RNA-based Personalized Cancer Vaccine for AML immunotherapy

Abstract Acute Myeloid Leukemia (AML) is a rapidly progressing malignancy relying primarily on chemotherapy as the frontline treatment option. Success of personalized neoantigen vaccines in melanoma patients have garnered great attention in the field of immunotherapy. We hypothesize that personalized cancer vaccine, in the form of mRNA, can be a viable treatment strategy for AML patients. Next generation sequencing was performed on both C57BL6Jinv mouse’s and C1498 cell line’s nucleic acid materials. Three different neoepitope prediction pipelines were utilized to identify potential immunogenic neoepitopes. Candidate neoepitopes within intron polyadenlyation sites (iPAS), retained intron (RI) sites and somatic mutation (SM) sites were found. A total of 285 SM-derived neoepitopes were identified. Top 38 highly expressed genes were selected for further analysis. Using Sanger sequencing, we confirmed 37 out of the 38 mutations. A total of 5 iPAS-derived neoepitopes were identified. A total of 26 RI-derived neoepitopes were identified in our first RNAseq analysis and a separate 476 RI-derived neoepitopes were identified in our second RNAseq analysis. With these neoepitope candidates, we narrowed down our selection and did a preliminary screen for immunogenic epitopes using a total of 71 synthetic peptides in an ELISpot assay. 10 neoepitope candidates were short-listed for further in-vivoevaluation. mRNA vaccine was synthesized and evaluated in a murine AML model. T cell responses against the 10 neoepitopes were analyzed to provide pre-clinical evidence for patient studies. Neoepitope screening platform with human PBMCs has also been established to validate the prediction pipelines in AML patients. Supported by grant from NMRC (MOH-CIRG21jun-0008)