Article Text
PDF
Abstract
Background Elimination of cancer cells by T cells is a critical mechanism of antitumor immunity and cancer immunotherapy response. T cells recognize cancer cells via engagement of T cell receptors with peptide epitopes presented by major histocompatibility complex (MHC) molecules on the cancer cell surface. Discovery of the full landscape of tumor antigens in any given individual will be important for understanding response to immunotherapy and optimizing strategies for antigen-specific immunotherapy development. Although T cell epitopes can be derived from antigen proteins coded for by multiple genomic sources, bioinformatics tools used to identify tumor-specific epitopes using DNA and RNA sequencing data have largely focused on epitopes derived from somatic variants.
Methods We report here an open-source workflow utilizing the Nextflow DSL2 workflow manager, Landscape of Effective Neoantigen Software (LENS), which predicts tumor-specific and tumor-associated antigens from single nucleotide variants (SNVs), insertions and deletions (InDels), gene fusions, splice variants, cancer testis antigens (CTAs), overexpressed self-antigens, viruses, and human endogenous retroviruses (hERVs). The main advantage of LENS is that it extends the breadth of genomic sources of tumor antigens that may be discovered using genomics data. Other advantages include modularity, extensibility, ease of use, incorporation of phasing and germline variant information in epitope identification, and harmonization of relative expression level and immunogenicity prediction across multiple genomic sources (table 1). LENS is open-source and freely available for academic use (https://gitlab.com/bgv-lens).
Results To demonstrate the utility of LENS, we present an analysis of the predicted antigen landscape in 115 acute myeloid leukemia (AML) samples. AML was chosen due its low somatic mutation rate and dearth of classical (SNV-derived) neoantigens. Predicted tumor antigens were distributed unevenly across genomic sources (figure 1). The highest degree of antigen sharing was found in those derived from aberrantly expressed endogenous retroviral genes, which were also highly-expressed in the RNA-seq data (figure 2). A small number of gene fusion-derived antigens were also highly expressed with high predicted binding affinity (figure 3).
Conclusions We expect that LENS will be a valuable platform and resource for analysis of polyepitopic T cell responses across the full immunodominance hierarchy, antigen selection for therapeutic neoantigen vaccination, and T cell epitope discovery, especially in cancers with few somatic mutations and increased importance of tumor-specific epitopes from genomic sources beyond single nucleotide variants. In AML specifically, our results support further development of hERV and gene fusion-based tools for immune monitoring and therapeutics going forward.
TCGA AML tumor antigen distribution by patient sample. Tumor antigen distributions across 115 TCGA-LAML samples are shown. Predicted neoantigen number ranges between 40 to 2,433 peptides with a median of 319 peptides per patient
Antigen sharing in AML. (A) The number of AML antigens predicted to be presented in multiple TCGA AML samples are shown along with their genomic sources. Antigens identified in more than 10% of samples are enclosed by the red box. (B) RNA expression of predicted AML antigens by genomic source
Fusion peptide read support vs. HLA Class I binding affinity across patients. Several predicted fusion-derived peptides show both relatively high binding affinity and high relative peptide abundance (blue). Others show either high expression and low affinity or low expression and high affinity (light red), or both low expression and low affinity (red)
Comparison between LENS and other Neoantigen workflows LENS generally offers similar or improved functionality compared to other popular neoantigen workflows. Specifically, LENS supports more tumor antigen sources, includes a harmonized estimate of antigen expression, does not require end-user pre-processing of input data, and is a both modular and extensible workflow. It is worth noting that all the included workflows are open-source and may subjectively be classified as modular and extensible as a result. We classify LENS as modular and extensible as these were desired features that drove design and development rather than consequences of code availability. The ‘Hybrid’ classification for nextNEOpi’s containerization metric is due to some components, like pVACtools and NeoFuse, having self-contained containers while other tools (e.g. variant callers) are all contained within a single container.