Abstract PR012: Divergent FOXA1 mutations drive prostate tumorigenesis and therapy-resistant cellular plasticity

Abstract Androgen receptor (AR) signaling is critical for the survival and proliferation of prostate cancer (PCa) cells, thus making androgen deprivation therapy (ADT; e. g. , castration) the mainstay for treatment. Notably, the oncogenic transcriptional functions of AR rely on a host of chromatin-binding regulatory proteins, which includes a pioneer transcription factor called FOXA1. FOXA1 de-compacts chromatin to enable DNA binding of AR and activation of target gene expression. Our lab found FOXA1 alterations to recur within three distinct structural classes in over 35% of metastatic castration-resistant PCa (mCRPC) cases in Caucasian White men. Subsequent studies reported that FOXA1 mutations are prevalent in over 40% of primary PCa in Chinese patients, thus positioning FOXA1 as a principal oncogene in this disease. Yet, the tumorigenic potential and pathobiology of FOXA1 alterations remain unexplored in vivo. Here, we have developed and characterized the first in-field transgenic mouse models that conditionally overexpress FOXA1 mutants in the prostate luminal epithelia. Our findings reveal that truncal FOXA1 class1 mutations (i. e. , wing 2 alterations) in a Trp53-null background triggered high-grade adenocarcinoma mimicking human disease, characterized by abnormal induction of NSD2 expression, extensive AR reprogramming to chimeric AR-half neo-enhancers, and sensitivity to androgen deprivation therapies. In contrast, FOXA1 class 2 mutations (i. e. , C-terminal truncations) that are acquired in human mCRPC do not drive prostate luminal transformation in mice. Instead, these mutations induce intra-luminal plasticity by reprogramming differentiated cells into a progenitor stem-like state within androgen-replete normal tissues—a phenomenon otherwise induced upon castration. Class 2 mutants lead to a 20-fold expansion of Ar+/Ck8+ luminal epithelia that gain expression of stemness markers such as Trop2, Ck4, and Psca. These stem cells are similar to the Club cells detected in the human prostate gland that have been implicated in driving resistance to ADT. Mechanistically, we found the cistromically-dominant Class 2 mutants to pioneer over 40, 000 neo-enhancer elements that were bound by stemness-associated transcription factors, like KLF5 and AP-1, which together instruct an androgen-insensitive luminal progenitor cell fate. Consistently, we found that Class 2-mutant mouse prostates showed minimal atrophy upon castration, with immunohistological assessment revealing a higher density of Ki67+ luminal epithelial cells relative to both wild-type and Class 1-mutant tissues. Class 2-mutant organoids also showed greater subcutaneous grafting ability in limiting-dilution assays in mice. Collectively, our data establish FOXA1 as a multifaceted oncogene in AR-dependent prostate cancers, in which divergent evolution of FOXA1 mutational classes distinctly drives either cancer formation or therapy-resistant, intra-luminal plasticity during disease progression. These findings also represent the first report of FOXA1-driven prostate adenocarcinoma in mice. Citation Format: Sanjana Eyunni, Rahul Mannan, Yuping Zhang, Eleanor Young, Qiuyang Zhang, Jie Luo, Matthew Pang, Somnath Mahapatra, Jean Ching-Yi Tien, James George, Mustapha Jaber, Hamzah Hakkani, Sandra E. Carson, Abigail J. Todd, Noshad Hosseini, Mahnoor Gondal, Ryan J. Rebernick, Xuhong Cao, Fengyun Su, Rui Wang, Rohit Mehra, Jing Li, Marcin Cieslik, Arul M. Chinnaiyan, Abhijit Parolia. Divergent FOXA1 mutations drive prostate tumorigenesis and therapy-resistant cellular plasticity abstract. In: Proceedings of the AACR Special Conference in Cancer Research: Innovations in Prostate Cancer Research and Treatment; 2026 Jan 20-22; Philadelphia PA. Philadelphia (PA): AACR; Cancer Res 2026;86 (2Suppl): Abstract nr PR012.