Mechanistic Dissection of Regulators of Cancer Plasticity Using High-Content Optical Pooled Screens at Clonal Resolution

Abstract Cancer treatment failure is often attributed to two processes: selection for pre-existing resistant cells and cellular plasticity that allows drug-sensitive cells to become resistant. Plasticity is a hallmark of cancer, but we still lack a clear definition and a mechanistic understanding of how it is controlled in tumors. These gaps are especially problematic in pancreatic ductal adenocarcinoma (PDAC), where epithelial-to-mesenchymal plasticity (EMP) is thought to drive metastasis, tumor initiation, and resistance to therapy. Causally dissecting plasticity requires large-scale perturbation studies that systematically target many genes and measure how they reshape cell-state dynamics at single-cell resolution over weeks to months. Such studies must follow enough cells to observe rare state transitions and clonal heterogeneity, yet current methods force a tradeoff between (i) sufficient molecular phenotyping, (ii) temporal duration, and (iii) cellular throughput, limiting our ability to map regulators of plasticity at scale. We set out to address both issues by (1) developing a screening platform that overcomes major limitations in existing paradigms to (2) enable discovery of the mechanistic underpinnings of cellular plasticity in pancreatic cancer. Using an engineered patient-derived PDAC model, we performed a paired single-cell CRISPRi transcriptomic screen (Perturb-seq) and high-throughput imaging-based optical pooled screen (OPS) of EMP across a panel of chromatin and transcriptional regulators. We identified factors responsible for distinct plastic behaviors by quantifying state transitions from a FACS-defined initial state linked to high-content imaging phenotypes for 150,000 clones, across 1000 genetic perturbations and 22 million cells after two weeks of sustained perturbation in culture. Crucially, this framework sorted regulators into distinct functional classes of plasticity control, including “maintenance” genes stabilizing the starting state, “transition” genes that biased state switching in one direction, and “catalytic” genes that regulated switching bidirectionally. Our methodology recapitulated many genes previously implicated in EMP and uncovered novel candidate “catalytic” targets that converge on H3K9me3-associated epigenetic reprogramming. This work clarifies how cell state is regulated at clonal resolution, identifies potential vulnerabilities in plastic tumor cells, and establishes a platform that can be extended to studies of cellular plasticity in other biological contexts. Citation Format: Mikolaj Godzik, Russell Walton, Michael Bogaev, Lynn Bi, Julien Dilly, Martin Jankowiak, Elisa Donnard, David T. Ting, Nir Hacohen, Eric S. Lander, Paul Blainey, Arnav Mehta. Mechanistic dissection of regulators of cancer plasticity using high-content optical pooled screens at clonal resolution [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 6798.