SEARCH

SEARCH BY CITATION

Keywords:

  • biopsy;
  • prostatectomy;
  • prostatic neoplasm

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

Study Type – Diagnostic (exploratory cohort)

Level of Evidence 3a

What's known on the subject? and What does the study add?

Initial transrectal 12-core biopsy has a small but definite risk of missing anterior significant prostate cancers irrespective of age, PSA, prostate volume and DRE findings.

Our study yields valuable information for diagnosis and treatment decision of prostate cancer based on transrectal 12-core biopsy.

OBJECTIVE

  • • 
    To characterize prostate cancers missed by initial transrectal 12-core biopsy.

PATIENTS AND METHODS

  • • 
    Between 2002 and 2008, 715 men with prostate-specific antigen levels in the range 2.5–20 ng/mL or abnormal digital rectal examination underwent three-dimensional 26-core prostate biopsy (i.e. a combination of transrectal 12-core biopsy and transperineal 14-core biopsy) on initial examination.
  • • 
    Of the 257 patients diagnosed with cancer, 120 patients subsequently underwent radical prostatectomy.
  • • 
    Cancers were grouped into TR12-negative cancers (i.e. not detected through transrectal 12-core biopsy but detected through transperineal 14-core biopsy) and TR12-positive (i.e. detected through transrectal 12-core biopsy) cancers.
  • • 
    Clinicopathological characteristics of the TR12-negative and TR12-positive cancers were evaluated.

RESULTS

  • • 
    TR12-negative cancers comprised 21% of the three-dimensional 26-core biopsy-detected cancers.
  • • 
    The frequency of cancers with a biopsy Gleason score ≤6 and that of cancers with a biopsy primary Gleason grade ≤3 was higher in TR12-negative cancers, at 58% and 83%, respectively, than in TR12-positive cancers, at 25% (P < 0.001) and 53% (P < 0.001), respectively.
  • • 
    The median number of positive cores in TR12-negative cancers was two out of 26.
  • • 
    TR12-negative cancers were more frequently located anteriorly than posteriorly.
  • • 
    The incidence of the TR12-negative cancers was not associated significantly with any clinical variable.

CONCLUSION

  • • 
    Many of the cancers missed by initial transrectal 12-core biopsy are probably low-grade and low-volume diseases, although initial transrectal 12-core biopsy has a small but definite risk of missing anterior significant cancers.

Abbreviations
3D26PBx

three-dimensional 26-core prostate biopsy

GS

Gleason score

RP

radical prostatectomy

TR12PBx

transrectal 12-core prostate biopsy

TP14PBx

transperineal 14-core prostate biopsy

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

In a pattern consistent with the worldwide trend toward choosing extended over non-extended prostate biopsy methods, transrectal 12-core prostate biopsy (TR12PBx) is currently one of the most preferred biopsy methods for detecting prostate cancers. A systematic review of prostate biopsy methods noted that TR12PBx strikes a satisfactory balance with sufficiently high rates of cancer detection and sufficiently low rates of biopsy-associated comorbidity, and that taking more than 12 cores adds no significant benefit [1]. TR12PBx also meets the criteria for initial biopsy provided by the representative clinical guidelines [2,3]. Yet several studies have reported that repeat biopsy after negative initial extended transrectal biopsy detects prostate cancer in 17–21% of men [4–6], suggesting that these initial extended transrectal biopsies may miss a substantial number of cancers.

