European Urology

1. Introduction

Despite efforts to improve the prostate cancer (PCa) detection rate using prostate biopsies, there exists a challenging cohort of patients with substantial risk factors for PCa who had a first negative biopsy set. If PSA increases and stays elevated or if atypical small acinar proliferation (ASAP) is found on biopsies, the scientific societies suggest repeat prostate biopsies, and a significant proportion of patients with PCa will be diagnosed on the second, third, or fourth set of biopsies [1], and [2]. Several authors suggest increasing the number of biopsies for this second set, with an extended or saturation protocol in order to increase the PCa detection rate [3], [4], [5], [6], [7], [8], [9], [10], [11], and [12].

Since 2001, our group has performed 21-needle prostate biopsies on every patient referred to us with an elevated PSA and/or an abnormal digital rectal exam (DRE) [12], and [13]. This protocol can be performed safely, and with minimal patient discomfort in the office using local anesthesia. A statistically significant increase of prostate cancer detection was observed [12].

However, for patients with a negative first extended biopsy protocol, the risk of missing a cancer is unknown: Can these patients be reassured of not having prostate cancer? What is the risk of having an aggressive disease after an extended protocol that is supposed to evaluate all prostate zones? The goal of this article is to focus on patients who had a negative first extended protocol and who were candidates for repeated biopsies. The risk of having PCa and the aggressiveness of these cancers were evaluated.

2. Materials and methods

Between March 2001 and May 2007, 2500 consecutive patients underwent an extended protocol of 21 biopsies. All patients were included prospectively in the clinical database. Of 953 patients who had a negative first extended prostate biopsy procedure, 231 patients underwent second, third, or more sets of 21-core biopsies of prostate. Indications for repeated biopsies were patients with high risk of PCa with prior PIN or ASAP on previous 21-core biopsies, prostate-specific antigen (PSA) persistently elevated (greater than 4ng/ml), increase of PSA during the follow-up, and persistent prostatic nodule on DRE. Clinical and pathologic data, including patient age, PSA, PSA density, transrectal ultrasound measured (TRUS) prostate volume, Gleason score, the number and location of positive cores, were analyzed using our computerized data base. PSA doubling time was calculated with the following method: PSADT = log2 × dT/(logB−logA) where A and B are the initial (A) and final (B) PSA measurements, and dT is the time difference between the calendar dates of the two PSA measurements [14].

The prostate needle biopsy procedure has been described previously [12]. Briefly, patients were prescribed enemas 1 d and 3 h before the procedure. A fluoroquinolone antibiotic was prescribed for 7 d, starting the day before the procedure. All patients were adequately informed of the mode of execution of the procedure and its potential complications. All patients received local anesthesia using a 22-gauge spinal needle that was passed through the biopsy guide channel and 10 cc 1% lidocaine was injected into each neurovascular bundle. Ultrasound prostate volume calculations were then performed. A total of 21 biopsies were taken, using 18-gauge biopsy needles and a spring-loaded biopsy gun, providing 17 mm length tissue cores. The patients were allowed to leave the hospital 2 h after the procedure. The biopsies were performed in the following order: First, six sextant medial biopsies at a standard 45° angle (numbers 1–6), then three biopsies in each lateral zone from base to apex at an 80° angle (numbers 11–16). Next, three biopsies were taken in each transitional zone from base to apex (numbers 111–116). Finally, three biopsies in the midline peripheral zone (numbers 7–9) (Fig. 1). Each prostate core was given a specific number according to the biopsy protocol and was analyzed separately [5]. For each patient with abnormal DRE or hypoechoic lesions, our 21-biopsy protocol included these areas, although we did not specifically biopsy these anomalies. A board-certified pathologist interpreted all slides. Dedicated uropathologists were involved in daily practice for prostate needle biopsy diagnosis.

Create Platinum Slide Series

Fig. 1 Mondor Hospital's 21-sample needle biopsy protocol including sextant biopsy, six biopsies in far lateral peripheral zone, six biopsies in transitional zone, and three in the middle peripheral zone.

For patients undergoing radical prostatectomy, tumour grade, positive surgical margins, and extracapsular extension were assessed. Prostate specimens were serially sectioned and totally submitted. Tumour volume in cm³ was calculated for every prostatic specimen as a summation of all tumour nodules using. We defined clinically significant tumours according to Epstein et al as a tumour volume of greater than 0.5 cm³ or less than 0.5 cm³ plus a Gleason score of 7 or greater and/or pT3 [6].

