Co-culture of patient-derived glioblastoma cell lines with autologous pbmcs for the investigation of oncolytic virus response

Abstract BACKGROUND The dismal prognosis of glioblastoma (GBM) patients, with a median survival of less than 15 months despite maximal therapy urgently warrants new therapeutic approaches. Clinical trials employing oncolytic viruses (OVs) have shown encouraging results, however, in each OV clinical trial only a small subset of patients responded to treatment. As inter-tumoral heterogeneity has been the key challenge in treating GBM, we hypothesized that development of an in vitro co-culture model for assessment of viral replication and subsequent immune response might ultimately predict in vivo OV efficacy for individual GBM patients. METHODS 20 patient-derived GBM cell cultures were tested with dose-ranges of two clinically relevant OVs (DNX2401 and rQnestin34.5 V1) to determine the EC50 values (half-maximal effective concentration). To discriminate between responders and non-responders six of these cultures were infected using MOI 10 for DNX2401 and 0.25 for rQnestin34.5V1, and co-cultured with autologous PBMCs. OV-induced changes in gene and protein expression of immune associated genes were assessed using targeted gene expression (NanoString Technology) and ELISA. RESULTS DNX2401 EC50 values ranged from 0.35 to > 600 and from 0.04 to 1.77 for rQnestin34.5V1. Induction of pro-inflammatory cytokines and chemokines differed per virus and per patient. Enhanced OV infection or oncolysis efficiency did not lead to increased levels of IFNγ production. Importantly, longer exposure to dexamethasone prior to PBMC isolation correlated with suppressed IFNγ production. CONCLUSION We established an autologous GBM cells/PBMCs co-culture model which reflects inter-tumoral heterogeneity in terms of cytokine induction and virus-specific changes in gene and protein expression upon OV infection. Immune activation is not directly related to degree of infection or oncolysis and can be hampered by prolonged prior dexamethasone use. These first results support our hypothesis that improved prediction of response rates in oncolytic virotherapy for GBM may require a personalized approach with an in vitro test system.