Abstract B062: Interactions between prostate cancer lineage plasticity drivers and the RB1/E2F axis in AR-independent acquired resistance to AR-pathway inhibitors

Abstract Introduction: Androgen pathway inhibitors have transformed advanced prostate cancer (PCa) treatment yet remain susceptible to acquired resistance. Progression of PCa to androgen receptor (AR) -independent subtypes such as neuroendocrine prostate cancer (NEPC) confers poor prognosis. This lineage transformation is mediated by epigenetic mechanisms. NEPC frequently inactivates the tumor suppressor and cell cycle regulators RB1 and TP53, suggesting crosstalk between reprogramming factors and the cell cycle. While this bridge has been characterized in multiple cell types, it is understudied in NEPC. Salient examples include the pluripotency transcriptional factor (TF) SOX2, which regulates CCND1, and INSM1, which directly binds Cyclin D1 protein. Here, we investigate these interactions using isogenic inducible models and in-silico analysis of patient tumors, to identify therapeutic vulnerabilities exposed by neuroendocrine (NE) differentiation. Methods: Data were obtained from published datasets (Beltran et al, 2016; Labreque et al, 2019; DepMap) and the Caris CodeAI database. Patient and cell line data was analyzed for survival (where applicable) and differential mRNA expression between groups stratified by mRNA quartile. Inducible systems of TFs SOX2, ASCL1, FOXA2, and INSM1 were established in AR-positive LNCaP, and AR-negative DU145 and PC3 cell lines. RB1 and TP53 knockout (KO) LNCaP lines were generated by Cas9 electroporation, and Cyclin D1-HA (Addgene 174154) was stably overexpressed. TFs were induced with 200ng/mL doxycycline for 24h up to 28d, and protein expression was assessed via western blot. Cell viability and proliferation were surveyed by CellTiter-Glo® assay and nuclear counts, respectively. Results: NEPC patient data analysis revealed INSM1, ASCL1, FOXA2, and SOX2 expression to be negatively correlated with Cyclin D1 and positively correlated with Cyclin E1/2. High SOX2, ASCL1, FOXA2, or INSM1 correlated with worse prognosis in NEPC, while high SOX2 or INSM1 correlated with improved survival in prostate adenocarcinoma (PRAD). Elevated Cyclin D1 predicted improved outcomes in NEPC. INSM1 correlated with RB1 mutation and expression of E2F family members, CDK inhibitors, and E-/A-type cyclins in NEPC. Inducible INSM1, SOX2, ASCL1, and FOXA2 models showed robust induction 48h post-doxycycline addition, accompanied by expression of NE lineage markers and diminution of proliferation in the PRAD model LNCaP. Isogenic RB1 and TP53 KO LNCaP models were generated and validated, along with a CCND1 overexpression model. Ongoing work examines whether these models respond differently to TF induction. Conclusions: Our analyses reveal significant correlations between expression of NE-lineage TFs and cell cycle-regulators in clinical datasets. Novel in vitro studies suggest these TFs suppress proliferation. Future work will uncover whether RB1/TP53 loss and Cyclin D1 expression affect tolerance of NE programs, and uncover vulnerabilities such as G2/M checkpoint reliance, CDK inhibitor upregulation, and metabolic alterations. Citation Format: Connor Purcell, Anais Sidonia, Yumiko Imai, Audrey Su, Tyler Roady, Praveen R. Srinivasan, Maximilian Pinho-Schwermann, Benedito A. Carneiro, Wafik S. El-Deiry. Interactions between prostate cancer lineage plasticity drivers and the RB1/E2F axis in AR-independent acquired resistance to AR-pathway inhibitors 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 B062.