Description
Predicted Neoantigens Serve as Attractive Targets for Tumor Infiltrating Lymphocytes in Pediatric Leukemias
Anthony E. Zamora1, Jeremy C. Crawford1, Ti-Cheng Chang2, Pradyot Dash1, Jesse Bakke3, Hossam A. Abdelsamed1, Mari H. Dallas4, Terrence L. Geiger5, Robert A. Carter5, Taosheng Chen2, Douglas R. Green1, Benjamin A. Youngblood1, Jinghui Zhang2, Paul G. Thomas1
1Department of Immunology, 2Department of Computational Biology and Bioinformatics, 3Department of Chemical Biology and Therapeutics, 4Department of Bone Marrow Transplantation and Cellular Therapy, 5Department of Pathology, St. Jude Children’s Research Hospital, Memphis, Tennessee, USA
Abstract: The use of tumor-reactive T lymphocytes as a means of adoptive immunotherapy has emerged as a promising therapeutic against various cancers. Currently, several approaches towards achieving tumor specificity are beginning to bear fruit, including the use of: tumor infiltrating lymphocytes (TILs), chimeric antigen receptors, and T cell receptor (TCR) engineered T cells. Although these adoptive cellular approaches have shown clinical utility in adult tumors with relatively high numbers of nonsynonymous mutations, it is unclear whether similar adoptive immunotherapies can be exploited in pediatric tumors and/or in tumors with lower mutational burdens. In this study, we aimed to determine the clonality, differentiation status, functionality, and phenotype of tumor infiltrating lymphocytes from pediatric patients with acute lymphoid leukemia (ALL), which on average has lower rates of nonsynonymous mutations compared to adult solid tumors. We hypothesized that although leukemias contain lower mutational burdens, the neoantigens that arise in this tumor type serve as immunodominant antigens capable of inducing robust antitumor T cell responses. In order to test our hypothesis, we began by generating a list of predicted neoantigens for each patient sample and determined the HLA-binding potential of each of these neoantigens using computational algorithms. We next determined the tumor-associated T cell receptor repertoire with single cell resolution, allowing us to determine the structure of tumor-associated TCR repertoires and their propensity to target neoantigen. Finally, we tested and validated each candidate predicted neoantigen using in vitro peptide stimulation assays in order to determine which elicited the best T cell responses measured by TCR activation. Our results show that although pediatric leukemias contain fewer nonsynonymous mutations, a high frequency of the predicted neoantigens are expressed and elicit an antitumor response. Additionally, we found evidence of clonal expansion in some patient tumors, while others had broader TCR diversity. Our study expands on the current knowledge of cancer immunotherapy by characterizing the phenotypic and functional status of TILs in pediatric leukemias and suggests that these tumors can induce a robust antitumor T cell response. Additionally, our findings provide a platform for designing therapeutic interventions, including engineered T cells, against identified neoepitopes, which serve as attractive antitumor targets.