- Email: [email protected]
Prospective Evaluation of an Extended 21-Core Biopsy Scheme as Initial Prostate Cancer Diagnostic Strategy
EUROPEAN UROLOGY 65 (2014) 154–161
available at www.sciencedirect.com journal homepage: www.europeanurology.com
Prostate Cancer
Prospective Evaluation of an Extended 21-Core Biopsy Scheme as Initial Prostate Cancer Diagnostic Strategy Guillaume Ploussard y,*, Nathalie Nicolaiew y, Charles Marchand, Ste´phane Terry, Francis Vacherot, Dimitri Vordos, Yves Allory, Claude-Cle´ment Abbou, Laurent Salomon, Alexandre de la Taille INSERM U955 Equipe 7, Departments of Urology and Pathology, CHU Henri Mondor, APHP, Cre´teil, France
Article info
Abstract
Article history: Accepted May 28, 2012 Published online ahead of print on June 9, 2012
Background: The debate on the optimal number of prostate biopsy core samples that should be taken as an initial strategy is open. Objective: To prospectively evaluate the diagnostic yield of a 21-core biopsy protocol as an initial strategy for prostate cancer (PCa) detection. Design, setting, and participants: During 10 yr, 2753 consecutive patients underwent a 21-core biopsy scheme for their first set of biopsy specimens. Intervention: All patients underwent a standardized 21-core protocol with cores mapped for location. Outcome measurements and statistical analysis: The PCa detection rate of each biopsy scheme (6, 12, or 21 cores) was compared using a McNemar test. Predictive factors of the diagnostic yield achieved by a 21-core scheme were studied using logistic regression analyses. Results and limitations: PCa detection rates using 6 sextant biopsies, 12 cores, and 21 cores were 32.5%, 40.4%, and 43.3%, respectively. The 12-core procedure improved the cancer detection rate by 19.4% ( p = 0.004), and the 21-biopsy scheme improved the rate by 6.7% overall ( p < 0.001). The six far lateral cores were the most efficient in terms of detection rate. The diagnostic yield of the 21-core protocol was >10% in prostates with volume >70 ml, in men with a prostate-specific antigen level < 4 ng/ml, with a prostatespecific antigen density (PSAD) <0.20 ng/ml per gram. A PSAD <0.20 ng/ml per gram was the strongest independent predictive factor of the diagnostic yield offered by the 21-core scheme ( p < 0.001). The 21-core protocol significantly increased the rate of PCa eligible for active surveillance (62.5% vs 48.4%; p = 0.036) than those detected by a 12-core scheme without statistically increasing the rate of insignificant PCa ( p = 0.503). Conclusions: A 21-core biopsy scheme improves significantly the PCa detection rate compared with a 12-core protocol. We identified a cut-off PSAD (0.20 ng/ml per gram) below which an extended 21-core scheme might be systematically proposed to significantly improve the overall detection rate without increasing the rate of detected insignificant PCa. # 2012 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Keywords: Prostate cancer Biopsy Detection rate Core number Low risk Insignificant
y These authors contributed equally to this article. * Corresponding author. INSERM U955 Eq07, Department of Urology, CHU Mondor, 51 avenue du Mare´chal de Lattre de Tassigny, 94000 Cre´teil, France. Tel. +33 149812254; Fax: +33 149812568. E-mail address: [email protected] (G. Ploussard).
0302-2838/$ – see back matter # 2012 European Association of Urology. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.eururo.2012.05.049
EUROPEAN UROLOGY 65 (2014) 154–161
1.
Introduction
155
prophylaxis and were prescribed enemas at 1 d and again at 3 h before the procedure. Three dedicated surgeons performed all biopsies. The
The optimal number of biopsy cores that should be taken as an initial biopsy strategy remains a controversial topic in the diagnosis of prostate cancer (PCa). Since its introduction by Hodge et al, random systematic ultrasound-guided transrectal needle biopsy has significantly improved the diagnosis of PCa in terms of detection rate and pathologic characterization of PCa before treatment decision making [1]. Studies have demonstrated that a traditional sextant technique may miss substantial numbers of PCas and that additional sampling of the lateral peripheral zone may increase the diagnostic yield [2–4]. To date, extended biopsy as defined by the National Comprehensive Cancer Network (sextant biopsies with at least four additional cores from the lateral peripheral zones) is clearly recommended at first biopsy [5,6]. However, the false-negative rate remains substantial. Several authors have already shown the benefit of saturation biopsies as an initial strategy [5,7–9], whereas other teams did not recommend saturation biopsies for cancer detection improvement [10–12]. One possible reason for these conflicting results is that accuracy of biopsy schemes depends on different parameters such as prostate volume, prostate-specific antigen (PSA) level, and digital rectal examination (DRE), suggesting the adaptation of the biopsy scheme to each individual patient [3]. The location of targeted cores (eg, far lateral peripheral zone) also must be taken into account. The main end point of our long-term prospective trial was to study the detection rate of PCa according to the biopsy-core number. An intermediate analysis based on the first 1000 consecutive patients was published in 2007 [7]. We aimed to confirm, or not, these findings after a 10-yr follow-up and to perform subgroup analyses.
21-sample biopsy protocol included prostate ultrasound examination to evaluate the prostate volume using the prolate ellipsoid formula. All patients received local anesthesia by injection of 5 ml 2% lidocaine into each neurovascular bundle through a 22-G spinal needle. A 18-G biopsy needle and a spring-loaded biopsy gun that could collect 17-mm-long tissue cores were used. The biopsies were performed in the following order: six sextant biopsies (standard 458 angle), then three biopsies in each peripheral zone (808 angle), then three biopsies in each transition zone (TZ), and finally three biopsies in the midline peripheral zone (Fig. 1). Thus the six-core scheme included the sextant biopsies. The 12-core scheme added the six additional lateral peripheral biopsies. The 21-core scheme added the three midline cores and the six TZ cores. Each prostate core was given a specific number according to the biopsy protocol and mapped for location. Each core was placed in its own container and analyzed separately. Two senior uropathologists reviewed the cores. Prebiopsy clinical and biologic data were abstracted from a computerized prospective database. Findings from pathologic assessment included the length of cores, the number of positive cores and their location, the percentage of cancer involvement in any positive core, and the biopsy Gleason score. Insignificant PCa was defined according to the Epstein criteria: PSA density (PSAD) 0.15 ng/ml per gram, Gleason score 6, fewer than three positive cores, and <50% cancer involvement in any core [13,14]. PCa eligible for active surveillance was defined according to two common definitions: (1) PSA <10 ng/ml, Gleason score 6, clinical T1c stage, and fewer than three positive cores; and (2) PSA <10 ng/ml, Gleason score 6, PSAD <0.20 ng/ml per gram, clinical T1c–T2a stage, and fewer than three positive cores. The main end point of this trial was to study the detection rate of PCa according to the biopsy-core number. The PCa detection rate of the 21-core protocol was compared with the rate from 6- and 12-core biopsy schemes. Results were also shown as percentage improvement in cancer detection rate. Due to the dependence of these cohorts, the McNemar test was used. Statistical significances were confirmed by the Cochran Q, W Kendall, and Friedman tests. Predictive factors of the diagnostic yield achieved by the 12-core scheme (PCa detected in the six far lateral cores, but not in the six sextant biopsy specimens) and 21-core scheme (PCa detected by the three midline biopsies or the six TZ cores, but not by the
2.
Materials and methods
12-core protocol) were studied using logistic regression taking into account PSA, PSAD, age, and clinical stage.
Between December 2001 and December 2011, 2753 consecutive patients
Clinicopathologic characteristics of PCa diagnosed or missed by each
suspicious for PCa prospectively underwent an extended 21-core biopsy
biopsy scheme were also assessed and compared. The normal distribu-
protocol as a first set of biopsies. Indications for prostate biopsy were
tion of continuous variables was confirmed prior to analysis, and
(1) abnormal DRE, regardless of PSA level; (2) a PSA level >4 ng/ml (or
continuous variables were reported as mean and standard deviation. For
3 ng/ml in patients <60 yr); and (3) a free:total PSA ratio (%fPSA) <10%.
independent parameters, the Student t test was used for quantitative
All patients had undergone a standardized 21-core biopsy protocol as
variables and the chi-square test (or a Fisher exact test, as appropriate)
previously described [9]. Clinical stage was determined prior to biopsy
was used for qualitative variables. Statistical analyses were performed
by the same three experienced urologists who subsequently performed
using SPSS software (IBM Corp, Armonk, NY, USA). The limit of statistical
the biopsy procedure. Patients were given a fluoroquinolone antibiotic as
significance was defined as p < 0.05.
[(Fig._1)TD$IG]
Fig. 1 – The 21-core biopsy scheme: six sextant biopsies (Sext) (standard 458 angle), three biopsies in each peripheral zone (Lat) (808 angle), three biopsies in each transition zone (TZ), and three biopsies in the midline (Mid) peripheral zone.
