Targeting Neoepitopes to Treat Solid Malignancies: Immunosurgery

Abstract

Successful outcome of immune checkpoint blockade in patients with solid cancers is in part associated with a high tumor mutational burden (TMB) and the recognition of private neoantigens by T-cells. The quality and quantity of target recognition is determined by the repertoire of 'neoepitope'-specific T-cell receptors (TCRs) in tumor-infiltrating lymphocytes (TIL), or peripheral T-cells. Interferon gamma (IFN-γ), produced by T-cells and other immune cells, is essential for controlling proliferation of transformed cells, induction of apoptosis and enhancing human leukocyte antigen (HLA) expression, thereby increasing immunogenicity of cancer cells. TCR αβ-dependent therapies should account for tumor heterogeneity and availability of the TCR repertoire capable of reacting to neoepitopes and functional HLA pathways. Immunogenic epitopes in the tumor-stroma may also be targeted to achieve tumor-containment by changing the immune-contexture in the tumor microenvironment (TME). Non protein-coding regions of the tumor-cell genome may also contain many aberrantly expressed, non-mutated tumor-associated antigens (TAAs) capable of eliciting productive anti-tumor immune responses. Whole-exome sequencing (WES) and/or RNA sequencing (RNA-Seq) of cancer tissue, combined with several layers of bioinformatic analysis is commonly used to predict possible neoepitopes present in clinical samples. At the ImmunoSurgery Unit of the Champalimaud Centre for the Unknown (CCU), a pipeline combining several tools is used for predicting private mutations from WES and RNA-Seq data followed by the construction of synthetic peptides tailored for immunological response assessment reflecting the patient's tumor mutations, guided by MHC typing. Subsequent immunoassays allow the detection of differential IFN-γ production patterns associated with (intra-tumoral) spatiotemporal differences in TIL or peripheral T-cells versus TIL. These bioinformatics tools, in addition to histopathological assessment, immunological readouts from functional bioassays and deep T-cell 'adaptome' analyses, are expected to advance discovery and development of next-generation personalized precision medicine strategies to improve clinical outcomes in cancer in the context of i) anti-tumor vaccination strategies, ii) gauging mutation-reactive T-cell responses in biological therapies and iii) expansion of tumor-reactive T-cells for the cellular treatment of patients with cancer.

Keywords: T-cell receptor; T-cells; TIL; antigens; immunotherapy; neoepitopes; precision medicine; vaccination.

Figure 1

Mutation analysis reveals immune-recognition profile in the TME. Whole-exome sequencing data allows for mining of private somatic mutations in tumor samples compared to healthy (non-transformed) tissue or cells, which is unique to each patient. The stringency of the filtering parameters applied in bioinformatics and statistical analysis of the sequencing data will greatly influence the number of mutations recovered, which are essential for downstream characterization of immune responses of T-cell products. Highly stringent parameters may yield a lower number of mutations albeit with an exceptional level of accuracy. Nevertheless, this approach suffers the risk of overlooking several infrequent mutations which also give rise to immunogenic (T-cell reactive) neoepitopes in the patient. On the contrary, reducing the stringency levels of analysis may reveal rare mutations which facilitate the identification of potentially immunogenic molecular targets recognized by certain TCRs capable of eliciting a biologically relevant anti-tumor immune response. The drawback in the latter scenario is that a high degree of false positive hits may be obtained and included in the final list of legitimate cancer-associated somatic mutations. Thus, a balanced yet wholistic approach is required to identify all immunogenic mutations in tumor tissue which will be instrumental in developing personalized cancer therapies.

