungi in the ICIs immunotherapy of cancer. Studies that use bacteria to enhance the therapeutic efficacy of immune checkpoint inhibitors may disrupt the original bacterial-fungal homeostasis, and these studies have almost without exception ignored commensal fungi. Therefore, the symbiotic fungi group will be a very promising area in subsequent research.

Oncolytic viruses (OVs) are an emerging class of cancer therapeutics with several advantages in cancer treatment ( Figure 1C ). Selectively infecting and destroying cancer cells while sparing normal cells, OVs result in minimal damage to healthy tissues (102, 103). This targeted approach stimulates an antitumor immune response by presenting tumor-associated antigens to the immune system (104). Additionally, OVs induce the expression of inflammatory and immunomodulatory cytokines in the tumor microenvironment, further enhancing the immune response (105). Efficacious against various malignant neoplasms, including drug-resistant lymphoproliferative diseases (106), OVs have shown promise in clinical trials with well-tolerated outcomes for cancer patients (107). Furthermore, OVs offer the potential for delivering multiple eukaryotic transgene payloads, inducing immunogenic cell death, and promoting antitumor immunity (108). Overall, this evolving approach in cancer therapy represents a promising avenue to provide added patients benefits.

Oncolytic viruses (OVs) have been extensively studied as promising cancer therapeutics. They were found to inhibit HCC growth, including directly dissolving tumor cells, exposing neoantigens, sequentially activating anti-tumor immunity (109). The most studied OVs are a vaccinia virus, JX-594 (110, 111). This engineered OVs, JX-594, have two genes inserted at TK gene region. One gene encodes human granulocyte-macrophage colony-stimulating factor (hGM-CSF) and the other gene encodes lac-Z (112, 113) to form this functional vaccinia virus. JX-594 can induce anti-tumor immunity and inhibit tumor blood vessels by promoting the maturation of myeloid cells and dendritic cells. Additionally, other appealing oncolytic vectors have been developed for the treatment of HCC, such as a new chimeric vector called recombinant VSV-NDV (rVSV-NDV) (114), and Ld0-GFP, based on herpes simplex virus type 1 (HSV-1) (115). Both vectors have shown excellent cancer cell killing ability and significantly improved the survival in the HCC mouse model from both in vivo and in vitro studies. Notably, the combination of oncolytic viruses with other immunotherapies like immune checkpoint blockade (ICB) and chimeric antigen receptors (CAR) has shown promising tumor treatment efficacy (116). Hence, the approach of combining the oncolytic viruses with ICB or CAR will bring new hope to HCC treatment.

Endogenous retroviruses (ERVs) are retroviral sequences that have become integrated into the genome of a host species. They are remnants of ancient retroviral infections that have been passed down through generations (117). ERVs can be found in the genomes of various organisms, including humans (118). The host has co-opted ERVs over time, and some ERV genes have been repurposed for host functions (119). ERVs are regulated by various mechanisms, including epigenetic modulation, and their dysregulation has been associated with neurological diseases, cancer, and inflammatory processes (120).

Endogenous retroviruses (ERVs) have several advantages in cancer immunotherapy ( Figure 1C ). First, ERV reactivation can induce an interferon response, known as viral mimicry, which sensitizes tumor cells to immunological recognition (121, 122). This viral mimicry state can be triggered by drugs or cellular changes in tumor cells (123). Second, ERV expression can lead to the production of tumor-specific antigens (TSAs). These TSAs, derived from translated Human ERV elements, contribute to the landscape of antigens recognized by the immune system (124, 125). They have been recognized by cytotoxic CD8+ T cells, resulting in cancer cell recognition. Third, the combination of viral mimicry and T-cell recognition can enhance the effectiveness of existing immune stimulatory therapies, such as checkpoint inhibition. In summary, the reactivation of ERVs in cancer cells has the potential to sensitize tumors to immunotherapy and improve treatment outcomes.

Recent advances have been made in the use of ERVs for cancer immunotherapy. Human ERVs represent 8% of the human genome and have been identified as potential tumor antigens for immunotherapy. Bioinformatic approaches have been developed to identify shared CD8+ T cell epitopes derived from cancer-associated HERVs in solid tumors. In vitro priming assays have confirmed the immunogenicity of these epitopes, leading to the induction of high-avidity CD8+ T cell clones. HERV-specific T cells, which have been shown to specifically recognize and kill tumor cells presenting HERV epitopes on HLA molecules, have also been identified among tumor-infiltrating lymphocytes from patients with breast cancer (126). Recent studies have shown that anti-ERV antibodies play a role in the effectiveness of ICB immunotherapy in mouse and human lung cancer, and nearly half of the LUAD patient samples showed antibody activity against the human endogenous retrovirus HERV-K. Among the 7 LUAD patients who received ICB, all had increased HERV-K envelope-reactive antibodies. Additionally, it was found that this increase in antibodies was positively correlated with survival after discontinuation of ICB (127).

