Abstract B021: Single-cell lineage tracing to reveal ecDNA dynamics in glioblastoma through treatment

Abstract Extrachromosomal DNA (ecDNA) has emerged as a major mechanism of dynamic oncogene amplification, contributing to intratumoral heterogeneity, treatment resistance, and poor clinical outcomes across multiple cancer types. This is especially pertinent in glioblastoma (GBM), where more than half of patients have ecDNA at diagnosis. Despite its clinical importance, the biological factors governing ecDNA copy number dynamics and inheritance across cell divisions remain poorly understood, limiting our ability to predict or therapeutically target ecDNA-driven tumor evolution. To address this gap, we developed a CRISPR/Cas9 base-editing lineage-tracing platform that enables high-resolution tracking of ecDNA molecules across GBM cell generations. By introducing heritable, multi-state edits into both ecDNA and chromosomal loci, we generate dense lineage information to reconstruct cellular phylogenies alongside the inheritance histories of individual ecDNA molecules. With this system, we achieve multiple independent lineage integrations per cell, shared between ecDNA molecules and genomic integrations, providing sufficient resolution to jointly model cell-level and ecDNA-level evolutionary dynamics. Applying this approach to GBM models subjected to ecDNA-targeted treatment (erlotinib) or radiation, we reveal how therapeutic pressures reshape ecDNA clonal structure, impact patterns of ecDNA segregation, and drive selective restriction or expansion of ecDNA-bearing lineages. Along with ecDNA lineage, our system enables recovery of transcriptomic information from the same cells. We observe shifts in cell state towards AC-like state in response to erlotinib or MES-like state in radiation treatment. Furthermore, ecDNA dynamics shifted: cells exposed to erlotinib reduced ecDNA, whereas irradiated cells initially maintained higher ecDNA levels. Transcriptomic differences were also detected between cells with higher or lower ecDNA levels, suggesting pathways for ecDNA maintenance. Together, this work establishes a scalable framework for directly observing ecDNA inheritance and evolution in cancer, providing new insights into how therapeutic interventions influence ecDNA-driven tumor heterogeneity and adaptation. By directly tracing the inheritance and evolution of ecDNA at single-molecule resolution, this work reveals how standard cancer therapies actively reshape ecDNA-driven clonal architecture and cellular states, providing a new framework for understanding and ultimately targeting ecDNA-mediated treatment resistance in GBM. Citation Format: Abigail C. Marshall, Aaron McKenna. Single-cell lineage tracing to reveal ecDNA dynamics in glioblastoma through treatment [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Brain Cancer; 2026 Mar 23-25; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2026;86(6_Suppl):Abstract nr B021.