Figure 2

TMB-directed immunotherapy approaches at the Champalimaud Centre for the Unknown. The schematic shows strategies aimed at therapeutic targeting of private (personalized and patient-specific) and shared (often driver) mutations. For personalized therapy, CD4⁺ and CD8⁺ T-cells from TIL or peripheral blood expressing a highly diverse TCR αβ repertoire recognizing a private neoepitopes can be procured. HLA-matched, healthy donor-derived TCRs have also been shown to recognize patient-specific neoepitopes (21). Personalized cancer vaccines, comprising private neoepitopes as a peptide formulation or as RNA constructs, promote durable immune responses in patients with advanced cancer. Autologous B-cells can be used as a source of APCs as well as cytokine producers, in addition to their differentiation into plasma cells to secrete tumor antigen-specific antibodies in vivo. Approaches targeting shared mutations serve as excellent ‘off-the-shelf’ options which can be used for larger groups of patients simultaneously. Cancer vaccines based on shared mutations are also clinically important, provided the patients’ HLA profiles are matched to the epitope binding characteristics. Antibodies derived from tumor-infiltrating B-cells or from peripheral blood B-cells targeting surface-bound shared neoantigens may mediating cellular cytotoxicity and aid in the development of CAR T-cells. Gene therapy to correct shared driver mutations may promote tumor susceptibility to immune attack. Immune checkpoint blockade has been placed between the two domains as its clinical activity targets both private and shared mutated targets. Similarly, NK, TCR γδ T-cells and possibly NKT T-cells or MAIT-cells may be instrumental in patients presenting with private and/or shared HLA pathway mutations and can be derived from allogeneic sources for treatment.

Figure 3

Schematic representation of the HLA class I and II pathways and T-cell activation. The HLA class I pathway is also known as the intrinsic pathway as it processes and presents endogenous antigens while antigens derived from the extracellular environment are processed and presented via the HLA class II (extrinsic) pathway. LMP2/7 are immunoproteasome subunits necessary for generating short epitopes (7-11 amino acids along), which are then loaded on the HLA class I molecule for presentation to CD8⁺ T-cells. The β2-microglobulin (β2M) is critical for the assembly and stable expression of HLA class I-peptide complexes on the cell surface. On the other hand, HLA class II molecules first exist with the class II-associated invariant chain (CLIP) for stability, which is then removed with assistance from the HLA-DMA/B complex, for loading of CD4⁺ T-cell epitopes generated via lysosomal degradation. Processed antigens are then presented by either HLA-II (extrinsic pathway) or HLA-I (intrinsic pathway), to T-cells to initiate an immune synapse followed by activation of the latter. Indeed, as a result of cognate antigen recognition, T-cells may produce one or a combination of effects: i) cellular proliferation (also involves IL-2), ii) increase in cytotoxicity (may be measured by surface CD107a induction assay), iii) induction of 4-1BB expression and/or iv) production of cytokines, such as IFN-γ, TNF-α, IL-2, IL-17c.

Figure 4

T-cell phenotype and functional-spatial differences. TIL were expanded from different regions from a pancreas cancer lesion metastatic to the liver, 5 regions were harvested in different proximity to the tumor center. Note the different homing/maturation phenotype based on CD45RA/CCR7 expression, central memory T-cells in the tumor periphery. Thus, the quality of the T-cell response (to neoepitopes) is also associated with the immune cell maturation status. Reactivity to (mutant) KRAS or mesothelin was tested by pre-incubation of TIL for 5 days followed by IFN-gamma production analysis. Exclusive KRAS recognition in the tumor center versus mesothelin recognition in the tumor periphery and in macroscopically cancer-negative tissue demonstrating that the selection of neoepitope specific T-cells depends on the anatomical location.

