The treatment landscape of metastatic castration-sensitive prostate cancer (mCSPC) has expanded since the introduction of androgen receptor signaling inhibitors (ARSI) and docetaxel in combination with androgen deprivation therapy (ADT). However, the progression to castration-resistant prostate cancer (CRPC) remains inevitable. Ideally, multiple metastatic tumor biopsies across the clinical course are reasonable to elucidate resistance to systemic therapies or explore novel treatment targets. However, many patients do not have accessible metastatic lesions. To overcome this limitation, we aimed to establish patient-derived xenografts(PDXs) from circulating tumor cells (CTCs) as an experimental model.

CTCs were isolated by removing white blood cells from whole blood using anti-CD15 and anti-CD45 antibody. The extracted CTC solution was mixed with Matrigel at a 1:4 ratio and plated onto a dome shape in a 24-well plate to create organoids. Organoids were maintained in a 37℃ incubator. After approximately 2 weeks, organoids were subcutaneously implanted into NOD/SCID mice to establish PDX models.

We made PDX from the organoids of isolated CTCs from 10 blood samples of 6 patients with mCSPC or mCRPC. Two PDX models were successfully established using CTCs collected from one patient before and after treatment with ADT plus abiraterone. Hematoxylin and eosin staining revealed that the morphology of the two tumor tissues obtained from the PDX model was prostatic adenocarcinoma. Immunohistochemistry revealed that the two tumor tissues expressed AR, PSA, and NKX3.1. Interestingly, western blotting showed that one tumor tissue of PDX collected after ADT plus abiraterone had different molecular weights of the AR protein, indicating the presence of AR variants.

We successfully established 2 PDX models of CTCs from one patient with metastatic prostate cancer. We will evaluate the genome and transcriptome characteristics and verify their applicability as a new experimental model.