ATIM-48 (LTBK-05). MULTIDIMENSIONAL PERSONALIZED RESPONSE ASSESSMENT TO MICROGLIA MODULATORS IN GLIOBLASTOMA BIOREACTORS

Abstract BACKGROUND Recently, strategies harnessing the non-neoplastic immune tumor microenvironment (iTME), consisting of macrophages and microglia (TAMs), as well as adaptive immune cells have been employed to treat glioblastoma (GBM). To evaluate the effect of local TAM-modulating therapies in combination with T-cell checkpoint inhibitors, we generated 3D GBM perfusion bioreactor cultures from patient-derived samples. We report patient- and tumor region specific responses to microglia modulators and checkpoint inhibitors using multidimensional fluorescent microscopy techniques, and multiplexed cytokine measurements, Subsequently, we aime at identifying responders versus non-responders as well as predictive markers of treatment response. METHODS Fresh, neuronavigated GBM biopsies from tumor center and periphery were placed into perfusion bioreactors and cultured for 7 days. Explants were treated with combinations of TAM and T-cell modulating drugs including anti-PD1 and anti-CD47 antibodies and their combination. Tissue was harvested for histology, and supernatants were processed for multiplexed cytokine analysis. Multidimensional CODEX technology analysis using a customized TAM centric 54 marker panel was implemented, and a map of individualized response criteria to specific immunotherapies developed. RESULTS Multiplex cytokine analysis showed a dominance of proinflammatory cytokines (CCL2, CCL3, CCL4 and PAI-1) in the periphery of the tumor at baseline. We further observed that the tumor periphery was more responsive to treatments confirming the efficacy of the treatment after tumor resection. Using CODEX, we identified specific cell types responding to the treatment and undergoing phenotypic changes. Moreover, dynamic shifting of T-cell checkpoint expression levels under treatment pointed to potential resistance mechanisms in a subset of tumors. Further, we identified region-specific cytokine release as a response to the treatment in a series of 8 patient-specific explant cultures. In summary, we present an in-depth profiling of the GBM-region specific iTME at baseline and document its dynamic response under innate/adaptive immune modulators using CODEX. CONCLUSION The proposed approach serves as a patient-tailored ex vivo “Clinical Trial” by stratifying the individual patient’s iTME response.