Summary The lack of in vitro prostate cancer models that recapitulate the diversity of human prostate cancer has hampered progress in understanding disease pathogenesis and therapy response. Using a 3D “organoid” system, we report success in long-term culture of prostate cancer from biopsy specimens and circulating tumor cells. The first seven fully characterized organoid lines recapitulate the molecular diversity of prostate cancer subtypes, including TMPRSS2-ERG fusion, SPOP mutation, SPINK1 overexpression and CHD1 loss. Whole exome sequencing shows a low mutational burden, consistent with genomics studies, but with mutations in FOXA1 and PIK3R1, as well as of DNA repair and chromatin modifier pathways that have been reported in advanced disease. Loss of p53 and RB tumor suppressor pathway function are the most common feature shared across the organoid lines. The methodology described here should enable the generation of a large repertoire of patient-derived prostate cancer lines amenable to genetic and pharmacologic studies.

Plasticity of the cell state has been proposed to drive resistance to multiple classes of cancer therapies, thereby limiting their effectiveness1–4. A high-mesenchymal cell state observed in human tumours and cancer cell lines has been associated with resistance to multiple treatment modalities across diverse cancer lineages, but the mechanistic underpinning for this state has remained incompletely understood1–6. Here we molecularly characterize this therapy-resistant high-mesenchymal cell state in human cancer cell lines and organoids and show that it depends on a druggable lipid-peroxidase pathway that protects against ferroptosis, a nonapoptotic form of cell death induced by the build-up of toxic lipid peroxides7,8. We show that this cell state is characterized by activity of enzymes that promote the synthesis of polyunsaturated lipids. These lipids are the substrates for lipid peroxidation by lipoxygenase enzymes8,9. This lipid metabolism creates a dependency on pathways converging on the phospholipid glutathione peroxidase (GPX4), a selenocysteine-containing enzyme that dissipates lipid peroxides and thereby prevents the iron-mediated reactions of peroxides that induce ferroptotic cell death8. Dependency on GPX4 was found to exist across diverse therapy-resistant states characterized by high expression of ZEB1, including epithelial-mesenchymal transition in epithelial-derived carcinomas, TGFβ-mediated therapy- resistance in melanoma, treatment-induced neuroendocrine transdifferentiation in prostate cancer, and sarcomas, which are fixed in a mesenchymal state owing to their cells of origin. We identify vulnerability to ferroptic cell death induced by inhibition of a lipid peroxidase pathway as a feature of therapy-resistant cancer cells across diverse mesenchymal cell-state contexts.

Some cancers evade targeted therapies through a mechanism known as lineage plasticity, whereby tumor cells acquire phenotypic characteristics of a cell lineage whose survival no longer depends on the drug target. We use in vitro and in vivo human prostate cancer models to show that these * Corresponding author. sawyersc@mskcc.org. SUPPLEMENTARY MATERIALS

Summary The prostate gland consists of basal and luminal cells arranged as pseudo-stratified epithelium. In tissue recombination models, only basal cells reconstitute a complete prostate gland, yet murine lineage-tracing experiments show that luminal cells generate basal cells. It has remained challenging to address the molecular details of these transitions and whether they apply to humans, due to the lack of culture conditions that recapitulate prostate gland architecture. Here we describe a 3D culture system that supports long-term expansion of primary mouse and human prostate organoids, composed of fully differentiated CK5+ basal and CK8+ luminal cells. Organoids are genetically stable, reconstitute prostate glands in recombination assays and can be experimentally manipulated. Single human luminal and basal cells give rise to organoids, yet luminal cell-derived organoids more closely resemble prostate glands. These data support a luminal multilineage progenitor cell model for prostate tissue and establish a robust, scalable system for mechanistic studies.

Summary The transition from castration resistant prostate adenocarcinoma (CRPC) to neuroendocrine prostate cancer (NEPC) has emerged as an important mechanism of treatment resistance. NEPC is associated with over-expression and gene amplification of MYCN (encoding N-Myc). N-Myc is an established oncogene in several rare pediatric tumors, but its role in prostate cancer progression is not well established. Integrating a genetically engineered mouse model and human prostate cancer transcriptome data, we show that N-Myc over-expression leads to the development of poorly differentiated, invasive prostate cancer that is molecularly similar to human NEPC. This includes an abrogation of AR signaling and induction of Polycomb Repressive Complex 2 signaling. Altogether, our data establishes N-Myc as an oncogenic driver of NEPC.