The foundations of tumor immunology were established by Lloyd J. Old when he described the concepts of immunosurveillance and identified both tumor antigens and escape mechanisms (Old, 1981; Old, 1985; Kaplan et al., 1998). They demonstrated that tumors could be rejected by transferring immune cells from immunized mice into naive mice providing early evidence for cancer immunotherapy using immune cells (Old et al., 1962). These and other early discoveries paved the way to a broad array of cancer treatments that could be classified as cancer immunotherapies. Cancer immunotherapy, put simply, is a type of treatment that harnesses the power of the immune system to recognize and attack cancer cells. There are several approaches that cancer immunotherapy can utilize, and these therapies have revolutionized cancer treatment in since their introduction (Kotla et al., 2009; Kronke et al., 2015). Currently, the two main clinical therapies are checkpoint inhibitors and adoptive cell therapy (ACT).

Checkpoint inhibitors are drugs that obstruct specific proteins on immune cells or cancer cells to prevent the suppression of immune responses against tumors. Current clinically approved targets for inhibitory checkpoints include CTLA-4 and PD-1. Researchers have found that by blocking these inhibitory checkpoint proteins, immune checkpoint inhibitor antibodies effectively take the brakes off the immune system, amplifying its cancer-fighting abilities (Esfahani et al., 2020).

Adoptive cellular therapy is based on the activation and manipulation of autologous T cells in an in vitro setting followed by reinfusion to achieve better tumor targeting. This modality has three main approaches: tumor-infiltrating lymphocyte therapy (TILs), T cell receptor (TCR) gene therapy, and chimeric antigen receptor (CAR) modified T cells (June et al., 2015).

Tumor infiltrating lymphocyte (TILs) therapy involves expanding autologous TILs from resected tumors ex vivo and infusing them back into patients after lymphodepletion. In animal models, TILs from various mouse tumors have demonstrated the ability to combat tumors in vivo (Forget et al., 2018). Human patient studies by Radvanyi et al. showed TILs from melanomas reliably recognize autologous tumors (Radvanyi et al., 2012). In various phase I/II clinical trials for metastatic melanoma, TIL therapy resulted in tumor responses nearing 50% (Andersen et al., 2016). As of early 2024, the Food and Drug Administration approved Lifileucel from Iovance Biotherapeutics (Amtagvi) as the first approved solid tumor derived autologous T cell immunotherapy.

T cell receptor gene therapy (TCR) serves as another avenue for ACT, where patient T cells are reprogrammed to identify tumor antigens by introducing genes that encode tumor-specific TCRs (D’Angelo et al., 2018). One of the notable targets in this therapy includes cancer germline antigens such as NY-ESO-1. NY-ESO-1 is expressed in up to 52% of melanomas (Goydos et al., 2001), neuroblastomas (Camisaschi et al., 2018), synovial sarcomas and mixoid and round cell liposarcomas, and ovarian cancer (Robbins et al., 2015). Research has shown promising results, with response rates reaching up to 30% in melanoma cases (Robbins et al., 2015). However, there are concerns that toxicities might arise from off-tumor reactivity.

Chimeric Antigen Receptor T-cell (CAR-T) therapy is a prominent form of adoptive cell transfer (ACT), using engineered T cells to target tumor antigens through chimeric antigen receptors. These chimeric receptors incorporate parts of various other receptors to optimize cellular function. The CAR-T cells have achieved remarkable success targeting CD19 in blood cancers, with high response rates in lymphoma (Davila et al., 2014; Miller and Maus, 2015). However, translating CAR-T cell efficacy to solid tumors poses numerous challenges like the tumor immunosuppressive microenvironment and extensive extracellular matrix surrounds solid tumors, often physically blocking CAR-T migration and tumor penetration (Martinez and Moon, 2019; Donnadieu et al., 2020).

