A heterogenous pharmaco-transcriptomic landscape induced by targeting a single oncogenic kinase

Over-activation of the epidermal growth factor receptor (EGFR) is a hallmark of glioblastoma. However, EGFR-targeted therapies have led to minimal clinical response. While delivery of EGFR inhibitors (EGFRis) to the brain constitutes a major challenge, how additional drug-specific features alter efficacy remains poorly understood. We apply highly multiplex single-cell chemical genomics to define the molecular response of glioblastoma to EGFRis. Using a deep generative framework, we identify shared and drug-specific transcriptional programs that group EGFRis into distinct molecular classes. We identify programs that differ by the chemical properties of EGFRis, including induction of adaptive transcription and modulation of immunogenic gene expression. Finally, we demonstrate that pro-immunogenic expression changes associated with a subset of tyrphostin family EGFRis increase the ability of T-cells to target glioblastoma cells. Single-cell RNA-seq libraries were generated using three-level single-cell combinatorial indexing RNA sequencing (sci-RNA-seq) of untreated or small molecule inhibitor exposed A172, T98G, U87MG, BT112, BT228 and BT333 glioblastoma brain cancer cells. Different cells and different treatments were hashed and pooled prior to sci-RNA-seq using the sci-Plex nuclear barcoding strategy. This nuclear barcoding strategy relies on fixation of barcode containing well-specific oligos that are specific to a given cell type, replicate, or treatment condition.