Establishment and characterization of a panel of prostate adenocarcinoma XPDX models representing naïve, chemotherapy- and radiotherapy-resistant patient populations.

251 Background: Preclinical models of prostate adenocarcinoma (PRAD) are challenging to establish and maintain. We previously reported (AACR2025) the successful establishment and characterization of ten parental and three conditioned prostate adenocarcinoma models in immunodeficient mice. Using this experience, we have expanded our platform with eight additional PRAD models from patients naïve to or progressing on chemo- and radiotherapies. These models were characterized for receptor expression, genomic alterations, and in vivo drug sensitivity to relevant therapies. Methods: XPDX models representing prostate cancer were established from primary (ST6347, ST7756, ST7888) or metastatic (STM637B, STM655, STM784, STM784B, STM784C) samples implanted into intact immunodeficient male mice. Resulting models were passaged serially and further developed in both intact and castrated mice until growth stabilization. Resulting models were characterized using genomic analysis, including WES and RNA seq , receptor expression, and in vivo drug sensitivity studies. For in vivo studies, docetaxel, abiraterone, and enzalutamide were evaluated at standard treatment regimens. Study endpoints included tumor volume and time from treatment initiation with %T/C values and tumor regression reported at study completion; a %T/C of ≤ 20% versus control was considered sensitive. Tumor regression (%T/C < 0%) versus Day 0 tumor volume was also reported. Results: Differential staining for AR, PSMA, and AR-V7 was reported with these models, consistent with patient pretreatment status. ST7756 and ST7888 were established from naïve patients and ST6347 from a patient treated with docetaxel. Genomic characterization of STM637B, a model from a patient post-targeted and lutetium (177Lu) vipivotide tetraxetan (LVT) treatments, identified TMPRSS2-ERG fusion and PTEN loss. FGFR3 and NF1 aberrations were present in STM655, a model derived from a patient following chemo and LVT therapies. Three models developed from longitudinal sampling designated STM784, STM784B, and STM784C, were established from the same patient following progression on various targeted therapies and LVT, with STM784C collected after additional treatment with a PSMA-targeting Actinium-225 radiotherapy. Differential sensitivity to abiraterone, enzalutamide, and docetaxel was found in evaluated models that aligned with patient pretreatment history. Conclusions: We have established and characterized an expanded panel of PRAD XPDX models including several representing patient post chemo- and radiotherapies. These models can be utilized as a valuable tool in better understanding prostate cancer and in developing novel therapies for drug and radio-resistant patients.