Abstract 6519: <i>In silico</i> discovery of fusion-specific RNA biomarkers in pediatric leukemias

Abstract Purpose: Pediatric leukemia remains a leading cause of cancer-related mortality in children, with fusion-driven subtypes such as KMT2A-rearranged and ETV6-RUNX1 leukemias accounting for a significant proportion of high-risk cases. Despite improvements in overall survival, fusion-positive leukemias are associated with poorer outcomes, suboptimal therapeutic response, and elevated relapse risk. A persistent challenge in pediatric oncology is delayed relapse diagnosis. Although minimal residual disease testing and bone marrow aspiration remain standard for diagnosis and monitoring, these approaches are invasive, costly, and often fail to capture dynamic oncogenic changes. Oncogenic chimeric transcription factors (OCTFs), formed by abnormal gene fusions, are key molecular drivers in many pediatric leukemias. Emerging transcriptomic data suggest that OCTF-driven tumors produce unique noncanonical RNA transcripts termed neogenes which may be highly cancer-specific and represent a promising new class of RNA biomarkers. The purpose of this study is to identify and characterize neogene transcripts in fusion-driven pediatric leukemias with the goal of evaluating their potential as subtype-specific RNA biomarkers for early detection and disease monitoring. Methods: We assembled a modular transcriptome pipeline to identify neogene candidates in fusion-driven pediatric leukemias. Publicly available RNA-seq datasets were aligned to the human reference genome using STAR (Spliced Transcripts Alignment to a Reference), followed by transcript assembly with Scallop. Unannotated transcripts were identified via Gffcompare and quantified post-normalization. Candidate neogenes were filtered based on novelty, cross-sample recurrence, and biological plausibility. Parallel quantification of non-leukemic fusion-positive and fusion-negative cancers and normal tissues were performed to assess for neogene specificity. Results: Seven pediatric leukemia subtypes were profiled: AML, AMoL, APL, CML, T-ALL, and preB/proB-cell leukemia. Preliminary analyses reveal that each subtype exhibits a distinct neogene signature not observed in non-leukemic fusion-positive cancers (Ewing sarcoma, rhabdomyosarcoma, and Desmoplastic Small Round Cell Tumor) or in common fusion-negative pediatric malignancies (neuroblastoma, nephroblastoma, and hepatoblastoma). Chromatin accessibility and histone modification markers at neogene loci provide evidence for active transcription and supports biological relevance of transcripts. Conclusion: In silico findings support the existence of subtype-specific neogene signatures in fusion-driven pediatric leukemias. These transcripts may serve as novel biomarkers for early detection, subtype classification, and treatment monitoring. Further validation is underway to assess clinical utility and translational potential. Citation Format: Elizabeth Tsuying Chang, Sean Lee. In silico discovery of fusion-specific RNA biomarkers in pediatric leukemias [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 6519.