Abstract 417: Defining molecular mechanisms of resistance to glioblastoma immunity using a novel CRISPR/Cas9 in vivo loss-of-function screening platform

Abstract

Abstract OVERVIEW: Despite the impact of cancer immunotherapy, expanding its clinical utility requires a rational method for identifying combination therapies and resistance mechanisms. This is pertinent to glioblastoma (GBM), where initial trials show uncertain response rates. Functional genomic screens used to identify new therapies and escape mechanisms were generally conducted in vitro, where interaction with the immune system is absent. Here we present the development of a high-throughput, in vivo, loss-of-function screening platform for GBM immune-escape mechanisms. METHODS: CT2A- and GL261-astrocytoma cells bearing Cas9-endonuclease were engineered to express a library of barcoded single guide RNAs (sgRNAs). These cells form tumors when implanted intracranially in immunocompetent mice. Tumor-bearing mice were treated with vaccination or PD-1 checkpoint blockade. Dropout of sgRNAs targeting putative immune evasion molecules or enrichment of sgRNAs mimicking resistance mechanisms were detected using next-generation sequencing at the time of tumor implantation and harvest post-immunotherapy. RESULTS: We first developed an in vivo, pooled, loss-of-function genetic screen using Cas9/CRISPR genome editing in mouse transplantable tumors subjected to titratable, selective immune-pressure. Screening 2,400 genes expressed by melanoma cells for those that synergize with or cause resistance to checkpoint blockade recovered known immune-evasion molecules PD-L1 and CD47. Novel immunotherapy targets validated individually, identifying essential pathways of immune-evasion. We then sought to recapitulate this approach in the CNS, and showed that 500-1000 genes can be functionally screened under graded immune-pressure. In optimized immune-competent CT2A and GL261 GBM models, tumors derived from cancer stem cell CD133hi-rich neurospheres were sensitive to immunotherapy and more aggressive and infiltrative than tumors derived from adherent tumor cells. A neurosphere based immune-competent model can be scaled up to a whole-genome screen due to a shorter experimental time requirement and improved engraftment allowing for functional screening of >1000 genes. We curated a GBM-specific library based on differential in vitro and in vivo gene expression profiles of CT2A and GL261 cells exposed to graded immune pressure. We are now screening this GBM-specific library in our optimized in vivo pooled loss-of-function genetic screen using immune-pressure titration to identify novel immunotherapy targets in GBM. CONCLUSIONS: This assay provides the first high-throughput method for systematically identifying resistance mechanisms and new candidate targets for immunotherapy in CNS tumors. Our optimized model could be scaled up to whole-genome loss of function screens, serving as an important tool for identification of next-generation and combination immunotherapies. Citation Format: Martha R. Neagu, Robert T. Manguso, Hans Pope, Maria C. Speranza, Gordon J. Freeman, John Doench, Arlene H. Sharpe, William Nicholas Haining. Defining molecular mechanisms of resistance to glioblastoma immunity using a novel CRISPR/Cas9 in vivo loss-of-function screening platform [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 417. doi:10.1158/1538-7445.AM2017-417