A biopsy simulation study to assess the accuracy of several transrectal ultrasonography (TRUS)-biopsy strategies compared with template prostate mapping biopsies in patients who have undergone radical prostatectomy.

Abstract

What's known on the subject? and What does the study add? Transrectal ultrasonography (TRUS)-guided biopsies can miss prostate cancer and misclassify risk in a diagnostic setting; the exact extent to which it does so in a repeat biopsy strategy in men with low-intermediate risk prostate cancer is unknown. A simulation study of different biopsy strategies showed that repeat 12-core TRUS biopsy performs poorly. Adding anterior sampling improves on this but the highest accuracy is achieved using transperineal template prostate mapping using a 5 mm sampling frame. To determine the effectiveness of two sampling strategies; repeat transrectal ultrasonography (TRUS)-biopsy and transperineal template prostate mapping (TPM) to detect and exclude lesions of ≥0.2 mL or ≥0.5 mL using computer simulation on reconstructed three-dimensional (3-D) computer models of radical whole-mount specimens. Computer simulation on reconstructed 3-D computer models of radical whole-mount specimens was used to evaluate the performance characteristics of repeat TRUS-biopsy and TPM to detect and exclude lesions of ≥0.2 mL or ≥0.5 mL. In all, 107 consecutive cases were analysed (1999-2001) with simulations repeated 500 times for each biopsy strategy. TPM and five different TRUS-biopsy strategies were simulated; the latter involved a standard 12-core sampling and incorporated variable amounts of error, as well as the addition of anterior cores. Sensitivity, specificity, negative and positive predictive values for detection of lesions with a volume of ≥0.2 mL or ≥0.5 mL were calculated. The mean (SD) age and PSA concentration were 61 (6.4) years and 8.5 (5.9) ng/mL, respectively.In all, 53% (57/107) had low-intermediate risk disease. In all, 665 foci were reconstructed; there were 149 foci ≥0.2 mL and 97 ≥ 0.5 mL in the full cohort and 68 ≥ 0.2 mL and 43 ≥ 0.5 mL in the low-intermediate risk group. Overall, TPM accuracy (area under the receiver operating curve, AUC) was ≈0.90 compared with AUC 0.70-0.80 for TRUS-biopsy. In addition, at best, TRUS-biopsy missed 30-40% of lesions of ≥0.2 mL and ≥0.5 mL whilst TPM missed 5% of such lesions. TPM under simulation conditions appears the most effective re-classification strategy, although augmented TRUS-biopsy techniques are better than standard TRUS-biopsy.