Heterogeneity of contemporary grade group 4 prostate cancer in radical prostatectomy specimens

Purpose The aim was to evaluate the prognostic role of sub-categories of ISUP 4 prostate cancer (PCa) on final pathology, and assess the tumor architecture prognostic role for predicting biochemical recurrence (BCR) after radical prostatectomy. Methods From a prospectively-maintained database, we included 370 individuals with ISUP 4 on final pathology. The main outcomes were to evaluate the relationship between different ISUP patterns within the group 4 with pathological and oncological outcomes. Binary logistic regression and Kaplan-Meier estimator were used to evaluate the role of the different categories (3 + 5, 4 + 4, 5 + 3) and tumor architecture (intraductal and/or cribriform) on pathological and oncological outcomes. Results Among the 370 individuals with ISUP considered for the study, 9, 85 and 6% had grade 3 + 5, 4 + 4 and 5 + 3 PCa, respectively. Overall, 74% had extracapsular extension, while lymph node invasion (LNI) was documented in 9%. A total of 144 patients experienced BCR during follow-up. After adjusting for PSA, pT, grade group, LNI and positive surgical margins (PSM), grade 3 + 5 was a protective factor (HR: 0.30, 95% CI: 0.13,0.68, p = 0.004) in predicting BCR relative to grade 4 + 4. Intraductal or cribriform architecture was correlated with BCR (HR: 5.99, 95% CI: 2.68, 13.4, p < 0.001) after adjusting for PSA, pT, grade group, LNI and PSM. Conclusions Patients with tumor grade 3 + 5 had better pathological and prognostic outcomes compared to 4 + 4 or 5 + 3. When accounting for tumor architecture, the sub-stratification into subgroups lost its prognostic role and tumor architecture was the sole predictor of poorer prognosis in terms of biochemical recurrence.

Automated summary

This study evaluated the prognostic role of sub-categories of ISUP 4 prostate cancer on final pathology, and assessed the tumor architecture prognostic role for predicting biochemical recurrence after radical prostatectomy.