Materials and methods

Study design

We conducted a retrospective multicenter cohort study including patients diagnosed with prostatic ductal adenocarcinoma across seven high-volume urological centers in Germany between 2014 and 2024. Clinical data were collected and analyzed collaboratively across the participating institutions, following standardized protocols.

Patient selection

A total of 82 patients with histologically confirmed PDA from seven high-volume urological centers in Germany were included. Only patients with histologically confirmed prostatic ductal adenocarcinoma were included, whereas cases with isolated intraductal carcinoma without ductal adenocarcinoma morphology were not considered part of the study cohort. Diagnosis was established on RP specimens and, where applicable, on prostate biopsy. If available, the percentage of the ductal component was also recorded. Patients with insufficient histopathologic characterization were excluded. The analyzed therapeutic data comprised all treatments after local therapy, biochemical recurrence, and survival until 2025.

Clinicopathological data collection

Baseline clinicopathological parameters were extracted from institutional databases. They included age at diagnosis, relevant secondary diagnoses, smoking status, preoperative PSA level, clinical T stage, biopsy ISUP grade, and risk classification according to the European Association of Urology (EAU) guidelines. For patients undergoing RP, pathological findings included tumor stage (pT), lymph node status (pN), surgical margin status, and ISUP from the final specimen. Whenever available, the percentage of ductal growth infiltration was assessed. Baseline staging modalities were performed according to institutional standards and evolving guideline recommendations during the study period and included conventional imaging (computed tomography and bone scintigraphy) and, in later years, PSMA-PET/CT where available. Data regarding adjuvant, salvage, or later therapies (androgen deprivation therapy (ADT), androgen receptor pathway inhibitors (ARPI), chemotherapy, or radiotherapy) were recorded.

Outcome measures

The primary endpoint was time to initiation of systemic therapy following local treatment, defined as the interval between RP or radiotherapy and the initiation of any systemic therapy after the diagnosis of metastatic disease or biochemical recurrence (ADT, ARPI, or chemotherapy).Due to the retrospective multicenter design and the heterogeneous availability and granularity of longitudinal PSA follow-up data across participating institutions, standardized PSA-based biochemical recurrence definitions could not be consistently reconstructed for the entire cohort. Therefore, initiation of systemic therapy was selected as a pragmatic clinically meaningful endpoint reflecting disease progression requiring therapeutic escalation in routine clinical practice. This endpoint included initiation of systemic therapy after biochemical recurrence, or in the context of metastatic disease progression, depending on the documented clinical indication at the treating institution. To avoid including patients undergoing immediate adjuvant radiotherapy or with PSA persistence, only patients receiving systemic therapy at least 6 months after local therapy were included in this analysis.

Secondary endpoints included overall survival (OS) and cancer-specific survival (CSS). Given the limited follow-up duration, survival outcomes were considered exploratory in nature. Additionally, due to the heterogeneity in follow-up documentation, non-curative biochemical recurrence was operationally defined as the initiation of systemic ADT, with or without additional therapy such as ARPI or chemotherapy, but without concomitant local radiotherapy.

Statistical analysis

Medians and interquartile ranges (IQRs) were reported for continuous variables; frequencies and proportions for categorical variables. Chi-square tests compared categorical variables (e.g., PDA at diagnosis vs. final pathology). Due to limited sample size and events, separate univariable Cox models identified predictors of earlier initiation of systemic therapy, including age, PSA, ISUP (biopsy and post-RP), and pathological T-stage. Results are presented as hazard ratios (HR) with 95% confidence intervals (CI). Kaplan-Meier curves were used to descriptively visualize time to systemic therapy initiation; time zero was diagnosis. Patients without initiation of systemic therapy at last follow-up were censored at the date of last clinical contact or death. Unpaired t-tests compared mean therapy duration. All analyses were conducted using Stata (Version Stata/SE 15.1, Stata Corp LLC, TX, USA). A two-sided p-value < 0.05 was considered statistically significant.

Given the limited number of events and overall sample size, all regression analyses should be considered exploratory and hypothesis-generating in nature.