Abstract The most deadly, primary brain tumor, glioblastoma (GBM), is driven by various genetic mutations that can differentially affect the tumor microenvironment and treatment response. Here we present murine models developed autochthonously using mosaic analysis with dual recombinase-mediated cassette exchange (MADR) combined with CRISPR-Cas9-mediated loss-of-function (LOF) of common patient tumor mutations. This molecular toolkit allowed us to easily model tumors driven by the most common mutation of epidermal growth factor receptor (EGFR), EGFRvIII, with different LOF mutation combinations of PTEN, CDKN2A, and NF1. Standard of care treatment of radiation therapy and temozolomide were used to identify how tumors with EGFRvIII and Nf1 mutations (NF1 group) responded compared to tumor without Nf1 mutations (non-NF1 group). Initial assessment identified significantly increased survival in treated NF1 compared to non-NF1 groups (85. 5 vs 59. 5 days, respectively; P 0. 01), with median survival doubled in NF1 treated versus control groups (24. 5 days), compared to non-NF1 treated versus control groups (12. 5 days). Interestingly, use of cBioPortal identified a cohort of patients with both EGFR and NF1 aberrations that had significantly increased progression-free survival, suggesting our models more accurately represent human GBM behavior. We therefore hypothesized that patients with NF1 LOF have enhanced responses to treatment leading to longer survival. With the dearth of acute post-treatment patient brain tumor samples, we utilized our murine modeling system to look at the acute effects of treatment. Histological examination showed both non-NF1 and NF1 tumors had decreased tumor volume at 7 days post-treatment, with the latter being more pronounced. In agreement with this observation, percentage of Ki-67+ proliferating cells was significantly less in the NF1 group compared to non-NF1 group at both 3 and 7 days post-treatment (P 0. 001 and P 0. 01, respectively). GBM patient tumor with NF1 LOF also identifies as having a divergent myeloid presence compared to other GBM subtypes. Single-cell RNA sequencing of acutely treated murine tumor samples identified differences in myeloid presentation between non-NF1 and NF1 groups. While non-NF1 treated tumor presented with higher proportions of both homeostatic microglia and peripheral macrophages at D3 post-treatment, levels were drastically lower at D7. NF1 treated tumor presented with the opposite, with low proportions of homeostatic microglia and peripheral macrophages at D3 that subsequently increased at D7. Additionally, immunofluorescent expression of the innate immune sensor TLR2 was considerably greater in NF1 treated tumor compared to non-NF1 tumor. Taken together, the data above indicate an increased inflammatory and peripheral macrophage immune presence, along with decreased tumor cell proliferation in NF1 tumor with treatment that may explain the increased survival for this group and help identify additional treatment strategies. Citation Format: Katie B. Grausam, Emily A. Hatanaka, Antonio C. Fuentes-Fayos, Eddie Lu, Sara Versales, Stephen L. Shiao, Joshua J. Breunig. Divergent treatment responses in genetically distinct EGFR-mutant models of glioblastoma 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 (6Suppl): Abstract nr B014.
Building similarity graph...
Analyzing shared references across papers
Loading...
Katie B. Grausam
Emily A. Hatanaka
Antonio C. Fuentes-Fayos
Cancer Research
Cedars-Sinai Medical Center
University of Córdoba
Building similarity graph...
Analyzing shared references across papers
Loading...
Grausam et al. (Mon,) studied this question.
www.synapsesocial.com/papers/69c37af0b34aaaeb1a67ce88 — DOI: https://doi.org/10.1158/1538-7445.brain26-b014