This observational study aims to develop and validate an artificial intelligence-based model using prostate cancer biopsy pathology to predict lymph node metastasis and distant metastasis in patients with prostate cancer. The main questions it aims to answer are:

Can artificial intelligence-assisted analysis of prostate cancer biopsy pathology accurately predict lymph node metastasis? Can the model accurately predict distant metastasis and assess metastatic risk in patients with prostate cancer?

Researchers aim to evaluate whether the model can provide additional information for clinical decision-making and surgical planning.

Participants will:

Provide prostate biopsy pathology specimens and related clinical information; Undergo assessment of lymph node and distant metastatic status based on clinical and imaging data; Be included in the development and validation of the artificial intelligence prediction model.

Inclusion Criteria:

- Male patients aged between 18 and 90 years; Patients who underwent prostate biopsy due to elevated prostate-specific antigen (PSA), abnormal digital rectal examination (DRE), or abnormal imaging findings, were pathologically diagnosed with prostate cancer, and had available prostate biopsy pathology specimens; Patients who underwent radical prostatectomy with extended pelvic lymph node dissection (ePLND) or pelvic lymph node dissection (PLND), with definitive pathological information regarding lymph node metastasis; Patients who underwent PSMA PET/CT, MRI, bone scintigraphy, or prostate MRI capable of identifying regional lymph node metastasis or distant metastasis; Adequate cardiac, pulmonary, hepatic, and renal function; Eastern Cooperative Oncology Group (ECOG) performance status of 0-1; Expected survival time greater than 1 year; Written informed consent signed by the patient or legally authorized representative.

Exclusion Criteria:

- History of other malignancies; Severe dysfunction of major organs, including cardiac, pulmonary, hepatic, or renal insufficiency, or an expected survival time of less than 1 year; Prostate biopsy pathology specimens with inadequate whole-slide image scanning quality or failure of quality control assessment; Patients with prostate cancer diagnosed from transurethral resection specimens.

The study population consists of patients with pathologically confirmed prostate cancer from multiple participating hospitals led by Xiangya Hospital. Eligible patients underwent prostate biopsy with available biopsy pathology specimens and relevant clinical and imaging data for assessment of lymph node and distant metastasis. Retrospective clinical and pathological data will be collected for development and validation of an artificial intelligence-based metastasis prediction model.

Except for essential information, no other information will be shared to protect patient privacy.