Overall, immunomodulatory drugs offer diverse approaches within cancer immunotherapy. Often used in combinations to maximize effectiveness, immunotherapeutic regimen design depends on the type of cancer, the stage, and the patient’s overall health (Wang et al., 2022). Additionally, emerging research suggests that a patient’s microbiome can significantly influence the effectiveness of cancer immunotherapy, leading to the development of therapies aimed to manipulate the microbiome and enhance the immune response against cancer.

The human microbiome, comprising trillions of microorganisms, has gained attention due to its pivotal role in health and disease, particularly its potential influence on cancer and cancer immunotherapy. The microbiome’s composition and diversity can significantly influence immune responses, making it an intriguing factor to investigate in the context of personalized cancer treatment (Gopalakrishnan et al., 2018a).

The intersections between the microbiome and cancer can be categorized at two levels (Jain et al., 2021). The first is between microbiome composition and the development of tumor, and the second between the microbiome and the immune system which in turn can direct the immune response to cancer immune therapy. The impact of microbes on the immune response and their potential as therapeutic targets in cancer treatment have been subjects of interest in recent research (Barukčić, 2017; Turna et al., 2017; Priya et al., 2022).

Metabolites and cytokines are one of the main methods of crosstalk between the microbiome and the immune system. The gut microbiota produces various metabolites and cytokines that can shape systemic immunity. For example, certain bacteria metabolize dietary fiber into short-chain fatty acids (SCFA) like butyrate, which have anti-inflammatory effects and promote T-cell differentiation (Maslowski et al., 2009; Kim et al., 2014). Gut microbe released SCFAs enter circulation and modulate systemic immune cell differentiation and function. SCFAs can promote regulatory and T helper cell 1 and 17 cell differentiation and hence affect the immune therapy activity (Kim et al., 2016). Also, the bacterium B. pseudolongum produces inosine which enhances checkpoint inhibitor efficacy (Mager et al., 2020). This effect could be modulated through the inhibition of ubiquitin-activating enzyme UBA6 in tumor cells which in turn augment its sensitivity to the cytotoxic activity of T cells (Zhang et al., 2022). Interestingly, some metabolites like butyrate can also have concentration-dependent opposite effects such that low levels induce T regulatory cells while higher levels boost CD8+ T cell effector activity. In the clinical setting, responsive cancer patients were reported to show a higher systemic SCFA level (Arpaia et al., 2013).

Several strategies, including fecal microbiota transplantation (FMT) and probiotic administration, have thus been adopted in attempts to enhance cancer immunotherapy efficacy. In a study by the Vétizou group (Mao et al., 2021) examining both mouse models and melanoma patient samples, it was revealed that the therapeutic efficacy of CTLA-4 blockade is dependent on key gut commensals such as Bacteroides thetaiotaomicron and B. fragilis. Furthermore, the presence of T cells specific for these bacteria was associated with superior CTLA-4 response rates. Depletion of these bacterial populations, either in germ-free mice or through antibiotic treatment, mitigated the therapeutic effect. However, repopulation using probiotics or T-cell-targeted approaches could restore responsiveness to CTLA-4 inhibition.

Along the lines of microbiota manipulation to enhance the cancer immunotherapy response, another study identified that Bifidobacterium administration, either through fecal transfer or oral routes, was able to halt melanoma progression similarly to PD-L1 antibody therapy. Furthermore, the combined effects of Bifidobacterium and PD-L1 blockade together led to a significant inhibition of tumor growth. This highlights the synergistic potential of microbial secretants and checkpoint inhibitors in overcoming cancer progression (Sivan et al., 2015). Another preclinical study in germ-free mice showed that FMT from responder patients can replicate patient response to checkpoint inhibitors; non-responder phenotype in mice is reversible with additional FMT from responders (Routy et al., 2018). Several clinical trials are now testing microbiome restoration to improve immunotherapy response. For example, in melanoma patients refractory to anti-PD-1 treatment, FMT was tested in combination with pembrolizumab. This combined therapy resulted in stabilized disease or tumor regression in 2 patients out of 3 [clinical trial: NCT 03341143]. Given these promising results, the mechanism behind the gut microbiome modulating the cancer immune therapy response is of great interest to optimize and direct the use of microbiome manipulation, either at the study design level or for therapeutic intervention.

