Resolving the Paradox of Colon Cancer Through the Integration of Genetics, Immunology, and the Microbiota

Abstract

While colorectal cancers (CRC) are paradigmatic tumors invaded by effector memory lymphocytes, the mechanisms accounting for the relative resistance of MSI negative CRC to immunogenic cell death mediated by oxaliplatin and immune checkpoint inhibitors has remained an open conundrum. Here, we propose the viewpoint where its microenvironmental contexture could be explained -at least in part- by macroenvironmental cues constituted by the complex interplay between the epithelial barrier, its microbial ecosystem, and the local immune system. Taken together this dynamic ménage-à-trois offers novel coordinated actors of the humoral and cellular immune responses actionable to restore sensitivity to immune checkpoint inhibition. Solving this paradox involves breaking tolerance to crypt stem cells by inducing the immunogenic apoptosis of ileal cells in the context of an ileal microbiome shifted towards immunogenic bacteria using cytotoxicants. This manoeuver results in the elicitation of a productive Tfh and B cell dialogue in mesenteric lymph nodes culminating in tumor-specific memory CD8⁺ T cell responses sparing the normal epithelium.

Keywords: Bacteroides fragilis; Fusobacterium nucleatum; colon cancer; ileum; immune checkpoint; immunity; microbiome.

Figure 1

Heterogeneity of colorectal cancer (CRC) leading to new classifications. Several factors contribute to the intrinsic immunogenicity of CRC. There are several classifications based on anatomical, genetic, and immunological parameters which allow for the dissection of the intertwined relationships between these components and to predict clinical outcome. Therapies can contribute to gears of intrinsic immunogenicity, by providing antigens and adjuvants. While oxaliplatin (OXA)-based chemotherapeutic regimen and anti-EGFR Abs can be considered “immunogenic” therapeutics, paving the way to a better efficacy of immune checkpoint inhibitors (ICI), anti-VEGF Abs could rather modulate the vascularization and distinct immunosuppressive cues of the tumor microenvironment (TME) [such as myeloid-derived suppressor cell (MDSC)]. Despite this knowledge, the combination of immunogenic cell death-mediating compounds with ICI failed to ameliorate CRC patients’ prognosis, at least in MSS CRC. Integrating the ileal microbiome in this equation has the potential to break tolerance to self-antigens of the crypts, by priming Tfh and B cell responses, instrumental to control tumor progression.

Figure 2

Immune contexture of primary and metastatic colorectal cancer (CRC). A non-exhaustive list of the main immune features contributing to the stability or acceleration CRC progression is aligned on the left and right colon respectively. Pre-existing tumor immunity, termed “immune contexture”, monitored by “immunoscoring” as well as transcriptome deconvolution represent strong and independent predictors of long-term progression free and overall survival in CRC. Tumors enriched with cytotoxic CD8⁺, CD4⁺, in particular Th1 and Tfh, and B cells, are associated with an IFN-γ response, the upregulation of immuno-inhibitory molecules and better clinical outcome (left). Other types of inflammation, characterized by IL-17 expressing T cells, FOXP3^hi Tregs and immunosuppressive myeloid populations are associated with worse clinical outcome. The composition of the ileal microbiome contributes to shift the balance between Tfh and Th17 cells.

Figure 3

Geographical distributions of immune and microbial components highlight specificities between the ileal and colonic segments. The most dominant microbial and immune cells accounting for the diversity and specificity of each organ are depicted. The main specialized functions (metabolic, immune, motility, etc.) for each segment are indicated. A graphical abstract is lining below the boxes where these cells and functional specificities are listed. Immune cells are found either clustered within organized lymphoid structures or scattered within intestinal epithelium (IE) and lamina propria (LP). AMP, anti-microbial peptides; B cell, B cell lymphocyte; DC, dendritic cell; ILC3, innate lymphoid cell type 3; IL-22, interleukin-22; Th1, T helper 1 cell; Th17, T helper 17 cell; Treg, regulatory T cell.