156
EUROPEAN UROLOGY 65 (2014) 154–161
Table 1 – Clinical, biologic, and pathologic characteristics of the entire patient cohort
Table 2 – Prostate cancer detection of the 6-, 12-, and 21-core biopsy schemes and their diagnostic yield compared with the previous scheme
n = 2753 n = 2753 Mean age, yr (SD) Age <60, % Mean prostate volume, ml (SD) Prostate volume >50 ml, % Mean PSA level, ng/ml (SD) PSA level <4 ng/ml, % PSA level >10 ng/ml, % Mean free-to-total PSA, % (SD) Mean PSAD, ng/ml per gram (SD) PSAD >0.20 ng/ml per gram, % Clinical stage >T1c Mean core length, mm (SD) HG-PIN, no. (%) ASAP, no. (%) PCa cases, no. (%) Mean positives cores, no. (SD) >2 positive cores, % Biopsy Gleason score, no. (%): 6 7 8 9 Dominant pattern 4 (%) Mean percent age of core involvement (SD)
64.2 (7.8) 31.6 46.4 (25.3) 35.5 12.5 (72) 10.8 22.9 16.3 (8.5) 0.296 (1.6) 36.2 318 (11.6) 11.9 (2.2) 65 (2.4) 37 (1.3) 1191 (43.3) 6.3 (5.3) 33.5 660 (55.5) 443 (37.2) 66 (5.6) 22 (1.9) 221 (19.4) 30.5 (20.8)
SD = standard deviation; PSA = prostate-specific antigen; PSAD = prostatespecific antigen density; HG-PIN = high-grade prostatic intraepithelial neoplasia; ASAP = atypical small acinar proliferation; PCa = prostate cancer.
3.
Results
The main characteristics of the patient cohort are listed in Table 1. The PCa detection rate of the six-core protocol was 32.5% (Table 2). This rate was increased by 19.4% by using the additional six far lateral cores (ie, the 12-core protocol). The 21-core protocol increased this rate by 6.7%. The
Diagnosis based on: 6-core protocol 12-core protocol 21-core protocol Diagnostic yield, no. 12 vs 6cores 21 vs 6 cores 21 vs 12 cores
PCa cases, no.
PCa detection rate
896 1111 1191
32.5 40.4 43.3 +19.4 +24.8 +6.7
McNemar test p value
0.004 <0.001 <0.001*
PCa = prostate cancer. p value also <0.001 by the Cochran Q, W Kendall, and Friedman tests.
*
diagnostic yield offered by the 12-core protocol was statistically significant compared with the detection rate offered by the use of only six sextant biopsies ( p = 0.004). The diagnostic yield offered by the 21-core protocol was statistically significant compared with the detection rate offered by the use of the 12-core protocol (six sextant biopsies and six far lateral cores; p = 0.004). Adding three midline zone cores to the 12-core protocol increased the detection rate by 2.9% (41.6%). Adding six TZ cores to the 12-core protocol increased the detection rate by 4.7% (42.4%). PCa was significantly associated with an abnormal DRE ( p < 0.001; odds ratio: 2.7), higher PSA level (19.5 vs 7.2 ng/ml; p < 0.001), higher PSAD (0.47 vs 0.17; p < 0.001), lower %fPSA (0.14 vs 0.18; p < 0.001), and lower prostate volume (41.6 ml vs 49.9 ml; p < 0.001). All these factors remained significant in a multivariable model. Characteristics of PCa diagnosed by each biopsy protocol are listed in Table 3. Increasing the biopsy-core number resulted in diagnosing PCa with more favorable prebiopsy
Table 3 – Clinicopathologic features of prostate cancers detected by each six-core region (sextant, far lateral, and transition zone)
Cases, no. Mean PSA, ng/ml PSA >10, % Mean free-to-total PSA, % Mean PSAD, ng/ml per gram PSAD >0.20, % Mean age, yr Clinical stage >T1c, % Prostate volume, ml >50 ml, % Positive cores, no. Mean percentage of core involvement Biopsy Gleason score >6, % Insignificant PCa, % PCa eligible for AS, %: PSA <10 ng/ml; GS 6, T1c, <3 PC PSA <10 ng/ml, GS 6, PSAD <0.20 ng/ml per gram, T1c–T2a, <3 PC
PCa detected by a 6-core protocol
PCa detected by a 12-core protocol
PCa detected by a 21-core protocol
896 23.4 36.2 13.9 0.558 57.6 66.4 18.5 40.9 25.1 7.7 33.3 52.8 4.2
1111 20.4 33.4 14.1 0.489 54.0 66.1 17.3 41.4 26.5 6.7 31.1 46.4 7.2
1191 19.6 32.5 14.1 0.469 52.1 66.1 17.0 41.6 26.9 6.3 30.5 44.4 8.3
18.2 13.7
21.2 16.6
11.0 8.6
PCa = prostate cancer; PSA = prostate-specific antigen; PSAD = prostate-specific antigen density; AS = active surveillance; GS = Gleason score; PC = positive cores.
157
EUROPEAN UROLOGY 65 (2014) 154–161
Table 4 – Comparisons of clinicopathologic features of prostate cancers detected by a six-core protocol with those detected only by the six additional far lateral cores (12-core) and only by the last nine additional cores (midline and transition zone) Diagnosis by 6-core protocol: PCa cases (n = 896) (1)
Diagnostic yield by 12-core (vs 6-core): PCa cases (n = 215) (2)
Mean percentage core involvement (SD) Median Biopsy Gleason score >6, %
33.3 (20.5) 30.0 52.8
21.9 (19.2) 15.0 19.4
Dominant grade 4 pattern, %
23.6
Mean positive cores, no. (SD) Median Insignificant PCa, % PCa eligible for AS, %: PSA <10 ng/ml; GS 6, T1c, <3 PC PSA <10 ng/ml, GS 6, PSAD <0.20 ng/ml per gram, T1c–T2a, <3 PC
7.7 (5.4) 6.0 4.2
5.4
Diagnostic yield by 21-core (vs 12-core): PCa cases (n = 80) (3) 22.3 (20.5) 15.0 16.7 8.0
2.17 (1.53) 2.0 20.9
1.44 (0.71) 1.0 25.3
11.0
48.4
62.5
8.6
34.7
53.6
p value
(1) (2) (1) (2) (1) (2) (1) (2) (1) (2)
vs vs vs vs vs vs vs vs vs vs
(2): (3): (2): (3): (2): (3): (2): (3): (2): (3):
<0.001 0.874 <0.001 0.592 <0.001 0.425 <0.001 <0.001 <0.001 0.503
(1) vs (2): <0.001 (2) vs (3): 0.036 (1) vs (2): <0.001 (2) vs (3): 0.006
PCa = prostate cancer; PSA = prostate-specific antigen; PSAD = prostate-specific antigen density; AS = active surveillance; GS = Gleason score; PC = positive cores.
parameters and biopsy pathologic features. The percentage of low-risk PCas eligible for active surveillance increased twofold from the six-core to the 21-core biopsy scheme (Table 4). PCas detected by the six-core protocol were significantly more aggressive than those detected only by the 12- or 21-core protocol. Significant differences were reported in terms of percentage of core involvement, rate of high biopsy Gleason score, number of positive cores, rate of insignificant PCas, and percentage of men eligible for active surveillance ( p < 0.001 for all). The biopsy Gleason score and the percentage of core involvement were
comparable between PCa detected by the 12-core protocol diagnostic yield (n = 215) and those detected by the 21-core protocol (n = 80). Nevertheless, PCas detected only by the 21-core protocol involved significantly fewer biopsy cores (1.44 vs 2.17; p < 0.001) and were more likely to be eligible for active surveillance (62.5% vs 48.4%; p = 0.036) than those detected by the 12-core scheme. Nevertheless, no difference in insignificant PCa was found between the two protocols. To assess the ideal location of biopsy cores, we studied the characteristics of PCa missed by each six-core location: six sextant, six far lateral, and six TZ cores (Table 5). Using
Table 5 – Clinicopathologic features of prostate cancer missed by each six-core region (sextant, far lateral, and transition zone) and their comparisons*
Cases, no. (%) Mean PSA level, ng/ml PSA >10, % Mean free-to-total PSA, % Mean PSAD, ng/ml per gram PSAD >0.20, % Mean age, yr Clinical stage >T1c, mean no. Mean prostate volume, ml >50 ml, % Mean positive cores, no. Mean percentage of core involvement Biopsy Gleason score >6, % Insignificant PCa, % PCa eligible for AS, %: PSA <10 ng/ml; GS 6, T1c, <3 PC PSA <10 ng/ml, GS 6, PSAD <0.20 ng/ml per gram, T1c–T2a, <3 PC
PCa missed by six sextant biopsies
PCa missed by six far lateral biopsies
PCa missed by six TZ biopsies
311 (25.8) 8.2 22.5 14.9 0.23 36.9 65.1 13.2 43.6 31.6 2.0 22.0 20.1 21.2
254 (21.0) 10.0 23.4 16.5 0.28 31.1 66.3 11.4 44.8 30.6 2.0 22.2 19.8 21.3
399 (33.1) 9.5 22.8 16.1 0.26 38.7 65.7 14.3 44.5 32.0 2.3 23.7 21.6 20.2
50.8 39.6
50.6 44.2
47.2 36.9
PCa = prostate cancer; TZ = transition zone; PSA = prostate-specific antigen; PSAD = prostate-specific antigen density; AS = active surveillance; GS = Gleason score; PC = positive cores. * p > 0.1 for all statistical comparisons.