Statistical analysis was performed using Statview 5.0 (SAS Institute, Cary, NC). Continuous covariates (age) were compared using Student's t test and covariates without normal distribution (PSA, prostate volume, PSA density) were compared using nonparametric tests (Mann-Whitney U test, Kruskall-Wallis test). In all analyses, two-sided hypothesis testing was carried out with probability values less than 0.05 deemed significant. Cox proportional-hazard regression was used to carry out the multivariate analysis.

3. Results

A total of 231 patients underwent repeated 21-core prostate biopsies. The baseline characteristics of the cohort are seen in Table 1. The mean number of repeated 21-core biopsies procedures was 2.4. Only eight (3.4%) of the 231 patients reported adverse events after the repeat extended 21-sample procedure, with fever and prostatitis in three patients, acute urinary retention in four patients, and rectal bleeding in one patient.

4. Discussion

The ideal strategy for prostate biopsy procedure has yet to be fully elucidated. The sextant biopsy (medial peripheral zone) proposed by Hodge et al is associated with a significant false-negative rate [15]. The prevalence of false-negative sextant biopsy ranges between 20% and 33% [15], and [17]. Recent studies suggest that modifications of the standard sextant biopsy technique by increasing the number of cores obtained or expanding the number of regions sampled may improve the detection of PCa at biopsy. Eskew et al observed that a five-region biopsy method that incorporated lateral and midline biopsy with traditional sextant cores improved the diagnostic yield by 35% [3]. Presti et al investigated a 12-core biopsy strategy, including sextant biopsies and laterally directed sextant biopsies, in a multipractice community study involving 2299 men. The laterally directed sextant biopsies detected 83% of cancers [4]. For urologists, there is a not yet clear recommendation for the follow up of patients with negative prostate biopsies. In this study, we repeated prostate biopsies for young patients with PSA progression or elevated PSA, for whom in multivariate analysis ASAP, prostate volume, and PSA density were predictors of cancer risk. However, a prospective study with a long follow-up should be conducted to answer this question.

In this study, we analyzed the performance of repeated extensive biopsies in 231 patients who had negative 21-sample biopsy procedure and who were at increased risk for PCa. Our overall diagnostic yield was 25.1%, which is similar to the 13.5% to 34% yield in other series [5], and [16]. Selection of high-risk patients at the discretion of the physician represents a bias of this study. However, an important difference is that these other series included patients who had only prior sextant or 12-sample biopsies. In our series, all patients underwent prior extensive 21-sample biopsies. We noted yields of 13% and 22.1% when 6 and 12 cores were taken respectively. In the literature, there is a debate on the TZ biopsies. In our repeated biopsies, we found that 12% of positive biopsies were on the transitional zone or on midline peripheral zone. This also leads us to propose these additional biopsies.

In terms of determining the ideal number of cores to obtain in difficult diagnostic cases, in our series, 88% of cancers were identified in the peripheral zone (1 biopsies). One case (1.7%) was identified from the TZ-only cores and the indication of TZ biopsies alone could be discussed. According to published reports, PCa is diagnosed on the basis of TZ biopsies only in 1.8% to 8% of cases [7], [8], [9], and [10]. In these studies, TZ biopsies were indicated in patients with prior negative biopsy procedures and elevated serum PSA levels associated with an enlarged, non-nodular prostate [7]. The prevalence of positive TZ biopsies was 1.5% in 274 men who underwent sextant plus TZ biopsies for elevated PSA levels, but this rate increased to 9.5% in the 116 patients who previously had negative sextant biopsies [7]. The latter prevalence was slightly lower in our study (1.7%), in which TZ biopsies were systematically performed since the first biopsy procedure. Recently, Walz et al [23] reported a high PCa detection rate of 41% with a 24-core protocol. Similarly to the initial biopsy protocol, it has been proven that more time and effort should be spent on lateral biopsies, which increase the cancer detection rate, whereas parasagittal biopsy provides a low yield on repeat biopsy [24].

PIN and ASAP in the initial biopsy were associated with a positive repeated biopsy rate of 6% and 42% in our patients, respectively. This rate is consistent with those in other series of repeated biopsy in which initial biopsy findings indicated PIN or ASAP. It is also important to note that patients with PIN detected on a first extended protocol have the same risk as men with a benign (no PIN or ASAP) biopsy of having PCa on the repeated biopsies (16% vs 17%) [18]. Consequently, some authors conclude that there is no need to perform repeat biopsy within the first year on men with PIN unless there are other factors worrisome for cancer, such as PSA increase [2]. Moreover, the small number of patients with PIN or ASAP is a limitation in our study. We are unable to assess whether presence of PIN and/or ASAP represents a risk factor for PCa on extended biopsy.