Figure 4

Instrumental links between intestinal dysbiosis and colon carcinogenesis. Risk factors contributing to coloractal cancer (CRC) initiation and development encompass the life style, diet, the exposure to environment (xenobiotics), host genetics and the gut microbiome (A). Several hypotheses have been formulated to account for the toxicity of the microbiome for the epithelium. A driver bacterium could recruit a consortium of disease-facilitating microbes (B). The “keystone hypothesis” suggests that specific bacteria lead to dysbiosis and pro-carcinogenic microenvironment (C). Mucus digestion by some bacteria could expose the epithelium to toxins produced by virulent bacteria organized in biofilms (D).

Figure 5

Pharmacological-microbiota interactions. Drugs and microbes can mutually influence each other. A synergy - exemplified by oxaliplatinum - can occur culminating in triggering the local immune system eventually controlling the tumor microenvironment. This synergy is severely compromised by antibiotics. (A) Drugs can shape gut microbiota and activate bacterial functions beneficial for the anticancer therapy. For instance, CTLA-4 blockade could enrich the ileal microbiome with distinct Bacteroides spp. that in turn fosters Th1 responses beneficial against cancer. (B) On the other hand, commensals can metabolize anticancer prodrugs, thus activating or inactivating their bioactive metabolite, sustaining or compromising the therapeutic effect. Conversely, increased recycling of bioactive compounds may be accelerated by enzymatic machineries associated with distinct microbes (such as camptothecin-11), generating severe side effects. (C).

Figure 6

Evidence pointing to a key role of the gut microbiota in immune checkpoint inhibitors (ICI)-mediated anticancer effects. Recently, pre- and clinical studies have shown the clinical significance of the composition of the gut microbiota in ICI efficacy. 1) Retrospective and prospective studies have highlighted the detrimental effect of antibiotics administration within the month preceding the start of immunotherapy in multiple stage III and IV cancers. 2) Metagenomic analysis (MGS) of patients’ stool composition predicts primary resistance to ICI in 1L and 2L melanoma, kidney and lung cancers. 3) Avatar mouse models where the intestines of tumor bearing rodents are colonized by human stools from cancer patients at diagnosis allowed to predict the response to immunotherapy in patients. 4) The identification of beneficial bacteria, such as Akkermansia muciniphila or Bifidobacterium pseudolongum for immunotherapy efficacy opens up prospects to restore cancer-associated dysbiosis in patients.

Figure 7

The role of Tfh and B cell orchestration for the efficacy of ICI in TMB high breast tumors or MSS CRC endowed with an immunogenic ileal microbiome. The efficacy of OXA in mouse model of CRC is driven by ileal features. On one hand, OXA induces apoptosis and the release of DAMPs following caspase-3/7-dependent ileal epithelial cell death of the crypts. The bacterial composition is critical to turn this cell death into immunogenic cell demise instead of a tolerogenic cell death. The ICD triggers the migration of CD103⁺ conventional dendritic cells (BATF3⁺ cDC1) to the mesenteric lymph node (mLN), which prime Tfh cells in an IL-1β and IL-12 dependent manner. Next, a crosstalk between Tfh and B cells occurs, leading to a systemic IgG2b response and an accumulation of Tfh and TILs into MSS or MSI colon tumors. The IgG2b responses are suspected to be directed either against bacteria or tumor cells. (A) Activation mechanisms between B cells and Tfh result in the generation of antibodies that elicit antibody-dependent cellular cytotoxicity (ADCC) towards tumor-associated antigens in mouse model of breast cancers endowed with a high tumor mutational burden (TMB) (B). Small intestine immunogenic species (B.pseudolongum or A.muciniphila) secrete the inosine, a metabolite activating Th1 and Tc1 cells through the A_2A. Conjointly with IL-12 co-stimulatory signal produced by cDCs in presence of antigens, inosine-stimulated Th1 and Tc1 secrete IFN-γ and increase ICIs efficacy in MSS or MSI mouse models of colon cancer (C).