SEARCH

SEARCH BY CITATION

Keywords:

  • prostate cancer;
  • needle biopsy;
  • tumour staging;
  • diagnostic error

Abstract

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

OBJECTIVE

To describe the localization and to assess the clinical implications of areas of undetected prostate cancer in radical prostatectomy (RP) specimens, focusing on patients with unilaterally negative preoperative biopsy cores.

PATIENTS AND METHODS

The study included 149 of 559 consecutive patients (26.7%) who had RP for prostate cancer. Unilateral prostate cancer was diagnosed from prostate biopsies, taken by several physicians, but ≥ pT2c disease was present in the RP specimen. The prostate was dissected by standardized transversal cuts and tumour areas were mapped by one genitourinary pathologist. To estimate the tumour size and location, areas of prostate cancer were transferred to a digital grid database representing the prostate by 794 units.

RESULTS

The most frequent location of undetected prostate cancer was in the dorsalateral region and in the apex of the prostate. The mean tumour volume of the false-negative lobe was significantly lower than contralaterally (18.9 vs 47.5 units, P < 0.001). In 36 of 149 patients (24.2%), the tumour volume on the negative biopsy side was equal or higher than on the positive biopsy side; in the final RP specimen, 60 patients (40.3%) had capsular involvement on the negative biopsy side.

CONCLUSION

Significantly many patients with newly diagnosed prostate cancer remain clinically understaged. The apical and dorsolateral region of the prostate are not adequately represented in current biopsy strategies. Undetected tumour areas are often clinically significant by size and capsular involvement, indicating a direct clinical implication when planning nerve-sparing RP or focal therapy. Our results show a continuing need for optimized and standardized biopsy protocols.


Abbreviation
RP

radical prostatectomy.

INTRODUCTION

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

TRUS-guided prostate biopsy is the standard method for detecting prostate cancer [1]. The pathological biopsy report guides treatment decisions and should provide reliable information on tumour stage, grade and the extent of the disease. Random systematic sextant biopsy has had several extensions and modifications since its introduction [2–4]. It was shown that adding peripheral biopsies significantly increase prostate cancer detection rates [5–7].

There has been a stage migration in prostate cancer, with increasingly many small-volume tumours being diagnosed at presentation [8]. Besides radical prostatectomy (RP) and radiotherapy still having a considerable risk of morbidity, this trend has led to the idea of targeting focal prostate cancer lesions within the prostate, leaving parts of the gland untreated and thus minimizing side-effects. This novel concept of ‘focal therapy’ has recently been offered to increasingly many patients supposed to have insignificant, often unilateral prostate cancer [9,10]. However, no robust data on the long-term oncological outcomes after such therapy are yet available. Moreover, the diagnostic accuracy of prostate biopsy is challenged by this concept.

Prostate biopsy accounts for significant under-grading in up to a third of all patients [11] and there are further limits in accurately predicting the extent of prostate cancer [12,13]. Tumour areas that remain undetected might be of small volume and irrelevant, but at least for sextant biopsies it has been suggested that cancers missed by biopsy are clinically significant [14]. The visualization and characterization of tumours in false- negative lobes of the prostate might contribute to further optimizing biopsy strategies, by directing the needle to areas with a high probability of missed prostate cancer.

We reviewed patients who recently had RP, to determine the rate and extent of unilaterally false-negative prostate cancer after TRUS-guided prostate biopsy. The aim of the study was to describe the pathological characteristics and as the exact location and possible clinical impact of missed tumour areas in patients with bilateral disease on final pathology.

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

The charts of 559 consecutive patients who had RP with regional lymphadenectomy at our department between July 2003 and July 2006 were reviewed. Of these, we studied 149 patients who were diagnosed with unilateral prostate cancer at prostate biopsy but with ≥ pT2c disease in the RP specimen. Biopsies were taken by several physicians and therefore not standardized, as 107 patients (71.8%) were referred from other institutions or urologists.

