Identifying therapeutic targets for recurrent glioblastoma using patient-derived longitudinal models
Recommended Citation
Datta I, Nuga O, Berezovsky A, Poisson L, deCarvalho A. Identifying therapeutic targets for recurrent glioblastoma using patient-derived longitudinal models. Neuro Oncol 2025; 27(Supplement 5):v356.
Document Type
Conference Proceeding
Publication Date
11-11-2025
Publication Title
Neuro Oncol
Keywords
phenotype, radiation therapy, gene expression, genetic transcription, mutation, glioblastoma, dna, dna methylation, genome, interferons, nude mice, epidermal growth factor receptors, rna sequence analysis, heterologous transplantation, mice, neoplasms, temozolomide, central serous chorioretinopathy, human leukocyte interferon, tumor growth, pten gene, nervous system development, amplification, cancer stem cells, epithelial to mesenchymal transition, cyclin-dependent kinase 4, whole exome sequencing, whole genome sequencing, crisis standards of care
Abstract
Recurrent GBM poses significant treatment challenges due to limited therapeutic options and reliance on primary tumor molecular profiles, often overlooking treatment-induced molecular and phenotypic changes in recurrent tumors. This study investigates genomic, DNA methylation, and transcriptomic differences between a primary GBM (HF3016) and its matched recurrent tumor post-temozolomide and radiotherapy (HF3016R), using neurosphere cultures and patient-derived xenografts (PDXs) to assess radiotherapy (RT) response. Genomic DNA was analyzed via whole-genome sequencing, whole-exome sequencing, and EPIC DNA methylation arrays. Bulk RNA sequencing was performed on HF3016 and HF3016R cancer stem cell (CSC) cultures (n=4) and orthotopic PDX models (n=3). Differential gene expression was analyzed using NOISeq (q=0.95, fold change >2) on TMM-normalized data, with pathway enrichment via Metascape 3.1. Tumor growth and RT sensitivity were evaluated in NCRNU athymic nude mice implanted intracranially with HF3016 or HF3016R cells (3x10^5 cells/mouse, N=9-10 per group), receiving 5 Gy RT daily for 3 days or no treatment. Survival was assessed using Kaplan-Meier curves and Log-rank tests. No genomic driver alterations (CDK4, EGFR, MYC ecDNA amplification, PTEN, p53 mutations) or DNA methylation mesenchymal subgroup changes were observed between primary and recurrent models. However, a transcriptional shift from proneural to mesenchymal subtype occurred in recurrent models, with enriched epithelial-mesenchymal transition and interferon response pathways in HF3016R, while nervous system development pathways were enriched in HF3016. HF3016R PDXs exhibited faster growth (p=0.002) and greater RT resistance (p=0.0021 for HF3016, p=0.061 for HF3016R) compared to HF3016. These findings highlight the importance of considering treatment-induced molecular changes in recurrent GBM for optimizing therapeutic strategies. This study reveals that while genomic and DNA methylation profiles remain stable between primary and recurrent GBM models, significant transcriptional reprogramming occurs. The enrichment of epithelial-mesenchymal transition and interferon response pathways in recurrent models is associated with their aggressive growth and increased radiotherapy resistance.
Volume
27
Issue
Supplement 5
First Page
v356