To clarify the incidence and clinical importance of cancers missed by TR12PBx, it is necessary to analyze the results obtained using biopsy protocols that include not only all of the TR12PBx sampling sites, but also additional sampling sites. To the best of our knowledge, there are currently three biopsy protocols that meet these requirements. The first is three-dimensional 26-core prostate biopsy (3D26PBx), a combination of transperineal 14-core prostate biopsy (TP14PBx) and TR12PBx (Fig. 1), introduced by our group [7–9]. In a previous analysis of 321 men examined through 3D26PBx, we reported that 3D26PBx increased cancer detection by 24% compared to TR12PBx [7]. The second is transrectal 21-core biopsy [10]. The transrectal 21-core biopsy can detect significantly more cancers (increased detection of 9.8%) than the TR12PBx. The third is transrectal 14-core biopsy (TR12PBx plus two extreme anterior apical biopsy sites) [11]. The addition of only two extreme anterior apical sampling sites to TR12PBx increased the cancer detection rate by 7.5%. Although these studies mainly focused on cancer detectability, characteristics of cancers missed by TR12PBx have not been fully assessed to date.

image

Figure 1. Transverse, sagittal and coronal projections of three-dimensional 26-core prostate biopsy (3D26PBx), a combination of transperineal 14-core prostate biopsy (TP14PBx) and transrectal 12-core prostate biopsy (TR12PBx). The sampling sites are named: anterior 1 (A1), anterior 2 (A2), posterior 1 (P1), posterior 2 (P2), anterolateral (AL), posterolateral (PL) and transition zone (TZ) in TP14PBx; parasagittal apex (pa), parasagittal midprostate (pm), parasagittal base (pb), lateral apex (la), lateral midprostate (lm) and lateral base (lb) in TR12PBx.

Download figure to PowerPoint

When a patient undergoes an initial TR12PBx and the result is negative for cancer, how much risk does he have for a clinically important cancer that is missed? What are characteristics of the cancers missed by TR12PBx? To address these questions, we evaluated the characteristics of cancers that were detected or missed by TR12PBx in a cohort of 715 men undergoing 3D26PBx.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

PATIENTS

Between June 2002 and June 2008, 757 men prospectively underwent 3D26PBx as an initial biopsy at our institutions because of higher PSA levels >2.5 ng/mL and/or abnormal DRE findings in a clinical setting. Patients were excluded if they had diabetes mellitus or any rectal disease (e.g. uncontrolled hemorrhoids) because of the high risk of infection or rectal bleeding. In principle, those with apparently palpable mass, age ≥75 years, PSA level ≥20 ng/mL or poor state of health were excluded from recommendation for 3D26PBx. Written informed consent was obtained from all patients and 3D26PBx was performed under spinal, general or, recently, local anaesthesia [12], as described previously [7–9]. Of these 757 patients, 42 were excluded from the current study because of palpable stage T3/4 tumours, PSA level ≥20 ng/mL or lack of baseline clinical data. A total of 715 patients were subjected for analyses.

PATHOLOGICAL EVALUATION

All biopsy and radical prostatectomy (RP) specimens were re-evaluated by a single pathologist according to the 2005 International Society of Urologic Pathology Consensus Conference on Gleason Grading [3,13,14]. Each biopsy core was individually labelled so that the location of cancer-positive cores could be analyzed. All RP specimens were processed as described previously [9]. Tumour volume, Gleason score (GS), pathological stage and location of each isolated cancer focus in the RP specimens were recorded. Significant cancer was defined as a tumour volume ≥0.5 mL and/or Gleason pattern 4/5 and/or extraprostatic extension. A significant cancer focus was defined as one fulfilling the above-mentioned criteria for significant cancer, and was extensively evaluated. For analysis of cancer location, the prostate was divided into anterior, posterior and apical regions. The apical region was defined as the most inferior 10 mm of the gland. The remaining part of the gland was divided into anterior and posterior regions at the height of the urethra [15]. When a significant focus lay astride two regions, it was assigned to both regions.

DATA ANALYSIS

All cancers were grouped into two mutually exclusive groups: TR12-negative (i.e. not detected through transrectal 12-core biopsy but detected through transperineal 14-core biopsy) and TR12-positive (i.e. detected through transrectal 12-core biopsy) cancers. The former group did not have cancer-positive cores within the TR12PBx scheme but had cancer-positive cores within the TP14PBx scheme, and the latter had cancer-positive cores within the TR12PBx scheme. These two groups were compared with regard to patient age, PSA level, prostate volume, DRE findings, biopsy GS and the number of positive cores. In patients treated with RP, the two groups were also compared with regard to RP GS, pathological stage, tumour volume, frequency of significant cancer and cancer location. The study cohort was categorized by age, PSA level, prostate volume and DRE findings to identify any patient subgroups in which TR12PBx did not exhibit sufficient cancer detection rates.

