Comparison of Gleason upgrading rates in transrectal ultrasound systematic random biopsies versus US-MRI fusion biopsies for prostate cancer.

Abstract

57 Background: US-MRI fusion biopsy (FB) showed that it improves the detection of clinically significant prostate cancer (PCa). A more accurate diagnostic method is desirable to avoid misclassification, which in turn is particularly important in appropriate decision making on treatment for PCa (active surveillance or focal therapy or radical treatment). We aimed to compare the Gleason upgrading (GU) rates and the concordance of the Gleason scores in the biopsy versus final pathology after the surgery in patients who underwent only TRUS systematic random biopsies versus US-MRI FB for PCa. Methods: A retrospective analysis of patients’ prospective collected data that underwent prostate biopsy and subsequent radical prostatectomy were included from January 2011 to June 2016 at our institution. The study cohort was divided into: US-MRI FB (Group A) and only TRUS systematic random biopsy (Group B). US-MRI FB was performed in patients who had a previous MRI with a focal lesion classified by Likert score ≥ 3, otherwise a TRUS systematic random biopsy was performed. All biopsies and surgical specimens were analyzed by the same uropathologist and MRIs were analyzed by two expert urological radiologists. Results: 73 men underwent US-MRI FB and 89 TRUS systematic random biopsy. The GU rate was higher in group B (31.5% vs 16.4%; p = 0.027). GU according to Gleason grade pattern was higher in Group B against Group A (40.4% vs 23.3%; p = 0.02). Analyses from separate Gleason grade pattern showed that Gleason score 3+4 presented less GU in group A (24.1% vs 52.6%; p = 0.043). The Bland-Altman plot analysis showed a higher bias in Group B compared to group A (-0.27 [-1.40 to 0.86] vs -0.01 [-1.42 to 1.39]). In the multivariable logistic regression the only independent predictor of GU was the use of TRUS systematic random biopsy (2.64 [1.11 – 6.28]; p = 0.024). Conclusions: The US-MRI FB appears to be related to a decrease in GU rate and an increase in the concordance between biopsy and final pathology in comparison to TRUS random biopsy, that leads to greater accuracy on diagnosis and better treatment decision.