All RP specimens were processed and evaluated by one expert genitourinary pathologist (R.G.). The prostate was fixed in buffered formalin (4%) after gross examination. It was dissected by standardized multiple transverse cuts, using a modification of the previously described technique of Bettendorf et al.[15]. Briefly, the specimens were weighed and measured in three dimensions; the prostate volume was calculated as length × width × height × 0.5. The apical portion was separated (5 mm section) and sectioned parasagittally. The basal region of the prostate was also separated from the specimen (tangential cut, thickness 2 mm); 5 mm slices were laminated from the apex to the basis. Macroscopic tumour areas and their relation to the prostate capsule were registered. Microscopic prostate cancer was outlined on histological haematoxylin and eosin-stained sections, and the total number of tumour areas, and data on capsular involvement and positive surgical margins, were recorded. All prostate cancer areas were mapped in a schematic standardized protocol. The tumours were graded and staged according to the Gleason and Helpap grading system [16,17]. Tumours were staged pathologically using the TNM staging system [18]. Significant Gleason sum upgrading was defined as a Gleason sum increase either from ≤6 to ≥7 or from 7 to ≥8 between the biopsy and RP specimens [11,19]. We accordingly defined a significant Gleason sum downgrading as a decrease either from ≥7 to ≤6 or from ≥8 to ≤7 between the biopsy and RP specimens.

For digital prostate mapping, all tumour areas documented in the representative transverse sections of the pathology report were transferred to a corresponding digital grid database to reconstruct a map of the entire prostate (Fig. 1). Thus, the different regions of the prostate were represented by a total of 794 units. To provide an exact anatomical location, the units involved with tumour areas were counted and summed, reflecting a relative estimate of the tumour volume rather than absolute measures.

image

Figure 1. Schematic transverse sections of the prostate as they appear in the pathology report (A) of RP specimens. Prostate cancer areas were transferred into a digital database, representing the prostate by a total of 794 units for precise mapping of tumour masses that remained undetected by prostate biopsy (B).

Download figure to PowerPoint

The hospital charts and physician records of each patient included were reviewed, abstracted and entered into a database. Descriptive statistics were calculated to characterize the patients. Data are expressed as the mean (sd) unless otherwise indicated. The number of tumour areas in the different prostatic lobes were compared using Student’s t-test. The tumour volume was correlated with pathological and clinical data (e.g. the presence of capsular involvement/extracapsular tumour growth, positive surgical margins, preoperative PSA levels, and significant upgrading) using the Pearson or Spearman rank correlation coefficient, as appropriate. All statistical tests were two-sided and statistical significance was assumed for P < 0.05.

RESULTS

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

Of 559 patients, 149 (26.7%) were diagnosed with unilateral prostate cancer by prostate biopsy but had bilateral disease in the RP specimen; the characteristics of these 149 patients are summarized in Table 1. The age of the patients was 64.5 (5.8) years and the PSA level before biopsy was 8.5 (5.4) ng/mL. Neoadjuvant hormonal therapy was applied in eight patients (5.4%).

Table 1.  The clinical and pathological characteristics of 149 patients with bilateral prostate cancer in RP specimens and preoperative unilaterally false-negative prostate biopsy results
VariableN (%) or n/N
  1. Significant Gleason sum upgrading was defined as a Gleason sum increase either from ≤6 to ≥7 or from 7 to ≥8 between the biopsy and RP specimens; similarly, significant Gleason sum downgrading was defined as a Gleason sum decrease either from ≥7 to ≤6 or from ≥8 to ≤7 between the biopsy and RP specimens.

Biopsy Gleason sum score (preoperative) 
 4  4 (2.8)
 5  8 (5.5)
 6 86 (59.3)
 7 34 23.5)
 8 10 (6.9)
 9  3 (2.1)
 Not available  4
Laterality of false-negative lobe 
 Right 74 (49.7)
 Left 75 (50.3)
Pathological Gleason sum score 
 5 24 (17.0)
 6 31 (22.0)
 7 75 (53.2)
 8  5 (3.6)
 9  6 (4.3)
 Neoadjuvant hormone therapy  8
Pathological grading (Helpap) 
 2a 60 (42.6)
 2b 64 (45.4)
 3a 16 (11.4)
 3b  1 (0.7)
 Neoadjuvant hormone therapy  8
Pathological tumour stage (pT) 
 2c103 (69.1)
 3a 35 (23.5)
 3b  9 (6.0)
 4  2 (1.3)
Pathological upgrading 
 Total 58 (38.9)
 Significant 51 (34.2)
Pathological downgrading 
 Total 27 (18.1)
 Significant 10 (6.7)
Positive surgical margins 
 Total (149) 36 (24.3)
 By stage 
 p2c 18 (17.5)
 p3a 13 (37.5)
 p3b  3/9
 p4  2/2

