Digital pathology for repeat prostate biopsy reassessment during active surveillance: an institutional experience.

INTRODUCTION

Repeat biopsy remains central to active surveillance (AS) because upgrading and tumor burden may change management.

METHODS

This retrospective paired workflow study included 38 men and 47 repeat biopsy sessions (2023-2025), comprising 846 biopsy cores and 871 H&E slides. Whole-slide imaging (WSI) and conventional microscopy were compared after a washout interval. QuPath was used only for pathologist-guided tumorlength and greatest percentage of cancer (GPC) documentation.

RESULTS

On the first eligible repeat biopsy, upgrading to ISUP Grade Group (GG) 2 occurred in 7/38 men (18.4%; 95% CI, 7.7- 34.3). No GG3, cribriform morphology, or intraductal carcinoma was identified. Protocol-linked reclassification occurred in 11/38 men (28.9%). Slide-level cancer-detection agreement was 856/871 (98.3%; kappa = 0.91), and raw patient-level agreement for upgrading was 36/38 (94.7%; kappa = 0.80). After adjudication, no upgraded patient remained missed. QuPathsupported measurement was interpretable in 79/101 positive core-level assessments (78.2%), with excellent agreement (ICC = 0.98) and shorter median measurement time (41 s vs 78 s).

CONCLUSIONS

WSI reproduced management-relevant conventional microscopy outputs in repeat AS biopsies. Its value was organizational and documentary, not autonomous cancer detection or grading.