158
EUROPEAN UROLOGY 65 (2014) 154–161
Table 6 – Detection rates of the 6-, 12-, and 21-core biopsy schemes stratified by prebiopsy parameters (ie, prostate volume, prostate-specific antigen [PSA], PSA density [PSAD], and clinical stage), and their improvement of diagnostic yield
Prostate volume, % <50 ml >50 ml >70 ml PSA, ng/ml, % >10 ng/ml <10 ng/ml <4 ng/ml PSAD, ng/ml per gram, % <0.15 <0.20 >0.20 Clinical stage, % T1c T2–4 Age, yr, % <65 >65 Combined factors**, % Yes No * **
Detection rate by 6-core scheme
Detection rate by 12-core scheme (plus diagnostic yield, %)
Detection rate by 21-core scheme (plus diagnostic yield, %) *
36.9 22.3 16.8
45.0 (+18.1) 29.3 (+23.8) 21.6 (+22.2)
48.0 (+6.2) 31.9 (+8.4) 24.4 (+11.3)
51.6 27.0 14.0
59.0 (+12.5) 34.9 (+22.7) 20.9 (+32.8)
61.6 (+4.2) 37.9 (+7.9) 24.7 (+15.3)
17.6 20.9 50.4
24.5 (+27.8) 28.3 (+25.9) 58.8 (+14.2)
27.3 (+10.4) 31.7 (+10.8) 61.1 (+3.8)
29.9 53.0
37.7 (+20.5) 61.3 (+13.5)
40.5 (+7.1) 64.5 (+5.0)
26.3 40.1
33.4 (+21.1) 48.8 (+17.8)
35.8 (+6.8) 52.3 (+6.7)
14.4 37.1
20.1 (+28.6) 45.5 (+18.3)
22.8 (+11.8) 48.6 (+6.4)
p < 0.001 (McNemar test) for all comparisons of PCa detection rates between 12- and 21-core protocols. Combined factors: PSAD <0.20 plus PSA <10 plus T1c plus prostate volume >50 ml.
six sextant cores only, six far lateral cores only, or six TZ cores only would have missed 25.8%, 21.0%, or 33.1% of PCa, respectively. Thus the six far lateral cores were the most efficient cores in terms of detection rate. There was a trend toward less aggressive PCa among PCas that would have been missed by a six-core protocol than those missed by the
six far lateral cores or the TZ cores; however, differences were not significant. Table 6 shows the detection of each biopsy protocol and the diagnostic yield of the 12- and 21-core protocols in terms of various prebiopsy parameters (ie, PSA, PSAD, volume, and clinical stage). The overall detection rate was
Table 7 – Prebiopsy predictive factors of the diagnostic yield offered by a 21-core protocol as compared with a 12-core protocol by univariable and multivariable analyses Diagnostic yield 21- vs 12-core (n = 80), %
PSA level >10 ng/ml PSAD >0.20 ng/ml per gram Clinical stage >T1c Volume >50 ml Age >65 yr Multivariate analysis Categorical variables PSA level >10 ng/ml PSAD >0.20 ng/ml per gram Clinical stage >T1c Volume >50 ml Age >65 yr Simplified model PSAD >0.20 ng/ml per gram Clinical stage >T1c Age >65 yr Continuous variables PSA level PSAD Clinical stage >T1c Volume Age
No
Yes
33.4 54.0 17.3 26.5 55.3
20.0 27.5 12.5 31.9 55.0
p value
OR (95% CI)
0.013 <0.001 0.354 0.327 0.969
0.499 0.323 0.684 1.298 0.991
(0.28–0.88) (0.18–0.56) (0.35–1.35) (0.77–2.20) (0.63–1.56)
0.730 0.001 0.502 0.756 0.983
1.139 0.309 0.766 0.911 1.006
(0.54–2.38) (0.16–0.62) (0.35–1.67) (0.51–1.64) (0.61–1.66)
<0.001 0.512 0.970 0.393 0.007 0.699 0.776 0.889
OR = odds ratio; CI = confidence interval; PSA = prostate-specific antigen; PSAD = prostate-specific antigen density.
0.335 (0.19–0.58) 0.771 (0.36–1.68) 1.010 (0.62–1.66)
EUROPEAN UROLOGY 65 (2014) 154–161
higher in men with a low prostate volume, high PSA level and PSAD, and an abnormal DRE, regardless of number of biopsy cores. The diagnostic yield of the additional biopsy cores (12 or 21 cores) increased markedly when the prostate volume increased and when the PSA level and PSAD decreased. The diagnostic yield of the 21-core protocol was >10% in prostate >70 ml, in men with a PSA <4 ng/ml, and with a PSAD <0.20 ng/ml per gram. This yield was statistically significant ( p < 0.001). Predictive factors of the diagnostic yield offered by the 21-core protocol compared with the 12-core scheme are presented in Table 7. In univariable analysis, PSA level and PSAD were significant predictors of such a diagnostic yield. In multivariable analysis taking into account PSA, PSAD, clinical stage, volume, and age, the only independent predictive factor of the diagnostic yield offered by 21 biopsy cores was a PSAD <0.20 ng/ml per gram ( p < 0.001). This independent value remained significant regardless of prostate-volume cut-off (ie, 50, 60, or 70 ml) and of PSA levels (ie, 4 or 10 ng/ml) used in the multivariable analysis. 4.
Discussion
To our knowledge, our cohort was the largest of any prospective study published for this indication [7]. Overall, we found a continuum of improvement of cancer detection in the current study with an increasing number of cores. These findings confirmed our previous results and those published in the most recent prospective study [3,7]. The greater the number of biopsy cores taken, the larger the sampling of the prostate, and thus the lower the risk of missing a cancer. Nevertheless, accuracy of biopsy schemes depends on different parameters. The primary studied parameter is prostate volume. Sampling accuracy tends to decrease progressively with an increasing prostate volume [3,6,15]. In our study, as compared with the diagnostic yield offered by 12-core protocol, which stopped increasing at a prostate volume cut-off of 50 ml, the diagnostic yield of the 21-core protocol kept increasing with the prostate volume. Other parameters such as PSA and DRE influenced the detection rate and the yield provided by extended biopsy schemes. These variations in diagnostic yield offered by saturation strategies were highlighted by Scattoni et al. [3]. They showed that the most advantageous biopsy scheme, in terms of number and location of cores, varied as a function of age, prostate volume, and DRE. Our results confirmed these data, as shown in Table 6. The diagnostic yield of saturation was highly modified by the presumed risk of PCa that could be evaluated by DRE, PSA level, and prostate volume. For example, the diagnostic yield of the 12-core protocol ranged from 12.5% to 33% according to the PSA level. The diagnostic yield of the 21-core scheme was increased more than threefold in patients with PSA level <4 ng/ml compared with those having a PSA level >10 ng/ml (4.2% vs 15.2%). As compared with Scattoni et al, we chose to compare only 12 versus 21 cores. Thus our flowchart was simple, using only PSAD as the decisionmaking variable. Using multivariable logistic regression, we
159
identified a PSAD cut-off of 0.20 ng/ml per gram, below which an extended 21-core scheme might be systematically proposed to significantly improve overall detection. We think that this dichotomy is easier to use for the urologist in clinical practice. PSAD was previously identified as an interesting predictive factor when deciding on the number of prostate biopsy cores required [16]. Our findings also highlighted the importance of both number and location of biopsy cores. First, we suggested that the number of biopsy cores was a more important factor of high-risk PCa detection improvement than their exact location. No significant differences in terms of highgrade cancers were seen between PCa missed by the sextant biopsies, by the six far lateral biopsy specimens, or by the TZ specimens. Thus the rate of high-risk PCa missed by a six-core scheme did not differ according to the location of these six cores. Second, in line with previously published data, we found that the far lateral peripheral cores were the most efficient for diagnosis and added the most to the cancer detection rate compared with sextant and TZ biopsies [2,3,17,18]. Only 21% of PCa would have been missed by the use of only six lateral cores compared with 26% and 33% if only sextant or TZ biopsies, respectively, were used. Thus the location of sampled tissue appeared highly relevant when considering the overall detection rate including low- and high-risk PCas. The location of biopsy cores was not as important for the detection of high-grade PCa when compared with the biopsy-core number. Nevertheless, sampling the most lateral periphery appeared highly relevant to detecting as many cancers as possible. In our previous work, the diagnostic yield offered by a 18-core protocol (12-core plus TZ) was compared with that offered by a 21-core protocol (12-core plus TZ plus midline). The conclusion was that the diagnostic yield offered by midline zone biopsy specimens was not significant compared with the 18-core procedure. Unfortunately, the diagnostic yield offered by a 15-core protocol (12-core plus midline) has not been studied. In the present study, the diagnostic yield offered by midline cores was moderate (3%) but significant as compared with the 12-core protocol. This yield was still lower than that offered by the six TZ biopsies, but we cannot conclude if this discrepancy might be explained by the different core location or just by a lower number of additional cores (ie, three vs six). Even if the detection rate increased with an increasing biopsy-core number, it is important to know if such an improvement in detection is clinically relevant. The counterpart is that the saturation biopsy strategies have been hypothesized to increase the potential risk of overtreating patients whose tumors pose very low mortality risk. However, our 21-core protocol did not statistically increase the rate of detected insignificant PCa as defined by the Epstein criteria [13,14]. Moreover, the role of prostate biopsies has changed. The actual importance of prostate biopsies has evolved from purely cancer detection to investigating how biopsy results can assist clinical management for patients. The inclusion of patients in active surveillance protocols emphasizes the necessity of accurate staging strategies. A larger amount of sampled tissue may
