Identification and Functional Validation of Neoantigen-Specific T Cells in Pediatric Patients with Fusion-Derived Acute Leukemias

Abstract

Leukemia emerges from the accumulation of genetic mutations in hematopoietic cells, resulting in formation of altered proteins. Some of these mutant proteins (neoantigens) are consequently processed and presented as neoepitopes through major histocompatibility complex (MHC) molecules expressed on the surface of leukemia cells, permitting tumor-specific immune targeting. While adoptive T cell therapies against high mutation burden tumors have shown promising results, less is known about leveraging tumor-specific T cells as an effective approach in pediatric patients, including high-risk leukemia. In this study, peripheral blood or bone marrow samples were obtained from 34 leukemia patients aged 0 -14 years old. The patient tumors had founder fusion genes, including 15 patients with acute lymphoblastic leukemia (ALL), 15 patients with acute myeloid leukemia (AML), and 4 patients with mixed phenotype acute leukemia (MPAL). Samples were collected at the time of diagnosis or relapse. While leukemia blasts were the most abundant cellular component in these samples, the frequency of lymphocytes ranged between 1%- 25%, with a median of 3.5%. KMT2A rearrangements were expressed with either AFF1, MLLT1, MLLT3, MLLT4, MLLT6, or MLLT10 fusion partner in 73.5% of the patients' samples in this study. This included (14/15) ALL, (10/15) AML, and (1/4) MPAL patients . T cells were enriched and expanded for a total of 14 days using a rapid expansion protocol culture system. This approach was successful on all tested samples irrespective of their mutation backgrounds or initial lymphocyte count, with a range of 133-4957fold increases in total cell number and average frequencies of 59.7% and 26.1% for CD4+ and CD8+ T cell, respectively. We further screened the T cell response by examining the expression of 4-1BB, OX40, and IFN-g upon coculture with their autologous leukemia cells, and additionally validated the T cell killing activity against the blasts by a cell killing assay using standard protocol. We found that the expanded T cells exhibited a robust immune activation when cocultured with the autologous leukemia cells from ALL patients with KMT2A rearrangements, indicating an endogenous cell response in all patients in the ALL cohort (100%). However, we observed a considerably reduced reactivity in the samples from the AML and MPAL patients. Interestingly, we observed that such diminished reactivity was associated with a marked downregulation in the expression of MHC class I and II molecules, and that pretreating the leukemia blasts with IFN-γ prior to coculturing with the expanded T cells was sufficient to induce the expression of MHC molecules and subsequent activation of the expanded T cells. Single-cell TCR sequencing of the sorted, tumor-reactive T cells identified the major clonotypes among these reactive T cell clones and the top 4 - 6 clonotypes were selected from 5 patients to evaluate reactivity against the leukemia blasts and examine their specificity against the corresponding driving fusions. We further generated TCR-expressing clones and subsequently cocultured each clone with leukemia blasts to validate their antigen recognition and reactivity by cytotoxicity assay. Interestingly, and in line with the findings from investigations on solid tumors, the leukemia- reactive T cells were consistently characterized by higher expression of both PD1 and CD39 (PD1 hi CD39 +), indicating the capability of utilizing both markers for the enrichment of T cell prior to expansion and subsequent implementation. Despite the high-risk prognosis and low mutation burden in our patient cohort, our findings demonstrate that patient T cells contain endogenous neoantigen-specific clones which can be therapeutically leveraged for targeting leukemia blasts. These results support the utilization of adoptive T cell therapy in the context of hematological malignancies especially given the less restrictive spatial segregation between tumor-specific T cells and leukemia cells compared to solid tumors, where T cell infiltration is a prerequisite for antitumor immunity. Overall, these findings suggest the possibility of leveraging leukemia-associated lymphocytes, fusion oncogene vaccines, and TCR-based therapy as novel immunotherapeutic approaches for mitigating pediatric leukemia, particularly in high-risk patients.