Pancreatic cancer–restricted cryptic antigens are targets for T cell recognition.

Translation of the noncoding genome in cancer can generate cryptic (noncanonical) peptides capable of presentation by human leukocyte antigen class I (HLA-I); however, the cancer specificity and immunogenicity of noncanonical HLA-I–bound peptides (ncHLAp) are incompletely understood. Using high-resolution immunopeptidomics, we discovered that cryptic peptides are abundant in the pancreatic cancer immunopeptidome. Approximately 30% of ncHLAp exhibited cancer-restricted translation, and a substantial subset were shared among patients. Cancer-restricted ncHLAp displayed robust immunogenic potential in a sensitive ex vivo T cell priming platform. ncHLAp-reactive, T cell receptor–redirected T cells exhibited tumoricidal activity against patient-derived pancreatic cancer organoids. These findings demonstrate that pancreatic cancer harbors cancer-restricted ncHLAp that can be recognized by cytotoxic T cells. Future therapeutic strategies for pancreatic cancer, and potentially other solid tumors, may include targeting cryptic antigens. Editor's summary: Cancer neoantigens are proteins expressed by tumor cells that can trigger an immune response that eliminates the tumor. A major challenge that limits their use as successful immunotherapy for solid tumors, particularly pancreatic cancer, is rarity of such cancer-specific proteins and peptides that are consistently expressed by most of the tumor cells. Ely et al. performed an immunopeptidomics study of human pancreatic cancers and identified a large number of unique peptides derived from noncoding regions of the genome (see the Perspective by Tuveson). Certain peptides were immunogenic and provided a source to generate T cell–specific immune responses that could kill pancreatic cancers in preclinical models. Some of the cancer-derived peptides were shared by multiple patients and were absent from normal pancreatic cells, raising the prospect of universal "off-the-shelf" therapies, as opposed to more challenging personalized approaches. —Priscilla N. Kelly INTRODUCTION: Cancer cells can translate ostensibly noncoding genetic elements to produce cryptic (noncanonical) peptides that can be presented by human leukocyte antigen class I (HLA-I). Cryptic peptides are poorly characterized in most solid tumors, where immunotherapies often yield limited efficacy. A deeper understanding of cryptic peptide presentation and immunogenicity could open new therapeutic avenues in solid tumors, such as pancreatic cancer. RATIONALE: Pancreatic cancer has been refractory to immunotherapies. A broader study of its antigen landscape, inclusive of noncanonical peptide sources, could identify antigens with potential cancer-restricted presentation. In this work, we investigated HLA-I–bound cryptic peptides in pancreatic cancer and assessed their immunogenic potential. We also developed a system to generate and functionally characterize T cell receptors (TCRs) that recognize these cryptic peptides. RESULTS: We performed high-depth immunopeptidomics on organoids and bulk tumors from pancreatic cancer patients and found that organoids markedly improved specificity for cancer-derived peptides. Despite constructing personalized proteogenomic search spaces for each patient, we found that only a few mutation-encoded peptides were detectable using immunopeptidomics. By contrast, we empirically identified more than a thousand cryptic peptides presented by HLA-I. These cryptic peptides originated from diverse genomic elements, such as long noncoding RNAs (lncRNAs), 5′ or 3′ untranslated regions (UTRs), and alternative reading frames (intORFs). We also evaluated a wide range of normal tissues, including healthy thymus, and nominated cryptic peptides with potential cancer restricted translation. Nearly 30% of cryptic peptides identified in pancreatic cancer were not detected in any healthy tissue profiled. We refer to these as cancer-restricted cryptic peptides. Many cancer-restricted cryptic peptides were shared by multiple patients, including nearly 50% of those predicted to bind to HLA-A*02:01. We evaluated the immunogenic potential of cryptic peptides using an ex vivo T cell priming and expansion platform. Cancer-restricted cryptic peptides demonstrated robust immunogenicity, on par with that of mutation-encoded neoepitopes. Through barcode-enabled antigen mapping, we isolated TCRs specific for cryptic peptides at single-cell resolution. We reconstructed a subset of these TCRs to confirm antigen specificity, characterize TCR avidity, and assess potential cross-reactivity. Using CRISPR-based TCR redirection (TCR-T), we demonstrated that T cells could recognize endogenous levels of cryptic antigens on tumor cells. Furthermore, TCR-T cells specific to cancer-restricted cryptic antigens could exert robust killing of patient-derived pancreatic cancer organoids both ex vivo and in vivo. CONCLUSION: We addressed two major gaps in the study of antigenic peptides in cancer—namely, the cancer restriction and immunogenicity of noncanonical peptides. These results demonstrate that aberrant translation in pancreatic cancer can give rise to cryptic antigens capable of recognition by T cells. Our data support that a subset of cryptic peptides are cancer restricted, immunogenic, and directly targetable on the surface of human pancreatic cancer cells, nominating them as potential therapeutic targets that warrant further investigation. Cryptic antigens in pancreatic cancer.: Aberrant translation of the noncoding genome results in noncanonical (cryptic) peptides that can be processed and presented by HLA-I in pancreatic cancer. Cryptic peptides are potently immunogenic, and cognate TCRs are capable of recognizing and killing patient-derived pancreatic cancer organoids both ex vivo and in vivo. nuORF, novel unannotated ORF; dORF, downstream ORF; Ribo-seq, ribosome profiling; CTLs, cytotoxic T lymphocytes; IFN-γ, interferon-γ; TNFα, tumor necrosis factor–α. [ABSTRACT FROM AUTHOR]