160
EUROPEAN UROLOGY 65 (2014) 154–161
achieve a more complete picture of the disease burden. Extended or saturation biopsies have been advocated to detect cancers that standard biopsies miss and also to characterize better PCa volume and prognosis [19–22]. This impact is of great interest in low-risk PCa patients eligible for active surveillance because the need for subsequent reassessment biopsies may be reduced, as might the risk of initial misclassification of the disease [23]. In our study, PCas diagnosed by this 21-core scheme had the same pathologic characteristics as those diagnosed by the 12-core protocol in terms of percentage of cancer involvement per core and biopsy Gleason score, but they were more likely to be eligible for active surveillance due to a lower mean number of positive cores. We previously demonstrated that about 30% of cancers fulfilling the active surveillance criteria in a 12-core protocol, but not in the 21-core scheme, exhibited unfavorable disease characteristics in the radical prostatectomy (RP) specimen [21]. Our previous findings suggested that the 21-core biopsy strategy provided a more accurate staging of PCa and that a more extended biopsy strategy provides a more accurate assessment. The 21-core protocol did not statistically increase the rate of detected insignificant PCa as defined by the updated Epstein criteria. To date, these criteria are the most widely used preoperative criteria, and more recent attempts to predict insignificant PCa after positive biopsies and before surgery are based on this updated definition [13,14]. One limitation of our study was that we could not confirm the clinical significance of cancer because all patients did not undergo RP. This point might be relevant because the Epstein criteria are not perfect and misclassify about 30% of patients who would have unfavorable pathologic features in an RP specimen [14]. Nevertheless, in a previous study analyzing the RP specimens of patients selected from the same cohort, we found that PCa diagnosed only in a 21-core protocol were at least as aggressive as PCa detected by the diagnostic yield offered by a 12-core scheme [24]. The safety of the 21-core protocol was previously studied in our first report [9]. The overall complication rate requiring hospitalization was 3%, including infection (1%), acute urinary retention (1.9%), and rectal bleeding. Macroscopic hematuria occurred in 84% of patients and lasted an average of 3 d. Minor rectal bleeding, lasting an average of 1 d, was reported by 45% of patients. Thus the morbidity rate of the 21-core procedure was close to that reported with the standard 12-core procedure.
have the same pathologic characteristics as those detected by the diagnostic yield offered by the 12-core protocol in terms of percentage of cancer involvement per core and biopsy Gleason score, but they were more likely to be eligible for active surveillance. Nevertheless, the 21-core protocol does not statistically increase the rate of detected insignificant PCa as defined by the Epstein criteria. Author contributions: Guillaume Ploussard 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: Ploussard, Abbou, Salomon, de la Taille. Acquisition of data: Ploussard, Marchand, Nicolaiew, Salomon, de la Taille. Analysis and interpretation of data: Ploussard, Nicolaiew, Terry, Vacherot, Vordos, Allory. Drafting of the manuscript: Ploussard. Critical revision of the manuscript for important intellectual content: Vacherot, Abbou, Salomon, de la Taille. Statistical analysis: Ploussard, Nicolaiew. Obtaining funding: None. Administrative, technical, or material support: None. Supervision: de la Taille. Other (specify): None. Financial disclosures: Guillaume Ploussard certifies 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.
References [1] Hodge KK, McNeal JE, Stamey TA. Ultrasound guided transrectal core biopsies of the palpably abnormal prostate. J Urol 1989;142:66–70. [2] Epstein JI, Walsh PC, Carter HB. Importance of posterolateral needle biopsies in the detection of prostate cancer. Urology 2001;57: 1112–6. [3] Scattoni V, Raber M, Abdollah F, et al. Biopsy schemes with the fewest cores for detecting 95% of the prostate cancers detected by a 24-core biopsy. Eur Urol 2010;57:1–8. [4] Singh H, Canto EI, Shariat SF, et al. Improved detection of clinically significant, curable prostate cancer with systematic 12-core biopsy. J Urol 2004;171:1089–92. [5] Scattoni V, Zlotta A, Montironi R, Schulman C, Rigatti P, Montorsi F. Extended and saturation prostatic biopsy in the diagnosis and characterisation of prostate cancer: a critical analysis of the literature. Eur Urol 2007;52:1309–22.
5.
Conclusions
[6] Chun FK-H, Epstein JI, Ficarra V, et al. Optimizing performance and interpretation of prostate biopsy: a critical analysis of the literature.
A 21-core biopsy scheme improves significantly the PCa detection rate compared with a 12-core protocol. This improvement of diagnostic yield is >10% in patients having a PSAD <0.20 ng/ml per gram, a prostate volume >70 ml, and/or a PSA level <4 ng/ml. We identified a PSAD <0.20 ng/ml per gram as the most predictive factor of the diagnostic yield offered by this 21-core protocol. The biopsy-core number might be selected based on this PSAD cut-off value. PCas diagnosed using this 21-core scheme
Eur Urol 2010;58:851–64. [7] Guichard G, Larre´ S, Gallina A, et al. Extended 21-sample needle biopsy protocol for diagnosis of prostate cancer in 1000 consecutive patients. Eur Urol 2007;52:430–5. [8] Meng MV, Elkin EP, DuChane J, Carroll PR. Impact of increased number of biopsies on the nature of prostate cancer identified. J Urol 2006;176:63–8. [9] De la Taille A, Antiphon P, Salomon L, et al. Prospective evaluation of a 21-sample needle biopsy procedure designed to improve the prostate cancer detection rate. Urology 2003;61:1181–6.
EUROPEAN UROLOGY 65 (2014) 154–161
161
[10] Jones JS, Patel A, Schoenfield L, Rabets JC, Zippe CD, Magi-Galluzzi C.
[18] Philip J, Hanchanale V, Foster CS, Javle P. Importance of
Saturation technique does not improve cancer detection as an
peripheral biopsies in maximising the detection of early prostate
initial prostate biopsy strategy. J Urol 2006;175:485–8.
cancer in repeat 12-core biopsy protocols. BJU Int 2006;98:
[11] Pepe P, Aragona F. Saturation prostate needle biopsy and prostate cancer detection at initial and repeat evaluation. Urology 2007;70: 1131–5. [12] Eichler K, Hempel S, Wilby J, Myers L, Bachmann LM, Kleijnen J. Diagnostic value of systematic biopsy methods in the investigation
559–62. [19] Numao N, Kawakami S, Yokoyama M, et al. Improved accuracy in predicting the presence of Gleason pattern 4/5 prostate cancer by three-dimensional 26-core systematic biopsy. Eur Urol 2007;52: 1663–9.
of prostate cancer: a systematic review. J Urol 2006;175:1605–12.
[20] Descazeaud A, Rubin M, Chemama S, et al. Saturation biopsy
[13] Bastian PJ, Mangold LA, Epstein JI, Partin AW. Characteristics of
protocol enhances prediction of pT3 and surgical margin status
insignificant clinical T1c prostate tumors. A contemporary analysis. Cancer 2004;101:2001–5. [14] Ploussard G, Epstein JI, Montironi R, et al. The contemporary concept of significant versus insignificant prostate cancer. Eur Urol 2011;60:291–303. [15] Brawer MK. The influence of prostate volume on prostate cancer detection. Eur Urol Suppl 2002;1:35–9. [16] Jeong H, Jeong BC, Kwak C, Lee E, Lee SE, Kim TB. A comparison of
on prostatectomy specimen. World J Urol 2006;24:676–80. [21] Ploussard G, Xylinas E, Salomon L, et al. The role of biopsy core number in selecting prostate cancer patients for active surveillance. Eur Urol 2009;56:891–8. [22] Capitanio U, Karakiewicz PI, Valiquette L, et al. Biopsy core number represents one of foremost predictors of clinically significant Gleason sum upgrading in patients with low-risk prostate cancer. Urology 2009;73:1087–91.
prostate cancer detection rates by 12 or 6 core biopsy at different
[23] Abouassaly R, Lane BR, Jones JS. Staging saturation biopsy in
prostate specific antigen densities in Korean men. World J Urol
patients with prostate cancer on active surveillance protocol. Urol-
2008;26:395–400.
ogy 2008;71:573–7.
