Gleason score on biopsy: is it reliable for predicting the final grade on pathology?

OBJECTIVE: To assess the correlation of the Gleason score on biopsy and the final pathology after radical prostatectomy (RP) for prostate adenocarcinoma. PATIENTS AND METHODS: In a retrospective analysis within a tertiary-care centre, the charts of 537 patients who had undergone radical prostatectomy from April 1989 to November 2000 were reviewed. The RPs were undertaken in one institution; 167 biopsies were taken and interpreted in the referring centres, and 355 were taken and interpreted in the authors' institution by up to 15 pathologists. All the final pathology specimens were interpreted by the same group of pathologists. The main outcome measures were: the pathological report of the biopsy including the primary and secondary Gleason grade; the final pathological grade (primary and secondary); the margin status; and the identification of the pathologist for the biopsy and final pathology. RESULTS: In all, 390 patients had inclusion criteria (the Gleason grade before and after RP) available. For the individual scores 38.2% of tumours were undergraded, 32.6% overgraded and only 29.2% had identical grading in preoperative biopsies and final specimens. When grouped into more meaningful categories (Gleason 2-4, 5-6, 7 and 8-10) the correlation improved, with 48.5% of patients remaining in the same group after RP. For 39 patients the same pathologist assessed the biopsy and final specimen; in these cases individual scores were identical in 49% and group scores were identical in 64%. CONCLUSION: Gleason grading of the prostate biopsy remains a poor predictor of pathological outcome. Assessment by the same pathologist reduces the discrepancy but over half the patients are under- or overgraded on final pathology. Clinicians should be aware of these limitations when using the biopsy Gleason grade in decision making.