Long-read, single-cell RNA sequencing of high-risk neuroblastoma samples - PROJECT SUMMARY/ABSTRACT Most patients with high-risk neuroblastoma (NB), the most common extracranial solid tumor of childhood, die despite intensive cytotoxic therapies and immunotherapy with dinutuximab, a chimeric monoclonal antibody targeting the disialoganglioside GD2. This underscores the need for new immunotherapeutics, with GPC2 emerging as one promising target for chimeric antigen receptor (CAR) T cells and antibody-drug conjugates (ADC). Nevertheless, alternative targets will undoubtedly be needed to safeguard against intra- and intertumoral heterogeneity and epitope loss. Given the limitations of proteomic approaches, such targets are likely to emerge from sophisticated transcriptomic analyses. The Gabriella Miller Kids First Pediatric Research Program (GMKF) has already generated high-quality bulk, short-read RNA-seq datasets corresponding to high-risk NB. The value of this unique resource could be further increased by accounting for widespread variations at the level of mRNA processing, including alternative splicing. Focusing on surface proteins as potential CAR/ADC targets, we compared and contrasted splicing patterns in 150+ NB samples vs adrenal gland (where NB often originates) and filtered for highly expressed genes and in-frame events affecting extracellular (or ecto-) domains. We identified dozens of local splicing variations (ectoLSVs) giving rise to potentially targetable neo-epitopes. Notably, many of these ectoLSVs mapped to microexons, very short (<51 nt), conserved, dynamically regulated cassettes, which account for much of the surfaceome diversity in the nervous system. As rigorous and reproducible as they are, our preliminary RNA-seq have two major limitations. The first limitation reflects the inherent shortcomings of short-read technologies, which are not very good at resolving complex mRNA splicing events. This limitation can be overcome by long-read RNA-seq, as implemented in the PacBio and Oxford Nanopore Technologies platforms. However, as commonly utilized, these platforms do not offer single-cell resolution capabilities. The second major limitation of bulk RNA-seq is the paucity of insights into tumor heterogeneity and evolutionary trajectories of individual clones. The goal of this proposal is to overcome these limitations by first performing bulk long-read RNA-seq on 20 high-risk NB patient samples and patient-derived xenografts (PDXs). Then single-nuclei isoform RNA sequencing will be performed on a subset of these specimens. Prioritizing microexon-related events will help filter out aberrantly spliced transcripts unlikely to yield functional proteins and elucidate alternative splicing events with possible roles in NB pathogenesis. Therefore, upon completion of this 2-year project we will be in the position to a) integrate all the long-read sequencing data (bulk and single-cell) with the GMKF ecosystem; and b) nominate additional non-canonical proteoforms, for immunotherapy with CAR T cells and ADCs.