Center for Cancer Systems Therapeutics (CaST)
Project 3: Developing novel methodologies for the systematic prioritization of compounds and compound combinations capable of abrogating tumorigenesis in vivo, by targeting mechanisms that induce irreversible collapse of tumor checkpoints presiding over the stability of cancer cell state.
Tumor checkpoints are generally implemented by the concerted activity of a handful of master regulator (MR) proteins. The overarching goal of this project is to develop and validate computational methods capable of prioritizing MR-targeting drugs and drug combinations that either implement or prevent specific tumor state transitions. This includes inducing irreversible commitment to cell death, preventing progression to a malignant tumor stage, and rescuing drug sensitivity.
Master regulators as therapeutic targets
Although master regulators are rarely mutated or differentially expressed, they are enriched in transcriptional regulators, making them a potentially novel class of therapeutic targets for tumors that either lack targeted inhibitors or are considered undruggable. Our preliminary data show that MR proteins can be used effectively as universal, multiplexed gene reporter assays to prioritize compounds and compound combinations that induce tumor checkpoint collapse, both in vitro and during tumor regression in vivo. Compounds emerging from these analyses can then be further validated and mechanistically characterized, using network-based approaches.
By comparing tumor checkpoints identified in regulatory models of cancer with perturbation profiles of small molecule compounds, it becomes possible to identify drugs capable of abrogating the activity pattern of cell state master regulators or of activating master regulators of cell death phenotypes.
This general approach requires the assessment of tumor checkpoint dependencies in patient tumor tissue. By analyzing RNASeq data following compound perturbations, we can identify compounds whose mechanism of action (MoA) affects specific MR proteins. The best way to infer compounds that induce tumor checkpoint collapse is to assess cellular response, by RNASeq profiling, at multiple sublethal, physiologically relevant concentrations.
Because such an undertaking would be cost-prohibitive using standard RNASeq technology, we will utilize an innovative technology called PLATESeq, developed in the Sims laboratory and provided by the Molecular Profiling Core (MPC) . By pooling the sequencing of multiple samples in the same experiment, PLATESeq reduces costs dramatically, to just $32/profile. Preliminary data show that PLATESeq produces virtually no reduction in accuracy, compared to Illumina TrueSeq 30M RNASeq, which costs 20-fold more. As a result, it dramatically increases the range of compounds that can be effectively screened.
Heterogeneity of tumor response to MR-targeted therapy
Project 3 is addressing a broad range of questions related to the heterogeneity of tumor response and to the mechanisms that allow tumor cells to compensate for MR-targeted therapy. For instance, we have already identified compounds capable of inverting tumor checkpoint activity at 6 hours following administration, with tumor checkpoint activity completely restored at 24 hours. Such changes are too fast to be the result of genetic selection or epigenetic reprograming, suggesting that MR-targeting could be uniquely suited to time-course based analysis to elucidate the specific adaptive mechanisms responsible for this behavior.
In addition, we will compare the mechanisms responsible for induced resistance following the targeting of tumor checkpoints to the mechanisms involved in traditional oncogene-targeted therapy (see project 2). To do so, we will compare the activity of inducible RNAi/CRISPR inhibition of tumor checkpoint MRs in vivo with the activity of compounds and combinations prioritized by the analysis.
We expect that if resistance is induced by the emergence of bypass mutations, we will observe a failure to maintain tumor checkpoint MR protein inhibition over time. Conversely, if drug resistance emerges because of tumor state reprograming, we will observe a complete shift in tumor checkpoint MRs, revealing one or more novel, stable states of the cancer cell. We propose that addressing tumor plasticity and the potential escape routes implemented by tumor state reprograming is going to be critically relevant to assessing therapeutic strategies for the chronic management of cancer in patients.