STATISTICAL ANALYSIS

All analyses were performed using JMP, version 7 (SAS Institute Inc., Cary, NC, USA). Continuous variables were analyzed using Mann–Whitney's U-test. Categorical variables were analyzed using the chi-squared test or Fisher's exact test. The Cochran–Armitage test was used to test for trends. P < 0.05 was considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

CHARACTERISTICS OF TR12-NEGATIVE CANCERS: ANALYSES ON BIOPSY SPECIMENS

Prostate cancers were detected through 3D26PBx in 257 (35.9%) of the 715 men. Of these 257 cancers, 53 (21%) were identified as TR12-negative cancers; in other words, the addition of the TP14PBx sites to the TR12PBx sites improved the cancer detection rate by 26%. Patient and tumour characteristics of TR12-positive and TR12-negative cancers are shown in Table 1. Compared to patients with TR12-positive cancers, patients with TR12-negative cancers had significantly larger prostates and a lower incidence of abnormal DRE. The frequency of cancers with biopsy GS ≤6 and that of cancers with biopsy primary Gleason grade ≤3 were higher in TR12-negative cancers, at 58% and 83%, respectively, than in TR12-positive cancers, at 25% (P < 0.001) and 53% (P < 0.001), respectively.

Table 1.  Patient and tumour characteristics
VariableAll patients (n= 715)Patients with cancer (n= 257)TR12-positve cancer (n= 204)TR12-negative cancer (n= 53)P
  1. Continuous variables are expressed as the median (interquartile range). GS, Gleason score; TR12-positive, transrectal 12-core biopsy-positive; TR12-negative, transrectal 12-core biopsy-negative

Age (years)66 (61–71)66 (63–72)68 (63–72)67 (63–73)0.914
PSA level (ng/mL)6.1 (4.7–8.5)7.0 (5.1–9.6)7.2 (5.1–9.8)6.8 (5.2–8.7)0.370
Prostate volume (mL)35 (27–47)29 (23–39)29 (22–37)32 (27–48)0.003
% Abnormal DRE162529110.009
Number of positive cores, n (range)5 (2–8)6 (3–9)2 (1–2)<0.001
Biopsy GS, n (%)     
 5–683 (32)52 (25)31 (58)<0.001
 3 + 469 (27)56 (27)13 (25) 
 4 + 344 (17)39 (19)5 (9) 
 8 – 1061 (24)57 (28)4 (8) 

In the 53 TR12-negative cancers, cancer-positive rates within the TP14PBx sampling sites are shown in Table 2. The farthest anterior sampling site (A1; Fig. 1) had the highest cancer-positive rate of 47%. There were six cores from the anterior sampling sites (A1, A2 and AL; Fig. 1) that detected significantly (P= 0.009) more cancers than six cores from the posterior sites (P1, P2 and PL; Fig. 1), indicating that TR12-negative cancers are located more frequently in the anterior region than in the posterior region.

Table 2.  TR12-negative cancer (n= 53) detection rates in each transperineal sampling site
Transperineal sampling site*% Cancer detectionP
  • *

    See Fig. 1.

  • According to a chi-squared test (A1 + A2 vs P1 + P2, A1 + A2 + AL vs P1 + P2 + PL). TR12-negative, transrectal 12-core biopsy-negative.

