NEW YORK – A Genentech team has demonstrated a method for teasing out treatment response and resistance mechanisms for clonal cancer cell populations that uses single-cell transcriptome sequencing and clonal fitness mapping.
"Current paradigms to characterize drug response and resistance depend heavily on endpoint assessment after prolonged drug exposure," co-corresponding and co-senior authors Xin Ye and Scott Foster, researchers with Genentech's discovery oncology department, and their colleagues wrote.
"While this process can be effective to identify mechanisms of resistance, it is limited in its ability to uncover pre-existing features and early adaptive changes that enable specific subpopulations to persist," they explained. "We reasoned that a system that simultaneously tracks the origin and compares the immediate responses/fate of tumor cells to different therapies could greatly accelerate drug response/resistance studies."
For a study published in Nature Biotechnology on Wednesday, the researchers followed clonal responses to treatment in hundreds of epidermal growth factor receptor (EGFR)-mutated lung cancer cells using a single-cell RNA sequencing method known as "tracking differential clonal response scRNA-seq," or TraCe-seq. The cells were treated with distinct EGFR-targeting therapies, including conventional kinase inhibitors that interfere with enzyme binding and more novel treatments that involve target protein degradation.
The approach relies on a combination of diverse cellular barcoding and sequential single-cell sequencing, the team explained. While the barcoding step makes it possible to follow clonal fitness in expanded, untreated cell cultures and in subsets of this culture exposed to different drugs, the scRNA-seq profiles provide a peek at the expression changes behind these cell responses.
"[W]e developed TraCe-seq to rapidly capture the origins and early adaptive processes underlying therapy response by simultaneously measuring clonal fitness and their transcriptional trajectories in a heterogeneous population subject to anti-cancer treatments," the authors explained, arguing that the approach "enables direct and comprehensive comparisons of different therapeutic modalities at subpopulation and single-cell resolutions and provides insights into pre-existing transcriptional features that dictate drug response and subsequent resistance."
After assessing TraCe-seq in several cell types, the investigators applied the method to EGFR-mutant lung cancer cells treated for four days with the EGFR inhibitor erlotinib (Genentech's Tarceva), an EGFR inhibitor/degrader compound called GNE-104, or a similar compound called GNE-069 that does not degrade the protein.
Based on barcode diversity patterns in pre- and post-treatment cells, along with 10x Genomics scRNA-seq profiles for thousands of untreated cells or cells exposed to each compound, they saw more pronounced post-treatment fitness in the lung cancer cells treated with the EGFR inhibitor/degrader GNE-104 compared with the cells treated with erlotinib or GNE-069, despite a dip in MAP kinase pathway activity in EGFR-mutant lung cancer cells exposed to all three drugs. Their findings suggested that targeted EGFR degradation resulted in more modest anti-cancer growth effects than more conventional inhibitor compounds.
"Our results suggest that targeted degradation is not always superior to enzymatic inhibition and establish TraCe-seq as an approach to study how pre-existing transcriptional programs affect treatment responses," the authors wrote.
The team delved into treatment response differences — as well as the transcriptional features found in cells that were vulnerable to treatment and in treatment-resistant cells — with corresponding expression data from treated and untreated cells subjected to scRNA-seq. When it came to EGFR inhibitor response, for example, the results highlighted a role for endoplasmic reticulum stress, offering clues for developing future treatment approaches.
Moreover, the authors suggested that a similar strategy may be used to interrogate treatment responses in other cancer cell types and therapeutic settings.
"We anticipate that the TraCe-seq approach will guide the development of future therapies by revealing unknown features that predict response and resistance to different molecular modalities or treatment combinations," the authors wrote, noting that "the TraCe-seq barcoded population could be sampled at multiple time points to gather transcriptional information with clonal resolution over time, thus providing a generally applicable approach for studying the evolution of heterogeneous populations under a variety of contexts."