[17] Presti JC, O’Dowd GJ, Miller MC, et al. Extended peripheral zone
[24] Ouzaid I, Xylinas E, Campeggi A, et al. Contemporary pathologic
biopsy schemes increase cancer detection rates and minimizes vari-
characteristics and oncologic outcomes of prostate cancers missed
ance in prostate specific antigen and age related cancer rates: results
by 6- and 12-core biopsy and diagnosed with a 21-core biopsy
of a community multipractice study. J Urol 2003;169:125–9.
protocol. World J Urol 2013;31:869–74.
available at www.sciencedirect.com journal homepage: www.europeanurology.com
Prostate Cancer
Prospective Evaluation of an Extended 21-Core Biopsy Scheme as Initial Prostate Cancer Diagnostic Strategy Guillaume Ploussard y,*, Nathalie Nicolaiew y, Charles Marchand, Ste´phane Terry, Francis Vacherot, Dimitri Vordos, Yves Allory, Claude-Cle´ment Abbou, Laurent Salomon, Alexandre de la Taille INSERM U955 Equipe 7, Departments of Urology and Pathology, CHU Henri Mondor, APHP, Cre´teil, France
Article info
Abstract
Article history: Accepted May 28, 2012 Published online ahead of print on June 9, 2012
Background: The debate on the optimal number of prostate biopsy core samples that should be taken as an initial strategy is open. Objective: To prospectively evaluate the diagnostic yield of a 21-core biopsy protocol as an initial strategy for prostate cancer (PCa) detection. Design, setting, and participants: During 10 yr, 2753 consecutive patients underwent a 21-core biopsy scheme for their first set of biopsy specimens. Intervention: All patients underwent a standardized 21-core protocol with cores mapped for location. Outcome measurements and statistical analysis: The PCa detection rate of each biopsy scheme (6, 12, or 21 cores) was compared using a McNemar test. Predictive factors of the diagnostic yield achieved by a 21-core scheme were studied using logistic regression analyses. Results and limitations: PCa detection rates using 6 sextant biopsies, 12 cores, and 21 cores were 32.5%, 40.4%, and 43.3%, respectively. The 12-core procedure improved the cancer detection rate by 19.4% ( p = 0.004), and the 21-biopsy scheme improved the rate by 6.7% overall ( p < 0.001). The six far lateral cores were the most efficient in terms of detection rate. The diagnostic yield of the 21-core protocol was >10% in prostates with volume >70 ml, in men with a prostate-specific antigen level < 4 ng/ml, with a prostatespecific antigen density (PSAD) <0.20 ng/ml per gram. A PSAD <0.20 ng/ml per gram was the strongest independent predictive factor of the diagnostic yield offered by the 21-core scheme ( p < 0.001). The 21-core protocol significantly increased the rate of PCa eligible for active surveillance (62.5% vs 48.4%; p = 0.036) than those detected by a 12-core scheme without statistically increasing the rate of insignificant PCa ( p = 0.503). Conclusions: A 21-core biopsy scheme improves significantly the PCa detection rate compared with a 12-core protocol. We identified a cut-off PSAD (0.20 ng/ml per gram) below which an extended 21-core scheme might be systematically proposed to significantly improve the overall detection rate without increasing the rate of detected insignificant PCa. # 2012 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Keywords: Prostate cancer Biopsy Detection rate Core number Low risk Insignificant
y These authors contributed equally to this article. * Corresponding author. INSERM U955 Eq07, Department of Urology, CHU Mondor, 51 avenue du Mare´chal de Lattre de Tassigny, 94000 Cre´teil, France. Tel. +33 149812254; Fax: +33 149812568. E-mail address: [email protected] (G. Ploussard).
0302-2838/$ – see back matter # 2012 European Association of Urology. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.eururo.2012.05.049
EUROPEAN UROLOGY 65 (2014) 154–161
1.
Introduction
155
prophylaxis and were prescribed enemas at 1 d and again at 3 h before the procedure. Three dedicated surgeons performed all biopsies. The
The optimal number of biopsy cores that should be taken as an initial biopsy strategy remains a controversial topic in the diagnosis of prostate cancer (PCa). Since its introduction by Hodge et al, random systematic ultrasound-guided transrectal needle biopsy has significantly improved the diagnosis of PCa in terms of detection rate and pathologic characterization of PCa before treatment decision making [1]. Studies have demonstrated that a traditional sextant technique may miss substantial numbers of PCas and that additional sampling of the lateral peripheral zone may increase the diagnostic yield [2–4]. To date, extended biopsy as defined by the National Comprehensive Cancer Network (sextant biopsies with at least four additional cores from the lateral peripheral zones) is clearly recommended at first biopsy [5,6]. However, the false-negative rate remains substantial. Several authors have already shown the benefit of saturation biopsies as an initial strategy [5,7–9], whereas other teams did not recommend saturation biopsies for cancer detection improvement [10–12]. One possible reason for these conflicting results is that accuracy of biopsy schemes depends on different parameters such as prostate volume, prostate-specific antigen (PSA) level, and digital rectal examination (DRE), suggesting the adaptation of the biopsy scheme to each individual patient [3]. The location of targeted cores (eg, far lateral peripheral zone) also must be taken into account. The main end point of our long-term prospective trial was to study the detection rate of PCa according to the biopsy-core number. An intermediate analysis based on the first 1000 consecutive patients was published in 2007 [7]. We aimed to confirm, or not, these findings after a 10-yr follow-up and to perform subgroup analyses.
21-sample biopsy protocol included prostate ultrasound examination to evaluate the prostate volume using the prolate ellipsoid formula. All patients received local anesthesia by injection of 5 ml 2% lidocaine into each neurovascular bundle through a 22-G spinal needle. A 18-G biopsy needle and a spring-loaded biopsy gun that could collect 17-mm-long tissue cores were used. The biopsies were performed in the following order: six sextant biopsies (standard 458 angle), then three biopsies in each peripheral zone (808 angle), then three biopsies in each transition zone (TZ), and finally three biopsies in the midline peripheral zone (Fig. 1). Thus the six-core scheme included the sextant biopsies. The 12-core scheme added the six additional lateral peripheral biopsies. The 21-core scheme added the three midline cores and the six TZ cores. Each prostate core was given a specific number according to the biopsy protocol and mapped for location. Each core was placed in its own container and analyzed separately. Two senior uropathologists reviewed the cores. Prebiopsy clinical and biologic data were abstracted from a computerized prospective database. Findings from pathologic assessment included the length of cores, the number of positive cores and their location, the percentage of cancer involvement in any positive core, and the biopsy Gleason score. Insignificant PCa was defined according to the Epstein criteria: PSA density (PSAD) 0.15 ng/ml per gram, Gleason score 6, fewer than three positive cores, and <50% cancer involvement in any core [13,14]. PCa eligible for active surveillance was defined according to two common definitions: (1) PSA <10 ng/ml, Gleason score 6, clinical T1c stage, and fewer than three positive cores; and (2) PSA <10 ng/ml, Gleason score 6, PSAD <0.20 ng/ml per gram, clinical T1c–T2a stage, and fewer than three positive cores. The main end point of this trial was to study the detection rate of PCa according to the biopsy-core number. The PCa detection rate of the 21-core protocol was compared with the rate from 6- and 12-core biopsy schemes. Results were also shown as percentage improvement in cancer detection rate. Due to the dependence of these cohorts, the McNemar test was used. Statistical significances were confirmed by the Cochran Q, W Kendall, and Friedman tests. Predictive factors of the diagnostic yield achieved by the 12-core scheme (PCa detected in the six far lateral cores, but not in the six sextant biopsy specimens) and 21-core scheme (PCa detected by the three midline biopsies or the six TZ cores, but not by the
2.
Materials and methods
12-core protocol) were studied using logistic regression taking into account PSA, PSAD, age, and clinical stage.
Between December 2001 and December 2011, 2753 consecutive patients
Clinicopathologic characteristics of PCa diagnosed or missed by each
suspicious for PCa prospectively underwent an extended 21-core biopsy
biopsy scheme were also assessed and compared. The normal distribu-
protocol as a first set of biopsies. Indications for prostate biopsy were
tion of continuous variables was confirmed prior to analysis, and
(1) abnormal DRE, regardless of PSA level; (2) a PSA level >4 ng/ml (or
continuous variables were reported as mean and standard deviation. For
3 ng/ml in patients <60 yr); and (3) a free:total PSA ratio (%fPSA) <10%.
independent parameters, the Student t test was used for quantitative
All patients had undergone a standardized 21-core biopsy protocol as
variables and the chi-square test (or a Fisher exact test, as appropriate)
previously described [9]. Clinical stage was determined prior to biopsy
was used for qualitative variables. Statistical analyses were performed
by the same three experienced urologists who subsequently performed
using SPSS software (IBM Corp, Armonk, NY, USA). The limit of statistical
the biopsy procedure. Patients were given a fluoroquinolone antibiotic as
significance was defined as p < 0.05.
[(Fig._1)TD$IG]
Fig. 1 – The 21-core biopsy scheme: six sextant biopsies (Sext) (standard 458 angle), three biopsies in each peripheral zone (Lat) (808 angle), three biopsies in each transition zone (TZ), and three biopsies in the midline (Mid) peripheral zone.
