[Transrectal biopsy scheme can predict incorrect histological grading in prostate cancer]

To identify factors that might explain why a prostate with a Gleason score (GS) <7 in the biopsy specimen can turn out to have a GS ≥7 in the surgical specimen.

We compared the GS of biopsy specimens with the GS of surgical specimens in 185 patients who underwent surgery for prostate cancer. We calculated the sensitivity, specificity, and predictive values for the GS of the biopsy specimens. We used Cohen's kappa to determine the degree of concordance between a GS of <7 and ≥7 for the biopsy specimen and the surgical specimen. Age, a family history of prostate cancer, total prostate-specific antigen (tPSA), digital rectal examination, prostate structure and volume, and the number of biopsy cores (biopsy scheme) were analyzed using multivariable logistic regression.

Histological study of biopsy specimens yielded high sensitivity (98%) but low specificity (49%) for GS ≤6 and low sensitivity (35, 26%) and high specificity (93, 99%) for GS=7 and GS ≥7, respectively. Cohen's kappa for the GS from the biopsy and surgical specimens was 0.43 (95% CI=30-56%). The biopsy scheme was the only predictor of discordance in the GS between the two techniques. Among the other variables included in the model, only tPSA showed a slightly significant association. Taking a scheme with less than 7 cores as a reference, we found no difference with 8 to 9 cores but we did find a difference with 10 to 11 cores and with 12 or more cores, with a prevalence ratio of 0.138 (95% CI=0.030-0.513) and 0.277 (95% CI=0.091-0.806), respectively.

The GS of the biopsy depends on the scheme. This factor must be taken into account when choosing a treatment option in patients with low tumor grade in biopsy specimens.