The innate immune system plays a vital role in the host’s resistance to pathogenic bacterial infections as the second line of defense. Innate immune cells recognize Pathogenic Molecule-related Patterns (PAMPS) on the surface of pathogenic bacteria through pattern recognition receptors. This theory, proposed by Janeway in 1989 (129), is based on the idea that pathogenic molecule-related patterns on the cell wall of pathogenic bacteria, such as lipopolysaccharide, muramyl dipeptide, lipoteichoic acid, and peptidoglycan, or intracellular nucleic acid molecules. And the pattern recognition receptors are the receptors present on the innate immune cells, such as Toll-like receptors (TLR) and NOD-like receptors (NLR), which can recognize exogenous or endogenous dangerous molecules. Once relevant danger signals are recognized, a series of inflammatory responses are initiated, leading to the production of inflammatory cytokines and mediation of immune cell phagocytosis to eliminate corresponding pathogenic bacteria or abnormal cells. The above important findings lead to an intriguing note that the phagocytosis, which can be exploited as a cargo for the pathogenic bacteria to evade immune surveillance.

Intracellular bacteria are bacteria that live inside the cells of their host organisms. They can be found in various types of hosts, including corals, protists, arthropods, marine invertebrates, mammals, immune cells and cancer cells. These bacteria have developed many strategies to survive and travel within the host cells (130, 131). The presence of intracellular bacteria can be found in immune cells and other cell types in tumor (132). In addition, intracellular bacteria have been found to promote tumor metastasis (133).

Intracellular bacteria have developed multiple mechanisms to evade the host immune system (134). Moreover, bacteria can interfere signaling of immune cells to suppress the host’s immune response. Mycobacterium tuberculosis evades the immune response by inhibiting the NF-κB signaling pathway through the secretion of Rv0222 after being engulfed in immune cells (135). Taken together the interaction of intracellular bacteria with the host immune system can impact tumor progression.

Bacteria have been found to contribute to the advanced stages of HCC. Moreover, the role of bacteria in immune checkpoint blockade (ICB) is increasingly recognized. Hence, it is of utmost importance to examine the correlation between specific bacteria and the evolution of cancer. In light of this, we provide an overview of groundbreaking studies that probe the linkage between particular bacteria and cancer and further delve into the consequences of bacteria in the domain of ICB treatment. These reports serve as valuable references for future HCC translational research ( Figure 1A ).

Fungi, such as Candida, Saccharomyces, and Malassezia, are essential components of human commensal microbiome, colonizing various niches within the human body. The gut and other mucosal surfaces are typical habitats for these fungi (17, 201, 202). Candida species predominantly colonize the oral cavity, while Malassezia dominates the skin (203, 204). Despite accounting for only 1% of the human commensal microbiome, fungi significantly impact the host’s health and disease (17). They can act as opportunistic pathogens and infect immunocompromised patients, including those with cancer (24). Notably, their effects should not be overlooked ( Figure 1B ). The host’s natural immune system identifies fungi through their microbe-related molecular patterns, notably the carbohydrate components of the fungal cell wall. The detection of these patterns are carried out by pattern recognition receptors, such as C-type lectin receptors, found on epithelial and myeloid cells. Receptors such as Dectin, Lectin, Mincle, Toll-like receptors, and Nod-like receptors (NLR) activate signaling mechanisms involving SYK/CARD9, SYK/PLCy2, MYD88, and TRIF when they bind (205). Consequently, this triggers the production of signaling molecules, including interleukin 1β(IL-1β), IL-6, IL-12, IL-23, TGF-β, and interferon γ, which stimulate the immune response of T helper 1(Th1) and Th17 cells (206).

The Hepatitis B Virus (HBV) contributes to HCC through several processes. The HBV X protein (HBx) is instrumental in the propagation and spread of HCC (222). HBx interacts with a protein called the suppressor of cytokine signaling 1 (SOCS1), which inhibits the breakdown of p65, a component of NF-κB. This interaction activates factors associated with the epithelial-mesenchymal transition (EMT), a significant process in the migration and invasiveness of cancer cells (223). HBV infection also promotes disease progression through mechanisms such as HBV gene integration, genomic instability, activation of cancer-promoting signaling pathways, and alteration of hepatocellular physiology (224, 225). Additionally, miR-142-3p, highly expressed in HBV-infected HCC patients, promotes HBV-infected M1-type macrophage ferroptosis, affecting the production of reactive oxygen species and accelerating HCC development. The complexities of HBV-induced HCC mechanisms are illuminated by these findings, which also suggest possible therapeutic targets for intervention HCV infection induces HCC also through a variety of mechanisms. Chronic inflammation, oxidative stress, insulin resistance, endoplasmic reticulum stress, hepatic steatosis, and liver fibrosis are all included in this category (226). HCV-related proteins, such as HCV core, E1, E2, NS3, and NS5A, dysregulate cell cycle and metabolism by modulating signal pathways (227). HCV infection also leads to genetic and epigenetic modifications, including host genetic factors, dysregulation of signaling pathways, and histone and DNA modifications (228, 229). In addition to causing genetic and epigenetic changes, such as alterations in host genetic elements, disruption in signaling pathways, and changes in histones and DNA, HCV infection also results in excessive production of reactive oxygen species (ROS) and diminished function of natural antioxidants. This may cause damage to the DNA, lipids, and proteins (230). ROS activates signaling cascades, and the activity of transcription factors are modulated, which leads to changes in gene expression associated with cell survival, proliferation, angiogenesis, invasion, and metastasis. These processes play a role in the advancement and evolution of HCV-related HCC.