Defining and targeting drivers of lineage plasticity in stem cell-like prostate cancer

Prostate cancer depends on androgen receptor (AR) signaling for growth, which is why androgen deprivation (castration) therapy is effective at early stages. However, many tumors eventually progress to a lethal form known as castration-resistant prostate cancer (CRPC). A subset of CRPC tumors bypass dependency on AR signaling by acquiring lineage plasticity, where prostate cancer cells transdifferentiate into alternate cellular states through epigenetic reprogramming. Neuroendocrine (NE) prostate cancer represents one well-known lineage plasticity phenotype. Nevertheless, most AR-independent tumors do not exhibit NE features and are defined as AR-negative/NE-negative or “double-negative prostate cancer” (DNPC). In a collaboration with Dr. Ekta Khurana's computational genomics lab at Weill Cornell Medicine, we recently classified CRPC into four epigenetic subtypes, including the well-established 1) AR and 2) NE, as well as the novel DNPC subgroups 3) WNT and 4) stem cell-like (SCL) (PMID: 35617398). We focused on the SCL subtype as it is the second most common group in CRPC patients and lacks therapeutic targets. Using functional genomic approaches, we found that YAP/TAZ/TEAD cooperates with FOSL1 to drive the SCL lineage and growth of SCL models. We therefore hypothesize that the heightened dependency on the YAP/TAZ/TEAD/FOSL1 transcriptional program represents a therapeutic vulnerability in CRPC-SCL. To test this, we exposed CRPC models to TEAD inhibitors and found robust growth suppression in SCL cells compared to non-SCL cells in vitro. To evaluate whether the TEAD inhibitors are on-target, we performed transcriptomic profiling in SCL models and observed downregulation of YAP/TAZ gene signature as well as FOSL1 expression, which phenocopies the effects of YAP/TAZ double knockdown. To define the cistromes of these factors upon TEAD inhibition, we performed ChIP-seq and observed reduced co-occupancy at consensus sites, suggesting the disruption of the YAP/TAZ/TEAD/FOSL1 transcriptional circuit by the small molecule compound. To determine whether these phenotypes are recapitulated in vivo , we will treat mice harboring CRPC-SCL xenografts with TEAD inhibitors to assess growth response and evaluate epigenetic and transcriptional response using single-nucleus Multiome (ATAC+RNA). These studies will establish whether small molecule inhibition of TEAD is a promising strategy for the treatment of CRPC-SCL and allow high-resolution analysis of cell state transitions, with a focus on loss of SCL-specific signatures and potential emergence of AR/NE programs as adaptive resistance mechanisms.