When we analyzed factors that may be associated with positive biopsy in our group, PSA density (PSAD) was found to be a strong predictor. Some authors have reported that larger prostates have a decreased rate of cancer detection. In our series, cancer was detected in only 14% of prostates larger than 60 cc, compared with 27% in those with less than 40 cc. Many larger prostates have elevated PSA, which is not secondary to a malignant process but rather a reflection of increased size, and in particular TZ size. While undersampling of larger prostates may explain this discrepancy, it is also possible that there is a lower prevalence of cancer in these large organs composed mostly of benign tissue [6]. Another possibility is the use of nomogram: Karakiewicz's group reported several nomograms to predict the risk of PCa in extended repeat biopsies and the risk of positive biopsies in the TZ [19], and [20]. These powerful and attractive statistical models seem very useful for clinicians but their uses are not widely accepted.

Some authors suggested that increasing the number of biopsies can lead to a treatment of 3% to 27% of clinically insignificant tumours [21], and [22]. In our series, 85.2 % of cancers diagnosed by repeated 21-core biopsy were significant (tumour volume >0.5 cc, Gleason ≥7, and/or pT3). It appears that in our series the detection of clinically insignificant cancers with repeated saturation biopsy is similar to that in other series in which the diagnosis of PCa was made with fewer biopsy cores (6–18 cores). Thus, it does not appear to increase the detection of insignificant cancers. Moreover, different studies have demonstrated that the use of the extended biopsy procedure has been beneficial in the pretreatment decision-making process, because an increased number of biopsies increase the Gleason concordance. In our study, no significant difference existed between the biopsy and prostatectomy Gleason (6.1 vs 6.7). Different authors [25], and [26] have demonstrated that, with a more extended biopsy procedure, the risk of significant upgrading decreases because of higher sampling density and more accurate pathologic biopsy evaluation. The two largest published cohorts [27], and [28] showed a rate of overall Gleason sum upgrading of 29.3% and 32.6% and a rate of significant upgrading of 32% and 28.2%, respectively. Prospective studies are needed for the evaluation of the aggressiveness of PCa detected on extended repeat biopsies in terms of progression and overall survival.

Djavan et al reported in 2001 an original work on the risk of PCa on repeated biopsies performed 6 wk after a negative set [29]. They found that cancer detection rates on biopsies 1, 2, 3, and 4 were 22% (231 of 1051), 10% (83 of 820), 5% (36 of 737) and 4% (4 of 94), respectively, and that 58.0%, 60.9%, 86.3%, and 100% of patients who had a radical prostatectomy had organ-confined disease on biopsies 1, 2, 3, and 4, respectively. There are some differences between this study and the present report: First the delay between the first and the subsequent biopsies is different; second, the definition of clinically significant PCa was not used. Finally, one question remains unanswered: what is the evolution of PCa detected at the biopsy 3, 4, or 5. We need more data on the aggressiveness of PCa detected on extended repeat biopsies in terms of progression and overall survival. Probably, the use of repeated extended biopsies can lead to detection of intracapsular PCa for which active surveillance can represent a potential option instead of radical treatment.

5. Conclusion

A negative first extended biopsy protocol should not reassure a patient of not having prostate cancer: second, third, and fourth extended 21-sample needle procedures led to a cancer detection rate of 18%, 17%, and 14% respectively. Cancer detected at these sets of biopsies appeared to be intracapsular disease in 75% of the cases. These were considered as significant cancer, mostly due to a Gleason score greater or equal to 7 in 85%.
Author contributions: Alexandre de la Taille had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: De la Taille, Salomon.

Acquisition of data: Bastien, Campos, Allory, Abbou, Patard, Hoznek, Yiou.

Analysis and interpretation of data: Terry, Vacherot, Nicolaiew, Robert.

Drafting of the manuscript: De la Taille, Campos.

Critical revision of the manuscript for important intellectual content: De la Taille, Campos, Robert, Nicolaiew.

Statistical analysis: De la Taille, Nicolaiew.

Obtaining funding: None.

Administrative, technical, or material support: De la Taille.

Supervision: De la Taille.

Other (specify): None.

Financial disclosures: I certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/ affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: None.