Cribriform pattern and IDC-P in prostate biopsies: prognostic relevance and reporting in metastatic disease

Abstract

Cribriform pattern and intraductal carcinoma of the prostate are recognized adverse histological features, yet their prognostic value in treatment-naïve metastatic disease remains uncertain. We conducted a single-center retrospective study of 183 biopsy-proven prostate carcinomas (105 with metastatic castration-sensitive prostate carcinoma and 78 non-metastatic high-grade cases) diagnosed between 2017 and 2024. Cribriform pattern, intraductal carcinoma of the prostate, and coagulative tumor necrosis were recorded per core and summarized as patient-level binary status and as semiquantitative proportions per cancer-positive core. Two multivariable logistic regression models (binary and semiquantitative) were fitted, and receiver operating characteristic (ROC) analysis evaluated the discriminatory performance of the cribriform proportion. Cribriform pattern and intraductal carcinoma of the prostate were more frequent in metastatic castration-sensitive prostate carcinoma. In the semiquantitative model, the cribriform proportion remained independently associated with metastatic status [odds ratio (OR) 1.29, 95% CI 1.07–1.55, p = 0.008; per 1.0 increase in the proportion, equivalent to OR 1.03 per 10%-point increase], whereas necrosis remained significant only in the binary model. The cancer-positive core rate and a lower total number of biopsy cores were predictive in both models, whereas prostate-specific antigen, intraductal carcinoma of the prostate, and Grade Group composition were not independent predictors. ROC analysis for the cribriform proportion yielded an area under the curve of 0.704, with a Youden Index cut-off of 0.445 (approximately half of cancer-positive cores), corresponding to a sensitivity of 57.1% and a specificity of 75.6%. These findings indicate that semiquantitative reporting of cribriform pattern – expressed as the proportion of cancer-positive cores – adds discriminatory information for metastatic status at presentation and could complement binary reporting in high-grade disease. From a clinical perspective, such evaluation may refine risk stratification at diagnosis and support treatment intensification strategies in very-high-risk patients.