For all 559 patients who had RP, false-negative biopsy results were less frequent in patients who had at least a 10-core biopsy than in those from whom nine or fewer biopsy cores were obtained (false-negative rate 17.2% vs 30.6%, respectively). In 139 of the 149 study patients, a mean (sd, range) of 9.1 (3.6, 4–21) cores were obtained per patient. In 37 of these 139 patients (26.6%) a sextant biopsy was taken, whereas 7–10 cores were taken in 72 patients (51.8%); in the remaining 30 patients (21.6%) ≥10 cores were obtained. There was no significant difference in the number of cores obtained between the positive and the false-negative lobe. The mean (range) number of positive cores on the left and right positive biopsy side was 2 (1–5) and 2 (1–9), respectively. In all, 55 patients (36.9%) were diagnosed with prostate cancer in a single biopsy core only. Over 90% of patients had a Gleason sum score of ≤7 both on biopsy and on RP.

The volume of the prostate specimens was 45.8 (17.8) mL. When comparing the Gleason sum score of the RP specimen with the results of the biopsy report, there was significant upgrading in 51 patients (34.2%). The pathological Gleason sum score was downgraded in 27 patients (18.1%), with significant downgrading in 10 (6.7%) (Table 1); 36 patients (24.3%) had positive surgical margins on histology (Table 1). In 60 patients (40.3%) an involvement of the prostate capsule on the false-negative side at biopsy was diagnosed. Nerve-sparing RP was performed on the false-negative side of 51 patients (34.2%). In this subgroup, 13 patients (25.5%) had positive surgical margins in the RP specimen.

The total distribution of prostate cancer in patients with false-negative biopsy results is shown in Fig. 2A; the dorsolateral regions were most commonly involved. The mean tumour volume of the false-negative lobe was significantly lower than of the positive lobe (18.9 vs 47.5 units, P < 0.001). An increasing tumour volume was significantly associated with a higher PSA level (P = 0.002), Gleason score, Helpap grading, pT stage (each P < 0.001), the presence of positive surgical margins (P = 0.002) and the occurrence of upgrading (P = 0.035). The most frequent location of undetected tumour areas was in the dorsalateral third and in the anterior apical region of the prostate (Fig. 2B). In the false-negative lobes, there were very small areas of prostate cancer (≤4 units) in 48 (32.3%) and large tumour masses (≥16 units) in 29 patients (19.5%). In 36 of 149 patients (24.2%) the tumour volume on the false-negative lobe was equal or higher than on the positive side.

image

Figure 2. (A) Distribution of the merged total tumour areas in RP specimens (colour-coded) in patients with unilaterally false-negative prostate biopsy results, in an anterior view. (B) Merged areas of prostate cancer areas in representative transverse sections of only false-negative prostate lobes (left, 75; right, 74) in an anterior view. Note the undetected involvement of the dorsolateral and apical portions of the prostate.

Download figure to PowerPoint

DISCUSSION

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

The pathological biopsy report guides treatment decisions in prostate cancer for nerve-sparing RP and regional lymphadenectomy. Despite more evidence reporting an increase in Gleason score and laterality concordance between TRUS-guided prostate biopsy and RP, considerably many patients remain under-graded and under-staged [11–13,20,21]. It was reported that an initial transrectal sextant or 10–12-core biopsy might miss tumour areas in up to 30% of patients, leading to false-negative biopsy results [20,21]. Depending on the patient cohort, the indications and the biopsy technique, even repeat biopsies yield a rate of up to 50% of undiagnosed prostate cancer [22,23].

In the present series, there was clinical under-staging in 26.7% of 559 consecutive patients recently treated with RP in our department. This value is markedly lower than in two recent reports, in which ≥65% of patients had prostate cancer on the negative biopsy side [24,25]. In our subgroup analysis of patients who were biopsied according to current recommendations 17.2% were still clinically under-staged.

