Total Gleason Pattern 4 Length Outperforms Grade Group and Clinical Models in Predicting Oncologic Outcomes in Grade...

Area of Science:

• Urology
• Oncology
• Pathology

Background:

• Risk stratification for localized prostate cancer primarily relies on Grade group (GG).
• In GG2-4 disease, risk assignment is based on the proportions of Gleason pattern 3 and pattern 4.
• Current methods may not fully capture the prognostic information contained within biopsy patterns.

Purpose of the Study:

• To evaluate if total pattern 4 length on prostate biopsy better predicts oncologic outcomes than Grade group, percent pattern 4, or established nomograms.
• To determine the added value of pattern 3 and clinical variables when total pattern 4 length is known.

Main Methods:

• Retrospective analysis of 2499 patients with GG2-4 prostate cancer undergoing radical prostatectomy.
• Calculation of discrimination for adverse pathologic stage (seminal vesicle or lymph node invasion) and biochemical recurrence (BCR).
• Comparison of total pattern 4 length against GG, percent pattern 4, and a multivariable clinical model.

Main Results:

• Total pattern 4 length demonstrated superior discrimination for adverse pathologic stage (AUC 0.779) compared to GG (0.658), percent pattern 4 (0.719), and the clinical model (0.762).
• Similar results were observed for BCR prediction, with total pattern 4 length outperforming GG (C-index 0.716 vs 0.662), percent pattern 4 (0.695), and the clinical model (0.699).
• Neither pattern 3 length nor the clinical model provided additional predictive discrimination when total pattern 4 length was considered.

Conclusions:

• Total Gleason pattern 4 length on biopsy is a more robust predictor of oncologic outcomes in GG2-4 prostate cancer than current stratification methods.
• Standard clinicopathologic variables do not enhance predictive accuracy beyond total pattern 4 length.
• Further research is needed to optimize pattern 4 quantification for clinical risk stratification algorithms.