Unique amino acid sequences at the junctions of fusion proteins translated from chimeric RNAs form neoantigenic peptide regions that can be developed to expand the potential repertoire of targets for therapeutic vaccines. We present a framework to identify immunogenic neoantigen candidates from fusion transcripts that can serve as targets for developing vaccines for the treatment and prevention of cancer. Our platform is based on two pipelines. The first rescues discordant paired-end reads discarded from high-throughput sequencing to build a database of chimeric RNAs from cancer samples. Major Open Reading Frames (ORFs) predicted from the RNA fusions from the database are processed through MHC binding predictor (MHCnuggets), a high-throughput MHC Class I and II neoantigen binding prediction program developed by Karchin et al. 2020. The second pipeline utilizes both well-characterized and rare MHC alleles to generate immunogenic neopeptides rank-ordered based on binding affinity measurements from in vitro experiments (half-maximal affinity or IC50). We present 20 novel fusions from 75 breast tumors, each from 3 subtypes TNBC, HER2+, and HR+. We also present a 3833 bp chimeric RNA resulting from readthrough transcription of a pseudogene into a gene located immediately 3’ followed by transplicing between exons 12 and 2. A total of 15 different 8-mer neoantigen peptides discovered from the fusion were predicted to bind to 35 unique MHC class I alleles with a binding affinity of IC50<500nM. All 15 peptides were assessed through an in vitro Enzyme-Linked Immunospot (ELISpot) assay and tested for CD8+ T cell response. The peptides determined to have the highest immunogenicity through the ELISpot Assay can serve as targets for developing tumor vaccines for breast cancer.