Currently, increasingly many small- volume tumours with minimal, potentially insignificant prostate cancer are diagnosed [8]. One consequence of this epidemiological trend is that more patients have been offered focal therapy with curative intent. It seems an attractive option, as it is supposed to minimize side-effects from prostate treatment and might be suitable for patients with unilateral prostate cancer according to the biopsy report [10]. Mouraviev et al.[26] reported a rate of 19.2% of unilateral prostate cancer in 1184 patients who had RP for pathological organ-confined disease suggesting that 20% of men were potentially amenable for hemi-therapy of the prostate. To date, long-term oncological outcomes of this approach cannot be determined, as the natural history of prostate cancer is long and only short follow-up periods are available [9]. Focal therapy might be successful if no or only small areas of prostate cancer are present contralaterally. However, in the present study there was a high rate of clinical under-staging with respect to the laterality of prostate cancer. Moreover, our findings showed an emerging high number (40.3%) of tumours that involve the prostatic capsule on the negative biopsy side. As over a third of our patients had nerve-sparing RP on the negative biopsy side, interfascial dissection seems inappropriate in these patients with regard to oncological safety. This finding is of striking clinical relevance, including the subgroup of patients biopsied according to current recommendations.

There was significant upgrading of the Gleason sum score in RP specimens over the biopsy results in 34.2% of the present patients. The largest series on upgrading in prostate cancer published to date comprised 4789 unselected men; there was significant upgrading in 28.2%. A higher PSA level, higher tumour stage and a higher Gleason sum score at biopsy were independent predictors of significant upgrading [11]. By contrast with these findings, the PSA level failed to predict significant upgrading in the present series of highly selected patients with unilateral prostate cancer (data not shown). However, the relatively high rate of Gleason sum upgrading in the present study might indicate the inadequacy of the baseline biopsy before RP, as most patients had fewer than 10 cores taken. In our patients and in other series these relatively high rates of significant Gleason sum score upgrading have decreased with the more recent use of the modified Gleason sum score [27,28].

According to our results from digital prostate mapping, current biopsy strategies miss tumour areas mainly in the dorsolateral segment in the lower third of the prostate, as well as in the apical region. Biopsy techniques should therefore be accordingly modified by directing the ultrasound probe and the needle more dorsally, as well as towards the apex. We identified only a few patients with very small areas of prostate cancer on the negative biopsy side, potentially being easily missed by biopsy. Indeed, nearly a quarter of the tumour volumes located on the negative biopsy side were equal or higher than on the contralateral lobe. This suggests some relevance of these areas, as tumour size has been identified as an important variable associated with the clinical significance of prostate cancer [29,30]. Modern imaging technology (e.g. MRI spectroscopy) could help in directing the needle towards areas of suspected cancer involvement [31] as an adequate number of cores taken might still miss prostate cancer of relatively high volume. Accurate prostate mapping, as used in the present study, can complement these imaging technologies by providing knowledge on the most frequent locations of missed tumour areas of prostate cancer. This is of importance, e.g. for patients in whom imaging studies remain negative despite an elevated PSA level.

Our descriptive study is limited because relatively few cores were taken during prostate needle biopsy, and did not meet current guideline recommendations in many patients. A standardized biopsy sampling according to the guideline recommendations and applied to all patients included might significantly lower the rate of undetected foci of prostate cancer. There is also a lack of clinical follow-up information on the rate of biochemical recurrence. However, as prostate cancer is known to be a multifocal disease in the vast majority of cases, we primarily aimed to provide information on the location of missed tumour areas in current biopsy practice. Our data can help to direct the needle to areas with a high probability of missed prostate cancer, by accurately visualizing all tumour locations, thus optimizing current biopsy strategies. These areas not only include the lateral portions of the gland, as recommended for the extension of the classical sextant protocol, but in particular the apical region of the prostate.

In summary, the apical and the dorsolateral region of the prostate are not adequately represented in current biopsy strategies. Clinical under-staging occurs in over a quarter of patients with prostate cancer, and tumour areas on the negative biopsy side are often clinically significant by size. Our findings indicate a direct clinical implication for nerve-sparing RP and recently introduced focal therapy for prostate cancer. There is still a need to modify and standardize biopsy protocols, and to integrate modern imaging devices, to establish an optimized diagnostic tool for prostate cancer.

REFERENCES

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