One of the commonly reported mechanisms for microbiome interaction with immunotherapy response is through the presentation of pro-inflammatory antigens. The gut microbiome interacts with host immune cells via pattern recognition receptors like Toll-like receptors, stimulating inflammatory responses that help shape tumor immunity (Sato and Kawakami, 2022). Gut bacteria provide antigens that are sampled by intestinal immune cells and transported to lymphoid tissues to initiate B cell and T cell responses. Cross-reactivity between microbial and tumor antigens can boost anti-tumor immunity not only locally but systematically (Mowat and Agace, 2014).

A study highlighted the concept of systemic circulation of bacterial antigens and their role in cross-reactivity with pancreatic tumor antigens (Daillère et al., 2016). It reported that translocation of Enterococcus hirae from the gut to lymph nodes/spleen during cyclophosphamide chemotherapy enhanced anti-tumor immune responses in mice. This effect was linked to an E. hirae bacteriophage antigen cross-reactive with a tumor antigen, highlighting microbiome antigen mimicry of tumor antigens as an immunomodulatory mechanism (Fluckiger et al., 2020).

In efforts to best study tumor pathophysiology and therapeutic responses, researchers employ a multitude of in vivo and in vitro platforms, each with their own benefits and limitations ( Table 2 ). Two-dimensional plate-based cell culture and animal-based cancer models utilize almost exclusively different methodologies to examine disease growth and progression.

Conventional in vitro platforms include numerous tumor cell lines used in 2D formats for genetic pathway analysis and drug screening. While these models contribute important findings to the field of cancer biology and produce high throughput results and easily replicable, cell lines can have limited validity. With continual passages, genetic and phenotypic morphology can drift from the original tumor composition (Torsvik et al., 2014; Yu et al., 2019; Quevedo et al., 2020). Additionally, lack of cell-cell and cell-extracellular matrix contact in traditional 2D cultures reduces cell capacity to faithfully model in vivo cancer-associated pathways and interactions (Luca et al., 2013).

In vivo cancer models include genetically engineered animals to mimic human tumors, patient and cell line derived tumor xenografts, and spontaneous tumor development in veterinary species. While these models more readily capture the complexity of tumor development in the whole organism, they can be limited by their reproducibility, cost, and construction time. Murine xenograft models, which allow for the study of human derived tumor and associated microenvironment in a living system, are expensive, require immunocompromised host animals, can take up to 6 months or longer to produce, and cannot support all tumor types (Abdolahi et al., 2022). More readily available genetically engineered animals, typically mice, serve an important role in tumor development and drug discovery research. However, their predictive value in translational research can be limited by the lack of etiologic similarity in tumor development and key microbiome and immune system structure differences from humans (Mestas and Hughes, 2004; Dawson et al., 2009). Spontaneous, naturally occurring tumor animal models, particularly in nonhuman primates, allow for effective translational study of tumor development and tumor, immune, microbiome interaction (Deycmar et al., 2023). Yet, these animals also incur significant expense, can have limited throughput, and not all human tumors are represented in other species. To bridge the limitations between the conventional 2D tumor cell culture models and in vivo models, tumor organoids are quickly gaining popularity.

