AI-Based Digital Pathology Model for Predicting Treatment Response and Prognosis in mHSPC

The goal of this multicenter observational study is to develop and validate an artificial intelligence-based digital pathology model to predict treatment response and prognosis in adult men with metastatic hormone-sensitive prostate cancer treated with androgen deprivation therapy plus androgen receptor signaling inhibitors. The main questions it aims to answer are: Can digital pathology-based artificial intelligence predict deep prostate-specific antigen response, defined as prostate-specific antigen less than 0.2 ng/mL at 6 to 7 months after treatment? Can digital pathology-based artificial intelligence predict progression to metastatic castration-resistant prostate cancer and time to metastatic castration-resistant prostate cancer? Researchers will compare artificial intelligence-defined low-risk and high-risk groups to evaluate whether the model can improve prediction of treatment response and prognosis beyond conventional clinicopathological risk stratification. Participants will not receive any additional intervention as part of this study. Researchers will retrospectively collect clinical characteristics, treatment records, prostate-specific antigen response data, follow-up outcomes, and archived prostate biopsy slides from participating centers. The biopsy slides will be scanned into whole-slide images and analyzed using artificial intelligence to develop and validate predictive models.

Inclusion Criteria: * Histologically confirmed prostate adenocarcinoma diagnosed as metastatic hormone-sensitive prostate cancer at initial diagnosis. Metastatic disease may include distant bone metastases, visceral metastases, and/or metastatic lymph node involvement according to the definition of metastatic hormone-sensitive prostate cancer at the time of diagnosis. * Received first-line systemic therapy consisting of standard androgen deprivation therapy, either surgical or medical castration, in combination with at least one androgen receptor signaling inhibitor, including but not limited to abiraterone plus prednisone, enzalutamide, or apalutamide. * Availability of diagnostic prostate needle biopsy specimens obtained at initial diagnosis, with formalin-fixed paraffin-embedded tissue sections available and digitized into high-resolution whole-slide images in .svs format. * Availability of complete clinical follow-up data, including serial prostate-specific antigen measurements after treatment, recorded monthly or quarterly when available, and documented information on progression to metastatic castration-resistant prostate cancer. In particular, prostate-specific antigen level at approximately 6 to 7 months after treatment initiation must be available. * Age 18 years or older at the time of diagnosis or treatment initiation. * Eastern Cooperative Oncology Group performance status of 0 to 2 at the time of first-line systemic treatment initiation. * Expected survival of more than 6 months at the time of first-line systemic treatment initiation. * Informed consent or ethical approval for research use of clinical data and pathology materials must have been obtained according to the requirements of each participating center. Exclusion Criteria: * Unavailable or unusable pathology slide images, including missing tissue blocks or slides, failure of slide digitization, or whole-slide images of insufficient quality for artificial intelligence analysis. * Pathology specimens obtained from surgical resection rather than diagnostic needle biopsy. To ensure consistency of tissue source, only prostate needle biopsy specimens obtained at initial diagnosis will be included. * Diagnosis of special histological subtypes or mixed malignancies that do not represent typical androgen receptor pathway-driven prostate adenocarcinoma, including cases with a prominent small-cell neuroendocrine carcinoma component. * Discontinuation of first-line androgen deprivation therapy plus androgen receptor signaling inhibitor within 6 months for reasons unrelated to disease progression, such as severe adverse events, intolerance, or patient refusal. Incomplete follow-up information, including inability to determine the date of progression to metastatic castration-resistant prostate cancer or lack of adequate follow-up data for outcome assessment. -Receipt of additional systemic therapies that may substantially affect prostate-specific antigen levels during the treatment course, such as poly adenosine diphosphate-ribose polymerase inhibitors, unless such therapies were initiated after the defined outcome assessment period or after progression.