Combinatorial drug treatment has been proposed as a strategy to overcome rare resistant clones. However, rapid progress in drug discovery and the numbers of possible combinations make it challenging to find suitable combinations. In this talk, we demonstrate how high-throughput (HT) single cell gene expression and multiomic analysis coupled with cell multiplexing enables high-throughput screening of multiple combinatorial conditions. Non-small cell lung cancer (NSCLC) cell lines are great models for a combinatorial drug screen as they possess multiple forms of resistant clones. Taking the advantage of the scale of Chromium Single Cell Gene Expression HT assay, we analyzed a total of 192 samples, or ~960,000 cells on a single microfludic chip. UMAP projection of all of the treatment and time point conditions showed that H1975 cells, bearing a EGFR mutation, are more responsive to combinatorial drug treatment than that of A549 cells, bearing a KRAS mutation. The cell cycle and DNA repair pathways were down-regulated as early as 4h and continued until 24hr. The main affected gene network by the combinatorial treatment are cell cycle checkpoints and DNA double-strand break repair. We see less effect in the single treatment conditions, suggesting the regulation of the networks is a result of synergistic regulation of multiple pathways. In a separate study, we profiled single cell transcriptomes and surface protein markers from 7 multiplexed primary NSCLC cell lines. Single cell multiomic studies are easily carried out with the HT platform. By adding protein detection alongside single cell transcriptome analysis, cell types can be better inferred, making cell annotation less challenging. The scale of HT provides an opportunity for identifying low occurrence tumor types, e.g., ELF3+/EPCAM+ tumor. The identification of rare tumor types is key to understanding evolution of drug resistance and metastasis clones. This kind of study will shed light on therapeutic discovery of actionable targets. Finally, we found the common signaling pathways shared by the primary NSCLC models and A549 cell line. This suggests that A549 would be a great option to perform future large-scale screening for the KRAS-mutated type of NSCLC. Overall, this study highlights the scalability of single cell approaches using Chromium Single Cell HT, and the applicability of multiomic technologies on the HT platform to enable the high resolution study of biology.
.jpg "Funien Tsai, PhD photo")