Longitudinal Single-Cell RNA Sequencing of Patient-Derived Primary Cells Reveals Drug-Induced Infidelity in Stem Cell Hierarchy

Chemoresistance is one of the major causes of cancer-related deaths. Here we used single-cell transcriptomics to investigate the modes of chemoresistance in patient-derived tumor cell populations that display varying degrees of phenotypic heterogeneity. We observed that higher degree of intratumor heterogeneity (ITH) favors selection of pre-existing drug-resistant cells. However, surprisingly we found that phenotypically homogeneous (epithelial) cell populations can engage covert epigenetic mechanisms to adapt under drug selection, leading to the de novo emergence of new (mesenchymal) cell states. We observed pre-existing poised chromatin marks on genes associated with mesenchymal cell-states in naïve cells and the gain of activation marks (H3K27ac) during the acquisition of resistance and metastasis. Further mechanistic interrogation revealed that drug-induced adaptation was acquired upon the loss-of stem-factor Sox2, and a concomitant gain of Sox9 expression. Strikingly we observed an enrichment of Sox9 at drug-induced de novoH3K27ac sites, suggesting that tumor evolution could be driven by stem cell-switch-mediated epigenetic plasticity. Most importantly, we validated the phenomenon of drug-induced cell states and stem cell-switch observed in our “in plateevolution” models in matched patients, as well as prospective patient cohorts, thereby emphasizing the prognostic power of the single cell approach on patient-derived models to predict tumor evolution in the clinic. Additionally, we also demonstrate that JQ1-mediated BRD-4 inhibition could reverse the drug-induced epigenetic plasticity, and therefore the emergence of resistance. Altogether our results provide novel mechanistic insights into the modes of tumor evolution and implicates the use of combinatorial epigenetic therapies to target the emergence of chemoresistance.