A147 
A228 
AL28 
P124 
P219 
PL19 
TZ19 
A1 + A2620.020
P1 + P240 
A1 + A2 + AL750.009
P1 + P2 + PL51 

CHARACTERISTICS OF TR12-NEGATIVE CANCERS: ANALYSES ON RP SPECIMENS

In total, 120 of the 257 (47%) patients underwent RP [16]. Characteristics of TR12-positive and TR12-negative cancers are shown in Table 3. Among the pathological variables analyzed in the RP cohort, the TR12-positive and TR12-negative groups differed with respect to cancer location: specifically, TR12-negative cancers were located less frequently in the posterior region than the TR12-positive cancers.

Table 3.  Patient and tumour characteristics of 120 men undergoing radical protectomy
VariableTR12-positive cancer (n= 104/204)TR12-negative cancer (n= 16/53)P
  1. Continuous variables are expressed as the median (interquartile range). GS, Gleason score; TR12-positive, transrectal 12-core biopsy-positive; TR12-negative, transrectal 12-core biopsy-negative.

Clinical   
 Age (years)67 (62–71)64 (60–70)0.245
 PSA level (ng/mL)6.7 (5.2–9.3)6.4 (4.7–8.4)0.362
 Prostate volume (mL)28 (22–34)29 (20–38)0.817
 % Abnormal DRE24250.666
Biopsy   
 Number of positive cores, n (range)5 (3–9)2 (1–3)<0.001
 GS, n (%)   
  5–617 (16)8 (50)<0.001
  3 + 432 (31)5 (31) 
  4 + 327 (26)1 (6) 
  8–1028 (27)2 (13) 
Radical protectomy   
 GS, n (%)   
  5–626 (25)3 (19)<0.001
  3 + 449 (47)9 (56) 
  4 + 316 (15)3 (19) 
  8–1013 (13)1 (6) 
 % Organ-confined disease73870.178
 % Tumour volume ≥0.5 mL80750.437
 % Significant cancer92870.396
 Significant cancer   
  % Located posteriorly8350<0.001
  % Located anteriorly75710.500
  % Located apically86790.331

SUBGROUPS IN WHICH TR12PBX IS INSUFFICIENT FOR CANCER DETECTION

To identify patient subgroups in which TR12PBx would be entirely insufficient for cancer detection and in which more sampling would be needed, we compared the incidence of TR12-negative cancers in subgroups defined by age, PSA level, prostate volume or DRE findings (Fig. 2). Although cancer detection rates of 3D26PBx were significantly higher in patient subgroups with higher age, higher PSA level, smaller prostates or abnormal DRE findings, the incidence of TR12-negative cancers did not differ significantly between any of the subgroups.

image

Figure 2. Bar graphs show detection rates of cancers of all types through 3D26PBx. The overlying line plots show detection rates of TR12-negative cancers (i.e. not detected through transrectal 12-core biopsy but detected through transperineal 14-core biopsy). Statistical analyses were performed using the Cochran–Armitage trend test.

Download figure to PowerPoint

COMPARISON OF CANCER CHARACTERISTICS BETWEEN NORMAL AND ABNORMAL DRE GROUPS

PSA screening has significantly increased the proportion of men who undergo prostate biopsy based on PSA findings alone. To evaluate the efficacy of TR12PBx in men with normal DRE in more detail, the cancer characteristics of TR12-negative cancers were analyzed according to DRE findings (Table 4). TR12-positive cancers in the normal DRE group tended to have lower biopsy cancer grade and fewer positive cores compared to those in the abnormal DRE group. By contrast, in TR12-negative cancers, biopsy cancer grade and the number of positive cores did not differ significantly between the groups.

Table 4.  Comparison of patients and tumour characteristics between normal and abnormal DRE cohorts
VariableAll patientsPPatients with cancerPTR12-positive cancerPTR12-negative cancerP
DRENormal (n= 602)Abnormal (n= 113)Normal (n= 192)Abnormal (n= 65)Normal (n= 145)Abnormal (n= 59)Normal (n= 47)Abnormal (n= 6)
  1. Continuous variables are expressed as the median. GS, Gleason score; TR12-positive, transrectal 12-core biopsy-positive; TR12-negative, transrectal 12-core biopsy-negative.

