Tumour neoantigen mimicry by microbial species in cancer immunotherapy

Abstract

Tumour neoantigens arising from cancer-specific mutations generate a molecular fingerprint that has a definite specificity for cancer. Although this fingerprint perfectly discriminates cancer from healthy somatic and germline cells, and is therefore therapeutically exploitable using immune checkpoint blockade, gut and extra-gut microbial species can independently produce epitopes that resemble tumour neoantigens as part of their natural gene expression programmes. Such tumour molecular mimicry is likely not only to influence the quality and strength of the body's anti-cancer immune response, but could also explain why certain patients show favourable long-term responses to immune checkpoint blockade while others do not benefit at all from this treatment. This article outlines the requirement for tumour neoantigens in successful cancer immunotherapy and draws attention to the emerging role of microbiome-mediated tumour neoantigen mimicry in determining checkpoint immunotherapy outcome, with far-reaching implications for the future of cancer immunotherapy.

Fig. 1. Understanding checkpoint immunotherapy responsiveness.

Depicted are important clinical parameters known to influence the response to checkpoint-based immunotherapy. The expression of the targets of immune checkpoint inhibitors (ICIs), the level of tumour mutational/neoantigen burden and the extent of T cell infiltration at baseline are established clinical predictors of ICI treatment response; the molecular contribution of the (gut) microbiome, however, is less well-studied. Epitope mimicry of tumour neoantigens is likely to explain the functional importance of microbial species in checkpoint immunotherapy responsiveness and arguably deserves prioritisation for scientific elaboration efforts. Note the interdependence of the various checkpoint-predictive markers and parameters, indicated with dashed lines. In brief, cancer cell-specific mutations are a requirement for tumour neoantigens, the number of which is thought to correlate with the overall tumour mutational burden. Both tumour neoantigens and tumour mutational burden influence the infiltration of tumours by T cells and are further associated with checkpoint molecule expression. The host microbiome also modulates the response to checkpoint blockade and is connected with tumour neoantigens through molecular mimicry. PD-L1, programmed cell death ligand 1.

Fig. 2. Strategies for studying microbial tumour neoantigen mimicry in mice and men.

a Tumour cells of known neoantigen status and corresponding engineering of gut commensals can be used to generate mouse models of enforced microbial tumour neoantigen mimicry. The effects of such mimicry on checkpoint immunotherapy responsiveness and tumour growth can be investigated and the quality and strength of the anti-cancer immune response can be assessed. System perturbations such as co-treatment with antibiotics can be introduced to fine-tune the degree of tumour neoantigen mimicry in these well-defined models. b Using whole-exome sequencing and RNA sequencing of tumour and blood samples from cancer patients treated with immune checkpoint inhibitors, tumour neoantigens binding to respective MHC class I molecules can be identified using bioinformatics. In parallel, shotgun sequencing of stool samples can characterise the microbial metagenome and identify the antigenic sequences present in the gastrointestinal tract. Using this information, the degree of immunological similarity between tumour cells and gut microbes can be determined and summarised as ‘tumour antigenic similarity’ (TAS). Using TAS-based patient stratification, the impact of microbial tumour neoantigen mimicry on treatment response and survival can be evaluated. Companion immune profiling using methods such as enzyme-linked immunospot assay and T cell receptor sequencing can be used to corroborate clinical survival data and indicate the potential underlying mechanisms of treatment response or failure. ELISPOT enzyme-linked immunospot assay, NGS next-generation sequencing, TAS tumour antigenic similarity, TCR T cell receptor, WES whole-exome sequencing.

Fig. 3. Indirect evidence suggests beneficial effects of tumour neoantigen mimicry.

a Scenarios of reduced or absent microbial diversity (representing little/no mimicry) are associated with poor responsiveness to immune checkpoint inhibitors (ICIs), whereas scenarios of reinstated or higher microbial diversity (representing relevant/substantial mimicry) are associated with favourable ICI responsiveness. b Schematic showing the putative role of microbial tumour neoantigen mimicry in checkpoint-based immunotherapy. Microbial species mimicking tumour neoantigens can independently elicit tumour-reactive T cells based on antigen-presenting cell (APC) cross-priming or default APC pathways (extracellular microbes), or native ‘self’ antigen processing in infected host cells (intracellular microbes). As shown in the green box at the right-hand-side of the panel, neoantigen molecular mimicry can result in overlapping CD8⁺ (and CD4⁺) T cell specificities which might potentiate ICI therapy and contribute to protective cancer immune surveillance through reinvigoration of tumour neoantigen-specific immunity. ABx antibiotic treatment, APC antigen-presenting cell, FMT faecal microbiome transplantation, ICI immune checkpoint inhibitor.