Figure 5

PCV development and immuno-analyses workflow at the ImmunoSurgery Unit. Formalin-fixed paraffin-embedded (FFPE) or fresh-frozen tissue samples prepared by the Pathology Unit at the CCC is submitted for WES of tumor DNA with the patient’s PBMCs as an internal control for downstream analysis, RNA-Seq is also sometimes performed to tumor RNA. The WES and RNA-Seq raw data is then analyzed at the ImmunoSurgery Unit at the CCU to predict private mutations followed by HLA class I and II binding prediction matched to the patients’ HLA restriction profile to select candidates for inclusion in the PCV formulation. Only HLA-binding, neoepitope-containing peptides but not the wildtype counterparts are considered. The same and also 15-mer equivalent but non-clinical grade peptides, alongside the corresponding native sequences, are used for gauging TIL and/or PBMCs reactivities based on IFN-γ production (the peptide is at a concentration 1ug/mL tested with 10e⁴ responder T-cells; the fixed T-cell number allows to compare results obtained at different timepoints or from tissues harvested from different tumor areas). This part of the immunological evaluation of the neoepitopes is used in the follow-up phase of the trial which aims to assess T-cell responses of patients to the PCV and TIL therapies (possibly also for patients receiving immune checkpoint inhibitors). A different platform is an ELISA panel comprising the patient’s neoepitopes in linear format to assess IgG reactivity using antibodies from serum as well as those secreted by TIB and PBMC-derived B-cell lines. Neoepitopes are also screened for TIL recognition since TIL are routinely generated to gauge for differences in TIL versus PBMC recognition. This will allow to describe whether selected neoepitopes are recognized in the tumor lesion that was used to identify the tumor neoepitopes (by NGS), it also allows to screen for differences in TIL recognition from tumor lesions harvested at different anatomical sites or at different timepoints in the course of the disease.

Figure 6

Schematic representation of the general molecular paradigm of neoantigen recognition in the TME. The process of transcription of DNA to RNA and then to protein (antigens) is prone to generate heterogeneity in the context of cancer, i.e., the same DNA molecule may be differentially transcribed (due to RNA alternative splicing or mutations) and then translated to different proteins isotype (also as a result of post-translation modifications) or there might be gene fusions that result in novel RNA transcripts. The heterogeneous expression of tumor antigens, as a result of spatial-temporal differences in DNA to antigen production, results in different antigens being presented to the immune system by HLA complexes (as well as whole antigens) at the cell surface of a tumor cell and, therefore, contributing to different sub-regional TMEs within the same tumor tissue sample. These are likely to be neoantigens, as they are not present in healthy (non-transformed) tissue. The TCR diversity (“adaptome”) will also change depending on the specific TME, i.e., different TCRs will be encountered depending on intratumoral spatial differences. Along the same lines, molecular structures associated with the microbiome present in the tumor tissue may cross-react with some T-cells, depending on the presence of absence of TCRs that recognize such microorganisms. The possible cross-reactivity, if present, may favor the expansion of the relevant immune-cell populations and, therefore, change the TCR repertoire.

Figure 7

Different immune-textures in cancer lesions. Starting point for WES and RNA-Seq. Definition and documentation of the immune cell infiltrate. Parallel slices of the paraffin-embedded tissues are procured and subjected to DNA and RNA analysis. Note the different patterns of CD3⁺ T-cell clusters (left) versus individual CD3⁺ T-cells in close proximity to tumor cells. RNA isolated from this tumor section would also allow for deep TCR-sequencing and allow to trace back individual TCR CDR3 motifs in case if neoepitope specific TCRs are identified.

Figure 8

Example of a standard immuno-histological analysis of a tumor sample at the Clinical Pathology Unit. Analysis of CD3⁺, CD4⁺ and CD8⁺ T-cell infiltrates along with tumor-associated CD68⁺ macrophages. Testing for MHC class I (HLA-A, B and C) expression to screen whether transformed calls can be recognized by CD8⁺ T-cells, general MHC class I loss would not support vaccination strategies of adoptive T-cell therapy targeting TCR alpha/beta T-cells as the immune effector population. CD47, PD-1 and PDL-1 expression to gauge immune escape. Examination of commonly shared, non-mutant TAAs (NY-ESO-1, survivin, mesothelin) to identify T-cell responses in TIL and in corresponding PBMCs. Expression analysis of TAAs aids in quality control concerning RNA-Seq (of the corresponding gene coding for the TAA) and deep TCR analysis of T-cells reacting to TAAs.