156
EUROPEAN UROLOGY 65 (2014) 154–161
Table 1 – Clinical, biologic, and pathologic characteristics of the entire patient cohort
Table 2 – Prostate cancer detection of the 6-, 12-, and 21-core biopsy schemes and their diagnostic yield compared with the previous scheme
n = 2753 n = 2753 Mean age, yr (SD) Age <60, % Mean prostate volume, ml (SD) Prostate volume >50 ml, % Mean PSA level, ng/ml (SD) PSA level <4 ng/ml, % PSA level >10 ng/ml, % Mean free-to-total PSA, % (SD) Mean PSAD, ng/ml per gram (SD) PSAD >0.20 ng/ml per gram, % Clinical stage >T1c Mean core length, mm (SD) HG-PIN, no. (%) ASAP, no. (%) PCa cases, no. (%) Mean positives cores, no. (SD) >2 positive cores, % Biopsy Gleason score, no. (%): 6 7 8 9 Dominant pattern 4 (%) Mean percent age of core involvement (SD)
64.2 (7.8) 31.6 46.4 (25.3) 35.5 12.5 (72) 10.8 22.9 16.3 (8.5) 0.296 (1.6) 36.2 318 (11.6) 11.9 (2.2) 65 (2.4) 37 (1.3) 1191 (43.3) 6.3 (5.3) 33.5 660 (55.5) 443 (37.2) 66 (5.6) 22 (1.9) 221 (19.4) 30.5 (20.8)
SD = standard deviation; PSA = prostate-specific antigen; PSAD = prostatespecific antigen density; HG-PIN = high-grade prostatic intraepithelial neoplasia; ASAP = atypical small acinar proliferation; PCa = prostate cancer.
3.
Results
The main characteristics of the patient cohort are listed in Table 1. The PCa detection rate of the six-core protocol was 32.5% (Table 2). This rate was increased by 19.4% by using the additional six far lateral cores (ie, the 12-core protocol). The 21-core protocol increased this rate by 6.7%. The
Diagnosis based on: 6-core protocol 12-core protocol 21-core protocol Diagnostic yield, no. 12 vs 6cores 21 vs 6 cores 21 vs 12 cores
PCa cases, no.
PCa detection rate
896 1111 1191
32.5 40.4 43.3 +19.4 +24.8 +6.7
McNemar test p value
0.004 <0.001 <0.001*
PCa = prostate cancer. p value also <0.001 by the Cochran Q, W Kendall, and Friedman tests.
*
diagnostic yield offered by the 12-core protocol was statistically significant compared with the detection rate offered by the use of only six sextant biopsies ( p = 0.004). The diagnostic yield offered by the 21-core protocol was statistically significant compared with the detection rate offered by the use of the 12-core protocol (six sextant biopsies and six far lateral cores; p = 0.004). Adding three midline zone cores to the 12-core protocol increased the detection rate by 2.9% (41.6%). Adding six TZ cores to the 12-core protocol increased the detection rate by 4.7% (42.4%). PCa was significantly associated with an abnormal DRE ( p < 0.001; odds ratio: 2.7), higher PSA level (19.5 vs 7.2 ng/ml; p < 0.001), higher PSAD (0.47 vs 0.17; p < 0.001), lower %fPSA (0.14 vs 0.18; p < 0.001), and lower prostate volume (41.6 ml vs 49.9 ml; p < 0.001). All these factors remained significant in a multivariable model. Characteristics of PCa diagnosed by each biopsy protocol are listed in Table 3. Increasing the biopsy-core number resulted in diagnosing PCa with more favorable prebiopsy
Table 3 – Clinicopathologic features of prostate cancers detected by each six-core region (sextant, far lateral, and transition zone)
Cases, no. Mean PSA, ng/ml PSA >10, % Mean free-to-total PSA, % Mean PSAD, ng/ml per gram PSAD >0.20, % Mean age, yr Clinical stage >T1c, % Prostate volume, ml >50 ml, % Positive cores, no. Mean percentage of core involvement Biopsy Gleason score >6, % Insignificant PCa, % PCa eligible for AS, %: PSA <10 ng/ml; GS 6, T1c, <3 PC PSA <10 ng/ml, GS 6, PSAD <0.20 ng/ml per gram, T1c–T2a, <3 PC
PCa detected by a 6-core protocol
PCa detected by a 12-core protocol
PCa detected by a 21-core protocol
896 23.4 36.2 13.9 0.558 57.6 66.4 18.5 40.9 25.1 7.7 33.3 52.8 4.2
1111 20.4 33.4 14.1 0.489 54.0 66.1 17.3 41.4 26.5 6.7 31.1 46.4 7.2
1191 19.6 32.5 14.1 0.469 52.1 66.1 17.0 41.6 26.9 6.3 30.5 44.4 8.3
18.2 13.7
21.2 16.6
11.0 8.6
PCa = prostate cancer; PSA = prostate-specific antigen; PSAD = prostate-specific antigen density; AS = active surveillance; GS = Gleason score; PC = positive cores.
157
EUROPEAN UROLOGY 65 (2014) 154–161
Table 4 – Comparisons of clinicopathologic features of prostate cancers detected by a six-core protocol with those detected only by the six additional far lateral cores (12-core) and only by the last nine additional cores (midline and transition zone) Diagnosis by 6-core protocol: PCa cases (n = 896) (1)
Diagnostic yield by 12-core (vs 6-core): PCa cases (n = 215) (2)
Mean percentage core involvement (SD) Median Biopsy Gleason score >6, %
33.3 (20.5) 30.0 52.8
21.9 (19.2) 15.0 19.4
Dominant grade 4 pattern, %
23.6
Mean positive cores, no. (SD) Median Insignificant PCa, % PCa eligible for AS, %: PSA <10 ng/ml; GS 6, T1c, <3 PC PSA <10 ng/ml, GS 6, PSAD <0.20 ng/ml per gram, T1c–T2a, <3 PC
7.7 (5.4) 6.0 4.2
5.4
Diagnostic yield by 21-core (vs 12-core): PCa cases (n = 80) (3) 22.3 (20.5) 15.0 16.7 8.0
2.17 (1.53) 2.0 20.9
1.44 (0.71) 1.0 25.3
11.0
48.4
62.5
8.6
34.7
53.6
p value
(1) (2) (1) (2) (1) (2) (1) (2) (1) (2)
vs vs vs vs vs vs vs vs vs vs
(2): (3): (2): (3): (2): (3): (2): (3): (2): (3):
<0.001 0.874 <0.001 0.592 <0.001 0.425 <0.001 <0.001 <0.001 0.503
(1) vs (2): <0.001 (2) vs (3): 0.036 (1) vs (2): <0.001 (2) vs (3): 0.006
PCa = prostate cancer; PSA = prostate-specific antigen; PSAD = prostate-specific antigen density; AS = active surveillance; GS = Gleason score; PC = positive cores.
parameters and biopsy pathologic features. The percentage of low-risk PCas eligible for active surveillance increased twofold from the six-core to the 21-core biopsy scheme (Table 4). PCas detected by the six-core protocol were significantly more aggressive than those detected only by the 12- or 21-core protocol. Significant differences were reported in terms of percentage of core involvement, rate of high biopsy Gleason score, number of positive cores, rate of insignificant PCas, and percentage of men eligible for active surveillance ( p < 0.001 for all). The biopsy Gleason score and the percentage of core involvement were
comparable between PCa detected by the 12-core protocol diagnostic yield (n = 215) and those detected by the 21-core protocol (n = 80). Nevertheless, PCas detected only by the 21-core protocol involved significantly fewer biopsy cores (1.44 vs 2.17; p < 0.001) and were more likely to be eligible for active surveillance (62.5% vs 48.4%; p = 0.036) than those detected by the 12-core scheme. Nevertheless, no difference in insignificant PCa was found between the two protocols. To assess the ideal location of biopsy cores, we studied the characteristics of PCa missed by each six-core location: six sextant, six far lateral, and six TZ cores (Table 5). Using
Table 5 – Clinicopathologic features of prostate cancer missed by each six-core region (sextant, far lateral, and transition zone) and their comparisons*
Cases, no. (%) Mean PSA level, ng/ml PSA >10, % Mean free-to-total PSA, % Mean PSAD, ng/ml per gram PSAD >0.20, % Mean age, yr Clinical stage >T1c, mean no. Mean prostate volume, ml >50 ml, % Mean positive cores, no. Mean percentage of core involvement Biopsy Gleason score >6, % Insignificant PCa, % PCa eligible for AS, %: PSA <10 ng/ml; GS 6, T1c, <3 PC PSA <10 ng/ml, GS 6, PSAD <0.20 ng/ml per gram, T1c–T2a, <3 PC
PCa missed by six sextant biopsies
PCa missed by six far lateral biopsies
PCa missed by six TZ biopsies
311 (25.8) 8.2 22.5 14.9 0.23 36.9 65.1 13.2 43.6 31.6 2.0 22.0 20.1 21.2
254 (21.0) 10.0 23.4 16.5 0.28 31.1 66.3 11.4 44.8 30.6 2.0 22.2 19.8 21.3
399 (33.1) 9.5 22.8 16.1 0.26 38.7 65.7 14.3 44.5 32.0 2.3 23.7 21.6 20.2
50.8 39.6
50.6 44.2
47.2 36.9
PCa = prostate cancer; TZ = transition zone; PSA = prostate-specific antigen; PSAD = prostate-specific antigen density; AS = active surveillance; GS = Gleason score; PC = positive cores. * p > 0.1 for all statistical comparisons.
