TMOD-12. ESTABLISHING A CLINICALLY RELEVANT MODEL OF MESENCHYMAL GLIOBLASTOMA (GBM) TO STUDY RESPONSE TO STANDARD OF CARE TREATMENT AND IMMUNE CHECKPOINT INHIBITION (ICI)

GBM is a devastating disease with peak incidence in the seventh decade. Pre-clinical models are essential for studying resistance mechanisms and screening novel therapies. However, historically these models have failed to predict response in humans. Current models seldom incorporate surgical resection, and commonly use young animals whose immune contexture differs from older patients. Here, we have established an orthotopic model employing the syngeneic mesenchymal-NFpp10a-cell line which incorporates surgical resection in aged mice. We further characterise response to ICI and temozolomide monotherapy. NFpp10a and GL261-cell lines were exposed in vitro to irradiation (0Gy/2Gy/5Gy) and response assessed using colony formation assays. NFpp10a formed significantly more colonies at 5Gy compared to GL261 at both day 10 (NFpp10a 5.167 vs GL261 1.4; p=0.0017) and day 14 (NFpp10a 3.5 vs GL261 0; p< 0.0001). Hence, NFpp10a displays increased radioresistance. Next, NFpp10a-luciferin expressing cells were orthotopically implanted into young (6-8weeks;n=16) and aged (18months;n=16) C57BL/6-mice. Weekly bioluminescence imaging (BLI) was performed to monitor growth. Mice undergoing resection showed a median 18.47-fold drop in BLI signal. We demonstrated resection survival advantage in aged mice (Resection:33.5 days vs Non-Resection:18 days, p= 0.0166) and showed young age to be a positive prognostic factor (Young:62 days vs Aged:22 days, p=0.0002). Subsequently, we orthotopically implanted NFpp10a-Luc2 cells into C57BL/6 mice and treated with temozolomide (n=24) or PBS control (n=23), and anti-PD1 (n=24) or IgG (n=23). We observed that temozolomide and anti-PD1 monotherapy had no impact on NFpp10-Luc2 growth (temozolomide-overall:p=0.9001, anti-PD1-overall:p=0.7933) or survival (temozolomide-overall:p=0.3035, anti-PD1-overall:p=0.6328). Overall, we have established an NFpp10-Luc2 mesenchymal-GBM model in aged mice which incorporates surgical resection and accurately displays significant resistance to temozolomide and anti-PD1 monotherapy. We are currently employing this model to study the efficacy of neoadjuvant anti-PD1 therapy. Mechanistic analyses with multiplex-immunohistochemistry, scRNA and whole exome sequencing are planned to interrogate treatment effects on the tumor microenvironment.