Several studies have been performed, mainly in Western countries, to identify the risk factors for cancer in men undergoing repeat prostate biopsy [11–13]. Borboroglu et al. demonstrated that the only statistically significant predictor of a positive repeat biopsy was PSAV (P < 0.001). Prostate cancer was detected in 64% of men with a PSAV of 1 ng/ml or higher in extensive transrectal ultrasound guided prostate biopsies (average of 22.5 cores) . Djavan et al. have suggested that percent-free PSA was the most accurate predictor of prostate cancer in settings where a repeat biopsy was performed . The authors carried out a large retrospective study, and concluded that repeat biopsies should be performed in patients with a percent free PSA of less than 30% or a transition zone PSAD of 0.26 ng/ml2 or greater . Ouyang et al. reported that the presence of atypia at initial biopsy is a strong predictor of malignancy in subsequent biopsies . On the other hand, in a Japanese study, Park et al. reported that total PSA, PSAD, PSAV, digital rectal examination, and TRUS findings were independent predictors of a positive repeat biopsy . Despite these data, there is currently no standard set of criteria governing the performance of repeat biopsies following a negative initial biopsy in Japan or in Western countries. In the current study, we have carried out a retrospective analysis of clinical and pathological data on positive and negative repeat biopsies in order to identify predictors of repeat biopsy outcome. Based on an ROC curve analysis, we determined cut-off values for PSAV and PSAD of 0.80 ng/ml/year and 0.30 ng/ml2, respectively. When at least one of these criteria was satisfied, we predicted 87% (20 of 23) of positive repeat biopsies. We also determined that 94% (46 of 49) of patients who had PSAV and PSAD values that were below the cut-off value underwent unnecessary repeat biopsies. Furthermore, the use of PSAV and PSAD criteria would have spared 44% (46 of 127) of patients from undergoing a repeat biopsy (Table 1). These results warrant additional studies to identify other variables that can be used in conjunction with PSAV and PSAD to predict the results of a repeat biopsy, and decrease the number of needless repeat biopsies performed.
We also carried out a retrospective analysis of the clinical and pathological characteristics and outcomes of patients who were diagnosed at initial and repeat biopsies. Miyake et al. demonstrated that there were no significant differences in the final pathological features of prostate cancers that were detected at initial and repeat biopsies . Although the authors did not analyze the outcome of patients who were diagnosed at repeat biopsy, they speculated that the biological behavior of the tumors that are detected at initial and repeat biopsies may be similar. In the current study, we found that patients who were diagnosed at a repeat biopsy had a significantly lower pathological T-stage than those who were diagnosed at an initial biopsy. However, the outcome after radical prostatectomy was similar between the two groups (Figure 2A). We also found that the number of positive cores at biopsy was not predictive of outcome after radical prostatectomy (Figure 2B).
There have been several recent studies evaluating the rate or duration of biochemical recurrence-free survival after radical prostatectomy in patients who are diagnosed at repeat biopsy. The largest retrospective study was carried out by Lopez-Corona et al. . The authors found that cancer was diagnosed in 1,042 patients at an initial biopsy and 315 at a repeat biopsy. Patients who were diagnosed at repeat biopsies and underwent radical prostatectomy had a higher rate of clinical T1c stage cancer and organ-confined disease than patients who were diagnosed at an initial biopsy (P < 0.0001) . However, despite the appearance of more favorable pathological features in tumors that were detected at a repeat biopsy, there was no difference in biochemical recurrence rate . Steiner et al. demonstrated that when prostate cancer was diagnosed at a repeat biopsy, a negative result at the initial needle biopsy was predictive of a lower pathologic stage and grade, as well as smaller tumor volume . However, patients who were diagnosed at a repeat biopsy did not have more favorable outcomes after radical prostatectomy. Our results agree with these previous studies, and indicate that patients who are diagnosed at a repeat biopsy include those with clinically insignificant and organ-confined cancer, and those with treatment delay. To our knowledge, this is the first study to demonstrate that biochemical recurrence-free survival after radical prostatectomy is similar in an Asian population of prostate cancer patients who were diagnosed at initial or repeat biopsies, similar to Caucasian populations.
The average number of cores per biopsy increased over the period of time examined in the current study, from 6 cores (between 1998 and 2003) to 10 cores (between 2004 and 2006). In addition, in the later period, there were more patients diagnosed with prostate cancer at a repeat biopsy (90/191 positive initial biopsy, 21/23 positive repeat biopsies). These variables represent a potential for bias in the current study, in terms of both the pathological features of the tumor and prognosis. However, our results warrant additional comprehensive analyses of patients that are diagnosed with prostate cancer at repeat biopsies.