Recent developments in biomaterials have resulted in more physiologically accurate culture methods that can be used to study complex human systems (Rossi et al., 2018). Specifically, advances in immune population and microbiome ex vivo models have allowed for composite models to examine interactions between the two systems ( Figure 1 ). Organoids have been used to model many aspects of the microbiome and immune environment including different bacterial species, viral infections, and effects on various cellular niches (Min et al., 2020). Anti-PD-1 and anti-CTLA-4 have been tested in microfluidic devices to model efficacy (Aref et al., 2018). Organoids and spheroid culture have also demonstrated their use as a model for studying ICI efficacy (Jenkins et al., 2018; Neal et al., 2018). Microbiome derived factors and viable immune responses can therefore be combined to create a model that can demonstrate the interplay between microbiome derived metabolites and ICI in an organoid system (Shelkey et al., 2022). Chip systems have even been constructed that incorporate microbiota compartments to produce immunomodulatory effects on the tissue present with specific species causing identifiable inflammatory responses through factors like metabolite production and reactive oxygen species (De Gregorio et al., 2022). It is even possible to modulate the microbiome in chip-based systems to cause inflammation and bacterial outgrowth, which can then be monitored in ways that incompatible with animal models (Kim et al., 2016). The high throughput nature of ex vivo culture makes it ideal for conducting large scale studies that work to isolate individual components underpinning the mechanisms of action for complex systems (Rae et al., 2021). Further advances in advanced cell culture models will continue to contribute to understanding immunotherapy-microbiome interactions.

Ex vivo culture of cells and tissue constructs is continuously advancing with applications in precision medicine, immune modeling, and organ system replication. Organoids have proven to be particularly effective at predicting cancer patient response to therapeutic treatment (Nagle et al., 2018; Mazzocchi et al., 2018; Votanopoulos et al., 2019a; Forsythe et al., 2021). Advances in microfluidic production have allowed multiple organ systems to be integrated in one construct to demonstrate how all of the different systems interact (Skardal et al., 2017; Ingber, 2022). These systems are also significantly better for continuous monitoring of cell populations of interest (Kim et al., 2016). Microbiome modulation and analysis in correlation with cancer immune therapy and cancer progression has been investigated in several clinical studies to evaluate the positive outcomes and drawbacks. These studies also aim to estimate the potential of antibiotics that affect gut microbiome composition and the subsequent influence on either cancer promotion or response to therapeutics. There have previously been difficulties in producing ex vivo culture models that are exposed to live bacterial populations, with tissue and live bacterial interactions being limited to animal models. Some researchers compensated by focusing on bacterial metabolites that could be used in aseptic culture; however, intestine-on-a-chip models have now been produced that are able to sustain equivalent microbiomes in culture with intestinal epithelium. These models replicate oxygen gradients, intestinal barriers, and can be utilized going forward for the controlled testing of therapeutics (Jalili-Firoozinezhad et al., 2019; Shin et al., 2019). Different versions of the gut-on-a-chip model can even model peristaltic movements while maintaining a normal microbial population (Kim et al., 2012). With the continuous advances in cell culture technology producing models that were previously impossible, it is likely that more ex vivo models better able to reproduce the complicated cancer immunotherapy-microbiome interactions will be produced. These models can more easily be leveraged for high-throughput studies both for mechanistic investigation and predictive precision medicine, resulting in better patient care. These results will continue to supplement and corroborate the ongoing clinical trials that aim to elucidate the interactions between the microbiome and cancer immunotherapy.

With several drugs in the clinical pipeline and more clinical trials in progress, immunotherapies are a promising therapeutic for a wide array of tumor types (Cloughesy et al., 2019; Ni et al., 2021; Rothschild et al., 2021; Vignali et al., 2022; Garon et al., 2023; Huang et al., 2023; Boesen et al., 2024). To continue to grow the list of available immunotherapies, advanced testing platforms are necessary. Tumor organoids, particularly those derived from human primary specimens, are an increasingly important platform to personalize current therapies and model various tumor microenvironment interactions to accelerate novel drug development. In particular, the capacity to co-culture microbiome, immune cells, and tumor organoids provides critical insight into complex interplays which regulate immunotherapy responses. As we highlight in this review, these dynamic mechanisms cannot be accurately modeled in traditional 2D culture systems, and animal models fail to provide fully translational findings. As microfabrication technologies continue to evolve and improve the organoid systems, these ex vivo assays are a crucial tool to innovate immunotherapy treatment strategies through understanding of the microbiome-immune-tumor interactions and improve patient outcomes.