Consistencies and inconsistencies of Gleason score and grade group system in prostate biopsy and radical prostatectomy

Abstract

The diagnostic of prostatic adenocarcinoma (PADK) is based on two scoring systems, namely the Gleason grading system and the grade group system - both of them being included in the pathology report. There is a discrepancy in the Gleason score given on the initial biopsy and prostatectomy specimen, respectively, that affects both patients who can benefit from active surveillance and those who cannot benefit from curative surgical treatment, and for whom therapeutic alternatives must be established. In this context, our study aimed at analyzing the diagnostic concordance between biopsies and subsequent prostatectomies, in order to establish the accuracy of prostate biopsy in predicting final Gleason score of the prostatectomy specimens. Material and methods: We analyzed 79 cases of PADK, initially diagnosed by prostate biopsy and subsequently treated by radical prostatectomy. The accuracy degree of the tumoral grading system assigned to the prostate biopsy, compared with the radical prostatectomy, was assessed by calculation of the Cohen's kappa concordance coefficient and calculation of sensitivity, specificity, positive prognostic value and negative prognostic value. Results: 37 cases (46.83%) presented the same Gleason score given for prostate biopsies and radical prostatectomies. In prostate biopsies, 32 cases (40.50%) were underestimated, compared to the Gleason score established by evaluating radical prostatectomy specimens, and 10 cases (12.65%) were overestimated. The highest agreement rate between the Gleason score in prostate biopsy and radical prostatectomy was found for Gleason score 6 (3+3) and 7 (3+4)-17 cases (21.51%) for each score category. The lowest agreement rate was observed for Gleason score 5, where no case showed a similar score for biopsy and prostatectomy. Statistical analysis indicated a kappa coefficient of 0.23, corresponding to a statistically significant satisfactory agreement (p = 0.01). The agreement on ISUP prognostic grade groups between prostate biopsy and radical prostatectomy indicated a slight increase in kappa coefficient -0.41, while maintaining a statistically significant moderate agreement (p = 0.0001). Conclusions: Our results demonstrate an increase in the predictive accuracy of prostate biopsy, as a result of the application of prognostic grade group system, which ensure the decrease of overestimating tendency of the old scoring system, ultimately leading to a preliminary characterization of potential lesions in prostatectomy specimens and a more effective treatment stratification of patients.