158
EUROPEAN UROLOGY 65 (2014) 154–161
Table 6 – Detection rates of the 6-, 12-, and 21-core biopsy schemes stratified by prebiopsy parameters (ie, prostate volume, prostate-specific antigen [PSA], PSA density [PSAD], and clinical stage), and their improvement of diagnostic yield
Prostate volume, % <50 ml >50 ml >70 ml PSA, ng/ml, % >10 ng/ml <10 ng/ml <4 ng/ml PSAD, ng/ml per gram, % <0.15 <0.20 >0.20 Clinical stage, % T1c T2–4 Age, yr, % <65 >65 Combined factors**, % Yes No * **
Detection rate by 6-core scheme
Detection rate by 12-core scheme (plus diagnostic yield, %)
Detection rate by 21-core scheme (plus diagnostic yield, %) *
36.9 22.3 16.8
45.0 (+18.1) 29.3 (+23.8) 21.6 (+22.2)
48.0 (+6.2) 31.9 (+8.4) 24.4 (+11.3)
51.6 27.0 14.0
59.0 (+12.5) 34.9 (+22.7) 20.9 (+32.8)
61.6 (+4.2) 37.9 (+7.9) 24.7 (+15.3)
17.6 20.9 50.4
24.5 (+27.8) 28.3 (+25.9) 58.8 (+14.2)
27.3 (+10.4) 31.7 (+10.8) 61.1 (+3.8)
29.9 53.0
37.7 (+20.5) 61.3 (+13.5)
40.5 (+7.1) 64.5 (+5.0)
26.3 40.1
33.4 (+21.1) 48.8 (+17.8)
35.8 (+6.8) 52.3 (+6.7)
14.4 37.1
20.1 (+28.6) 45.5 (+18.3)
22.8 (+11.8) 48.6 (+6.4)
p < 0.001 (McNemar test) for all comparisons of PCa detection rates between 12- and 21-core protocols. Combined factors: PSAD <0.20 plus PSA <10 plus T1c plus prostate volume >50 ml.
six sextant cores only, six far lateral cores only, or six TZ cores only would have missed 25.8%, 21.0%, or 33.1% of PCa, respectively. Thus the six far lateral cores were the most efficient cores in terms of detection rate. There was a trend toward less aggressive PCa among PCas that would have been missed by a six-core protocol than those missed by the
six far lateral cores or the TZ cores; however, differences were not significant. Table 6 shows the detection of each biopsy protocol and the diagnostic yield of the 12- and 21-core protocols in terms of various prebiopsy parameters (ie, PSA, PSAD, volume, and clinical stage). The overall detection rate was
Table 7 – Prebiopsy predictive factors of the diagnostic yield offered by a 21-core protocol as compared with a 12-core protocol by univariable and multivariable analyses Diagnostic yield 21- vs 12-core (n = 80), %
PSA level >10 ng/ml PSAD >0.20 ng/ml per gram Clinical stage >T1c Volume >50 ml Age >65 yr Multivariate analysis Categorical variables PSA level >10 ng/ml PSAD >0.20 ng/ml per gram Clinical stage >T1c Volume >50 ml Age >65 yr Simplified model PSAD >0.20 ng/ml per gram Clinical stage >T1c Age >65 yr Continuous variables PSA level PSAD Clinical stage >T1c Volume Age
No
Yes
33.4 54.0 17.3 26.5 55.3
20.0 27.5 12.5 31.9 55.0
p value
OR (95% CI)
0.013 <0.001 0.354 0.327 0.969
0.499 0.323 0.684 1.298 0.991
(0.28–0.88) (0.18–0.56) (0.35–1.35) (0.77–2.20) (0.63–1.56)
0.730 0.001 0.502 0.756 0.983
1.139 0.309 0.766 0.911 1.006
(0.54–2.38) (0.16–0.62) (0.35–1.67) (0.51–1.64) (0.61–1.66)
<0.001 0.512 0.970 0.393 0.007 0.699 0.776 0.889
OR = odds ratio; CI = confidence interval; PSA = prostate-specific antigen; PSAD = prostate-specific antigen density.
0.335 (0.19–0.58) 0.771 (0.36–1.68) 1.010 (0.62–1.66)
EUROPEAN UROLOGY 65 (2014) 154–161
higher in men with a low prostate volume, high PSA level and PSAD, and an abnormal DRE, regardless of number of biopsy cores. The diagnostic yield of the additional biopsy cores (12 or 21 cores) increased markedly when the prostate volume increased and when the PSA level and PSAD decreased. The diagnostic yield of the 21-core protocol was >10% in prostate >70 ml, in men with a PSA <4 ng/ml, and with a PSAD <0.20 ng/ml per gram. This yield was statistically significant ( p < 0.001). Predictive factors of the diagnostic yield offered by the 21-core protocol compared with the 12-core scheme are presented in Table 7. In univariable analysis, PSA level and PSAD were significant predictors of such a diagnostic yield. In multivariable analysis taking into account PSA, PSAD, clinical stage, volume, and age, the only independent predictive factor of the diagnostic yield offered by 21 biopsy cores was a PSAD <0.20 ng/ml per gram ( p < 0.001). This independent value remained significant regardless of prostate-volume cut-off (ie, 50, 60, or 70 ml) and of PSA levels (ie, 4 or 10 ng/ml) used in the multivariable analysis. 4.
Discussion
To our knowledge, our cohort was the largest of any prospective study published for this indication [7]. Overall, we found a continuum of improvement of cancer detection in the current study with an increasing number of cores. These findings confirmed our previous results and those published in the most recent prospective study [3,7]. The greater the number of biopsy cores taken, the larger the sampling of the prostate, and thus the lower the risk of missing a cancer. Nevertheless, accuracy of biopsy schemes depends on different parameters. The primary studied parameter is prostate volume. Sampling accuracy tends to decrease progressively with an increasing prostate volume [3,6,15]. In our study, as compared with the diagnostic yield offered by 12-core protocol, which stopped increasing at a prostate volume cut-off of 50 ml, the diagnostic yield of the 21-core protocol kept increasing with the prostate volume. Other parameters such as PSA and DRE influenced the detection rate and the yield provided by extended biopsy schemes. These variations in diagnostic yield offered by saturation strategies were highlighted by Scattoni et al. [3]. They showed that the most advantageous biopsy scheme, in terms of number and location of cores, varied as a function of age, prostate volume, and DRE. Our results confirmed these data, as shown in Table 6. The diagnostic yield of saturation was highly modified by the presumed risk of PCa that could be evaluated by DRE, PSA level, and prostate volume. For example, the diagnostic yield of the 12-core protocol ranged from 12.5% to 33% according to the PSA level. The diagnostic yield of the 21-core scheme was increased more than threefold in patients with PSA level <4 ng/ml compared with those having a PSA level >10 ng/ml (4.2% vs 15.2%). As compared with Scattoni et al, we chose to compare only 12 versus 21 cores. Thus our flowchart was simple, using only PSAD as the decisionmaking variable. Using multivariable logistic regression, we
159
identified a PSAD cut-off of 0.20 ng/ml per gram, below which an extended 21-core scheme might be systematically proposed to significantly improve overall detection. We think that this dichotomy is easier to use for the urologist in clinical practice. PSAD was previously identified as an interesting predictive factor when deciding on the number of prostate biopsy cores required [16]. Our findings also highlighted the importance of both number and location of biopsy cores. First, we suggested that the number of biopsy cores was a more important factor of high-risk PCa detection improvement than their exact location. No significant differences in terms of highgrade cancers were seen between PCa missed by the sextant biopsies, by the six far lateral biopsy specimens, or by the TZ specimens. Thus the rate of high-risk PCa missed by a six-core scheme did not differ according to the location of these six cores. Second, in line with previously published data, we found that the far lateral peripheral cores were the most efficient for diagnosis and added the most to the cancer detection rate compared with sextant and TZ biopsies [2,3,17,18]. Only 21% of PCa would have been missed by the use of only six lateral cores compared with 26% and 33% if only sextant or TZ biopsies, respectively, were used. Thus the location of sampled tissue appeared highly relevant when considering the overall detection rate including low- and high-risk PCas. The location of biopsy cores was not as important for the detection of high-grade PCa when compared with the biopsy-core number. Nevertheless, sampling the most lateral periphery appeared highly relevant to detecting as many cancers as possible. In our previous work, the diagnostic yield offered by a 18-core protocol (12-core plus TZ) was compared with that offered by a 21-core protocol (12-core plus TZ plus midline). The conclusion was that the diagnostic yield offered by midline zone biopsy specimens was not significant compared with the 18-core procedure. Unfortunately, the diagnostic yield offered by a 15-core protocol (12-core plus midline) has not been studied. In the present study, the diagnostic yield offered by midline cores was moderate (3%) but significant as compared with the 12-core protocol. This yield was still lower than that offered by the six TZ biopsies, but we cannot conclude if this discrepancy might be explained by the different core location or just by a lower number of additional cores (ie, three vs six). Even if the detection rate increased with an increasing biopsy-core number, it is important to know if such an improvement in detection is clinically relevant. The counterpart is that the saturation biopsy strategies have been hypothesized to increase the potential risk of overtreating patients whose tumors pose very low mortality risk. However, our 21-core protocol did not statistically increase the rate of detected insignificant PCa as defined by the Epstein criteria [13,14]. Moreover, the role of prostate biopsies has changed. The actual importance of prostate biopsies has evolved from purely cancer detection to investigating how biopsy results can assist clinical management for patients. The inclusion of patients in active surveillance protocols emphasizes the necessity of accurate staging strategies. A larger amount of sampled tissue may
160
EUROPEAN UROLOGY 65 (2014) 154–161
achieve a more complete picture of the disease burden. Extended or saturation biopsies have been advocated to detect cancers that standard biopsies miss and also to characterize better PCa volume and prognosis [19–22]. This impact is of great interest in low-risk PCa patients eligible for active surveillance because the need for subsequent reassessment biopsies may be reduced, as might the risk of initial misclassification of the disease [23]. In our study, PCas diagnosed by this 21-core scheme had the same pathologic characteristics as those diagnosed by the 12-core protocol in terms of percentage of cancer involvement per core and biopsy Gleason score, but they were more likely to be eligible for active surveillance due to a lower mean number of positive cores. We previously demonstrated that about 30% of cancers fulfilling the active surveillance criteria in a 12-core protocol, but not in the 21-core scheme, exhibited unfavorable disease characteristics in the radical prostatectomy (RP) specimen [21]. Our previous findings suggested that the 21-core biopsy strategy provided a more accurate staging of PCa and that a more extended biopsy strategy provides a more accurate assessment. The 21-core protocol did not statistically increase the rate of detected insignificant PCa as defined by the updated Epstein criteria. To date, these criteria are the most widely used preoperative criteria, and more recent attempts to predict insignificant PCa after positive biopsies and before surgery are based on this updated definition [13,14]. One limitation of our study was that we could not confirm the clinical significance of cancer because all patients did not undergo RP. This point might be relevant because the Epstein criteria are not perfect and misclassify about 30% of patients who would have unfavorable pathologic features in an RP specimen [14]. Nevertheless, in a previous study analyzing the RP specimens of patients selected from the same cohort, we found that PCa diagnosed only in a 21-core protocol were at least as aggressive as PCa detected by the diagnostic yield offered by a 12-core scheme [24]. The safety of the 21-core protocol was previously studied in our first report [9]. The overall complication rate requiring hospitalization was 3%, including infection (1%), acute urinary retention (1.9%), and rectal bleeding. Macroscopic hematuria occurred in 84% of patients and lasted an average of 3 d. Minor rectal bleeding, lasting an average of 1 d, was reported by 45% of patients. Thus the morbidity rate of the 21-core procedure was close to that reported with the standard 12-core procedure.