Age (years)65670.01068670.22669670.26167650.508
PSA level (ng/mL)6.06.70.0226.97.40.2177.07.40.2706.87.30.715
Prostate volume (mL)36320.03230280.43529300.67134260.035
Number of positive cores 36<0.00157<0.001220.251
Biopsy GS ≤6 (%) 36220.03128190.15260500.488
Biopsy primary Gleason grade ≤3 (%) 66480.01060440.03783830.733

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

In the present study, we evaluated the characteristics of cancers missed by initial TR12PBx, more precisely, cancers missed by TR12PBx in patients who underwent 3D26PBx (TR12-negative cancers), and thus showed the diagnostic performance of TR12PBx. Initial TR12PBx missed 21% of cancers that were detectable through 3D26PBx; however, it should be noted that more than half of TR12-negative cancers had a biopsy GS ≤6, and most of them had a biopsy primary Gleason grade ≤3. Furthermore, the median number of positive cores in TR12-negative cancers was only two out of 26, suggesting that a substantial number of TR12-negative cancers can be expected to be low-grade and low-volume diseases.

Although our RP cohort is highly selective, most TR12-negative cancers treated with RP were significant cancers. Yet 87% were organ-confined disease and 75% were primary Gleason grade 3 cancers with favourable prognosis. Combined with the biopsy findings, this indicates that TR12-negative cancers have lower malignant potential than TR12-positive cancers, and that most of them can be expected to be organ confined, although a small number of TR12-negative cancers appear to be significant cancers that would exhibit biological aggressiveness.

The characteristics of the location of TR12-negative cancers are clearly shown in the present study. Our analysis of positive transperineal sites in TR12-negative cancers confirmed that TR12-negative cancers were located in the anterior portion of the gland rather than the posterior portion. The cancer location of TR12-negative cancers in RP specimens also supports this notion, suggesting that TR12PBx would be insufficient to detect anterior cancers. The results obtained in the present study are similar to those obtained by Moussa et al.[11], who reported that the addition of only two extreme anterior apical cores to TR12PBx transrectal sampling improved cancer detection by 7.5% and that these two cores achieved the highest rate of unique cancer detection. They therefore introduced the 14-core biopsy scheme (TR12PBx biopsy plus two extreme apical cores) as an initial biopsy to detect more anterior apical cancers.

On the basis of these findings, simply adding more transrectal sampling sites from the bottom of the prostate gland to TR12PBx would not increase its cancer detection rate efficiently. Indeed, several studies have tested transrectal extended biopsy methods using more than 12 cores; however, most of these so-called ‘saturation’ transrectal biopsy protocols did not outperform TR12PBx [17–19]. The transrectal 21-core biopsy [10] can identify more cancers than TR12PBx can, although the 9.8% increase in cancer detection that results from the nine additional samplings appears to be inefficient. In the present study, the addition of two far-anterior transperineal sampling to the TR12PBx improved its cancer detection rate by 11%. From these results, we consider that a few additional samplings in the anterior apical portion are effective for detecting cancers missed by TR12PBx.

In recent analyses on men with a PSA level <20 ng/mL without locally advanced tumours on DRE findings, there were no significant differences among subgroups defined by age, PSA level, prostate volume or DRE in the incidence of TR12-negative cancers. Therefore, we could not identify any subgroup in which TR12PBx would be entirely insufficient for cancer detection and in which more sampling would therefore be needed. There is the potential concern that TR12PBx may probably miss anterior aggressive cancers in men with normal DRE, although our analysis of cancer characteristics according to DRE findings shows that a large proportion of TR12-negative cancers probably consist of low-grade and low-volume disease, regardless of DRE findings. We consider that these results are sufficient grounds to eliminate such concerns.

