Abstract 3933: Modeling treatment naïve and castration-resistant metastatic prostate cancer in genetically engineered mouse models

Abstract Metastatic prostate cancer, the major cause of prostate cancer lethality, presents as several distinct subtypes. Treatment naïve metastatic prostate cancer, often called oligometastatic disease, develops within an androgen-intact context and, given its low frequency in the population, remains understudied. The more prevalent type of metastatic disease arises from castration-resistant prostate cancer (CRPC) as a consequence of treatment failure in response to androgen deprivation therapy (ADT). Both types of prostate cancer metastases have eluded curative treatments, and the lack of accurate molecular and phenotypic in vivo models has been a roadblock to developing effective treatments. We have developed genetically engineered mouse models (GEMMs) that recapitulate advanced prostate cancer through disruption of molecular processes that are typically altered in late-stage human prostate cancer. In particular, we have generated GEMMs that carry loss of function of the tumor suppressors, Nkx3.1, Pten and p53 alone or together with loss of DNA damage repair (DDR) genes Brca1, Brca2, or Atm. We have studied these GEMMs in parallel with mice that have both Nkx3.1 and Pten loss and gain of function of oncogenic Kras. The phenotype of these GEMMs were evaluated in androgen intact and deprived conditions to recapitulate treatment naïve and castration resistance contexts. Each of these GEMMs develop prostate cancer with distinct characteristics in the presence or absence of androgens. GEMMs with Pten loss displayed a non-metastatic indolent phenotype of prostatic intraepithelial neoplasia. When paired with p53 knockdown or mutation, the GEMMs showed phenotypic variety (adenocarcinoma, squamous cell, sarcomatoid) and had moderate metastasis incidence (∼50%). GEMMs that additionally have loss of function of DDR genes developed more aggressive prostate cancer with increased penetrance to organs and bone, modeling the metastatic tropism observed in human patients. Most notably, the Brca GEMMs that underwent castration revealed higher metastatic incidences overall, with increases in the lung, liver, and bone, indicating increased CRPC penetrance. Additionally, constitutive activation of Kras resulted in an aggressive adenocarcinoma phenotype with fully penetrant metastatic disease and the highest incidence of bone metastasis (∼40%). Interestingly, castration of Kras GEMMs led to a significant reduction in liver and bone metastases. To summarize, through the loss of function of tumor suppressor genes, DDR genes, and gain of function of Kras in murine models, we have successfully modeled treatment naïve and castration-resistant metastatic prostate cancer. These GEMMs are a valuable resource to identify therapeutic targets in advanced prostate cancer for various subtypes of metastatic prostate cancer in androgen intact and castrated contexts. Citation Format: Jeannine Turgeon, Florencia Picech, Jie Zheng, Cory Abate-Shen. Modeling treatment naïve and castration-resistant metastatic prostate cancer in genetically engineered mouse models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3933.