have the same pathologic characteristics as those detected by the diagnostic yield offered by the 12-core protocol in terms of percentage of cancer involvement per core and biopsy Gleason score, but they were more likely to be eligible for active surveillance. Nevertheless, the 21-core protocol does not statistically increase the rate of detected insignificant PCa as defined by the Epstein criteria. Author contributions: Guillaume Ploussard 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: Ploussard, Abbou, Salomon, de la Taille. Acquisition of data: Ploussard, Marchand, Nicolaiew, Salomon, de la Taille. Analysis and interpretation of data: Ploussard, Nicolaiew, Terry, Vacherot, Vordos, Allory. Drafting of the manuscript: Ploussard. Critical revision of the manuscript for important intellectual content: Vacherot, Abbou, Salomon, de la Taille. Statistical analysis: Ploussard, Nicolaiew. Obtaining funding: None. Administrative, technical, or material support: None. Supervision: de la Taille. Other (specify): None. Financial disclosures: Guillaume Ploussard certifies 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.
References [1] Hodge KK, McNeal JE, Stamey TA. Ultrasound guided transrectal core biopsies of the palpably abnormal prostate. J Urol 1989;142:66–70. [2] Epstein JI, Walsh PC, Carter HB. Importance of posterolateral needle biopsies in the detection of prostate cancer. Urology 2001;57: 1112–6. [3] Scattoni V, Raber M, Abdollah F, et al. Biopsy schemes with the fewest cores for detecting 95% of the prostate cancers detected by a 24-core biopsy. Eur Urol 2010;57:1–8. [4] Singh H, Canto EI, Shariat SF, et al. Improved detection of clinically significant, curable prostate cancer with systematic 12-core biopsy. J Urol 2004;171:1089–92. [5] Scattoni V, Zlotta A, Montironi R, Schulman C, Rigatti P, Montorsi F. Extended and saturation prostatic biopsy in the diagnosis and characterisation of prostate cancer: a critical analysis of the literature. Eur Urol 2007;52:1309–22.
5.
Conclusions
[6] Chun FK-H, Epstein JI, Ficarra V, et al. Optimizing performance and interpretation of prostate biopsy: a critical analysis of the literature.
A 21-core biopsy scheme improves significantly the PCa detection rate compared with a 12-core protocol. This improvement of diagnostic yield is >10% in patients having a PSAD <0.20 ng/ml per gram, a prostate volume >70 ml, and/or a PSA level <4 ng/ml. We identified a PSAD <0.20 ng/ml per gram as the most predictive factor of the diagnostic yield offered by this 21-core protocol. The biopsy-core number might be selected based on this PSAD cut-off value. PCas diagnosed using this 21-core scheme
Eur Urol 2010;58:851–64. [7] Guichard G, Larre´ S, Gallina A, et al. Extended 21-sample needle biopsy protocol for diagnosis of prostate cancer in 1000 consecutive patients. Eur Urol 2007;52:430–5. [8] Meng MV, Elkin EP, DuChane J, Carroll PR. Impact of increased number of biopsies on the nature of prostate cancer identified. J Urol 2006;176:63–8. [9] De la Taille A, Antiphon P, Salomon L, et al. Prospective evaluation of a 21-sample needle biopsy procedure designed to improve the prostate cancer detection rate. Urology 2003;61:1181–6.
EUROPEAN UROLOGY 65 (2014) 154–161
161
[10] Jones JS, Patel A, Schoenfield L, Rabets JC, Zippe CD, Magi-Galluzzi C.
[18] Philip J, Hanchanale V, Foster CS, Javle P. Importance of
Saturation technique does not improve cancer detection as an
peripheral biopsies in maximising the detection of early prostate
initial prostate biopsy strategy. J Urol 2006;175:485–8.
cancer in repeat 12-core biopsy protocols. BJU Int 2006;98:
[11] Pepe P, Aragona F. Saturation prostate needle biopsy and prostate cancer detection at initial and repeat evaluation. Urology 2007;70: 1131–5. [12] Eichler K, Hempel S, Wilby J, Myers L, Bachmann LM, Kleijnen J. Diagnostic value of systematic biopsy methods in the investigation
559–62. [19] Numao N, Kawakami S, Yokoyama M, et al. Improved accuracy in predicting the presence of Gleason pattern 4/5 prostate cancer by three-dimensional 26-core systematic biopsy. Eur Urol 2007;52: 1663–9.
of prostate cancer: a systematic review. J Urol 2006;175:1605–12.
[20] Descazeaud A, Rubin M, Chemama S, et al. Saturation biopsy
[13] Bastian PJ, Mangold LA, Epstein JI, Partin AW. Characteristics of
protocol enhances prediction of pT3 and surgical margin status
insignificant clinical T1c prostate tumors. A contemporary analysis. Cancer 2004;101:2001–5. [14] Ploussard G, Epstein JI, Montironi R, et al. The contemporary concept of significant versus insignificant prostate cancer. Eur Urol 2011;60:291–303. [15] Brawer MK. The influence of prostate volume on prostate cancer detection. Eur Urol Suppl 2002;1:35–9. [16] Jeong H, Jeong BC, Kwak C, Lee E, Lee SE, Kim TB. A comparison of
on prostatectomy specimen. World J Urol 2006;24:676–80. [21] Ploussard G, Xylinas E, Salomon L, et al. The role of biopsy core number in selecting prostate cancer patients for active surveillance. Eur Urol 2009;56:891–8. [22] Capitanio U, Karakiewicz PI, Valiquette L, et al. Biopsy core number represents one of foremost predictors of clinically significant Gleason sum upgrading in patients with low-risk prostate cancer. Urology 2009;73:1087–91.
prostate cancer detection rates by 12 or 6 core biopsy at different
[23] Abouassaly R, Lane BR, Jones JS. Staging saturation biopsy in
prostate specific antigen densities in Korean men. World J Urol
patients with prostate cancer on active surveillance protocol. Urol-
2008;26:395–400.
ogy 2008;71:573–7.
[17] Presti JC, O’Dowd GJ, Miller MC, et al. Extended peripheral zone
[24] Ouzaid I, Xylinas E, Campeggi A, et al. Contemporary pathologic
biopsy schemes increase cancer detection rates and minimizes vari-
characteristics and oncologic outcomes of prostate cancers missed
ance in prostate specific antigen and age related cancer rates: results
by 6- and 12-core biopsy and diagnosed with a 21-core biopsy
of a community multipractice study. J Urol 2003;169:125–9.
protocol. World J Urol 2013;31:869–74.