Recently, the value of MRI for detecting prostate cancers and determining their location has been extensively studied [20]. Lawrentschuk et al.[21] retrospectively analyzed patients with anteriorly predominant tumours on MRI who had undergone prostate biopsy and reported that MRI would be useful in the detection of anterior tumours that are difficult to detect using transrectal biopsy. It appears that MRI is a promising tool for detecting anterior cancers, although its cost is high.

Pathological evaluation in the present study is based on the 2005 International Society of Urologic Pathology Consensus [13,14]. Fused glands, ill-defined glands with poorly formed glandular lumina and most of the cribriforms are previously categorized into Gleason pattern 3 but, in the 2005 consensus, are categorized into Gleason pattern 4. Furthermore, two new modifications to the Gleason scoring system are recommended in the evaluation of biopsy specimens. One is that any high-grade pattern, no matter how small quantitatively, should be incorporated into the GS, although any secondary grade that occupies <5% of the specimen would not have been reported under the previous system. The other modification is that all higher tertiary grade components of the tumour, which were previously ignored, should be incorporated into the GS. Accordingly, the rates of high-grade patterns scored according to the 2005 consensus are higher than those scored under the previous system. This phenomenon has been confirmed in a study by Billis et al.[22] showing that GS that had been scored under the previous system were upgraded by re-evaluation under the 2005 consensus in 26.7% of the biopsy specimens. Similarly, the GS of some of the patients in the present study would have been lower if they had been evaluated under the previous system. We consider that these findings strengthen our view that many TR12-negative cancers can be expected to be low-grade diseases.

The present study does not indicate that we should aim to actively detect TR12-negative cancers in all candidates for initial prostate biopsy because a substantial number of TR12-negative cancers are low-grade and low-volume diseases. Overdiagnosis and overtreatment are now issues of major concern in the management of prostate cancer. Draisma et al.[23] have reported that the rate of overdiagnosis of prostate cancer has been estimated at 23–66% of screening-detected cancers. The main purpose of prostate biopsy is not only to detect more prostate cancers, but also to detect more life-threatening cancers. Even if we missed a case of prostate cancer at an initial biopsy, we would be able to determine the need for a repeat biopsy through the PSA test in most cases, and most cancers detected by repeat biopsy are manageable and not life-threatening [24]. If the goal of screening were simply to detect life-threatening cancers, the addition of sampling sites to the TR12PBx protocol would not be essential in all candidates for initial biopsy.

Another important purpose of biopsy, however, is to accurately characterize any tumours to allow for a more informed treatment decision-making process. If a custom treatment is to be devised for each individual patient, more sampling is required to generate more information, although more sampling leads to a greater detection of indolent cancers. This clinical dilemma makes it difficult to determine the optimal biopsy scheme. We now consider that the addition of anterior sampling sites to the initial TR12PBx would be a reasonable option for younger men with a long life expectancy or for men with suspected anterior cancer as assessed by MRI. On the basis of the analysis in subgroups divided by DRE findings, the finding that a DRE was normal does not mean that additional sampling of the anterior prostate should be performed. At repeat biopsy after negative TR12PBx, however, anterior samplings are highly recommended.

The present study has several limitations that should be considered. Given that 3D26PBx does not identify all cancers, it is possible that TP12PBx may fail to detect an even greater percentage of cancers than reported in the present study. We recommend transperineal sampling for the detection of TR12-negative cancers, although we realize that the transperineal approach may be unfamiliar to many urologists. A recently reported technique for simple and effective local anaesthesia would render transperineal extended biopsy more feasible [12]. Furthermore, because of the limited duration of the follow-up in the present study, we could not report the oncological outcome of TR12-negative cancers. Long-term observation will be required to acquire a better understanding of the diagnostic performance of TR12PBx.

In conclusion, TR12PBx missed 21% of cancers that were detected by 3D26PBx on initial biopsy. Although many of the undetected cancers were expected to be low-grade and low-volume diseases, it should be noted that the initial TR12PBx has a small but definite risk of missing anterior significant cancers.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES