Main

The identification of seminal vesicle invasion and lymph node metastasis have an important role in determining a patient's prognosis after radical prostatectomy, and may guide appropriate subsequent therapy by use of a number of predictive models.1, 2, 3, 4 However, in spite of the large and increasing numbers of radical prostatectomies performed in developed countries in the last 20 years, there is considerable variation in the pathological handling of the seminal vesicles in radical prostatectomy specimens and of associated lymph nodes. Subtle differences in the guidance issued, or in omissions of detail, may lead to significant reporting differences.5, 6, 7, 8, 9 The consequences of this are considerable, leading to great difficulty in comparing results from one institution to another (or even one pathologist to another) and recent surveys have exemplified these problems.10 From the patient's perspective, the risks of recurrence may be given, which are misleading, and potentially inappropriate therapy instituted. The International Society of Urological Pathology therefore considered both these areas of contention as part of a consensus conference.

International Society of Urological Pathology Survey and Consensus Conference

In order to identify the methods and practices most commonly employed by urological pathologists worldwide, a web-based survey on handling and reporting of radical prostatectomy specimens was distributed to 255 members of the International Society of Urological Pathology.

The International Society of Urological Pathology survey was followed up with a consensus conference held in conjunction with the 2009 Annual Scientific Meeting of the United States and Canadian Academy of Pathology held in Boston, Massachusetts. The aim was to obtain consensus relating to the handling, staging and reporting of radical prostatectomy specimens. Those who completed the electronic survey were invited to attend the consensus conference that was held on 8 March. The detailed process related to the survey and consensus meeting are reported in an introductory paper.11

Representatives from five working groups appointed to coordinate the consensus process presented background information and results from detailed literature reviews to the meeting. The survey results were then discussed with the objective of achieving agreement on controversial issues. Following this, a ballot was taken on a series of questions that were designed to address these controversial issues and an achievement of 65% agreement on voting was considered to be consensus. With a few exceptions, there was agreement between the survey results and the consensus conference results. The results of the individual working group reports related to specimen handling, T2 substaging, extraprostatic extension, seminal vesicle and lymph node involvement and margin status are reported separately.12, 13, 14, 15

Many recommendation of this consensus conference have already been incorporated into international guidelines, including the recent College of American Pathologists protocol and checklist for reporting adenocarcinoma of the prostate and the structured reporting protocol for prostatic carcinoma from the Royal College of Pathologists of Australasia.16, 17

Seminal vesicle invasion

Macroscopic Assessment

Seminal vesicle invasion by prostatic adenocarcinoma (pT3b) has generally been shown to be a predictor of poor prognosis after radical prostatectomy,18, 19, 20, 21 and is commonly associated with other forms of extraprostatic extension such as infiltration into surrounding adipose tissue. Despite this, the literature relating to seminal vesicle infiltration by tumor was greatly distorted by differences in how the seminal vesicles are sampled and assessed. This issue was considered in a comprehensive literature review22 that revealed large differences in major series, in both the percentages of cases with seminal vesicle invasion and in the 5-year recurrence-free survival that ranged from 5 to 60%.19, 20, 21, 23, 24, 25, 26, 27 The main conclusions from the literature search were that conformity to one definition and a uniform approach to seminal vesicle invasion is essential for the analysis and comparison of future radical prostatectomy series, in order to derive staging information that provides meaningful data for survival analysis and prognostic tables.

Survey and Consensus Conference Results

There was no survey consensus as to the most appropriate method for dissection of the seminal vesicles (Table 1). Of survey respondents, 45% reported that they sliced seminal vesicles transversely, whereas 33% sliced longitudinally and 18% employed a combination of these techniques. There were similar degrees of disagreement as to whether seminal vesicles should be submitted completely, with 38% practicing complete embedding whereas 52% only partially embedding the seminal vesicles, and the remainder completely embedding only in some cases. The methods of complete sampling were similarly highly variable among the respondents to the survey, although there was general agreement that sections taken from the junction of the prostate and the seminal vesicle should be mandatory. It was noted that the tip or mid portion of the seminal vesicle was sampled by a minority and the conference achieved consensus by agreeing that complete embedding of the seminal vesicles should not be mandatory (Table 1).

Table 1 Comparison of survey results with consensus conference results
Full size table

Based on the survey results, it is recommended that the junction of the seminal vesicle and the prostate should always be assessed for contiguous spread of tumor. This should be considered the minimum necessary to adequately sample the seminal vesicles. Remaining blocks to assess the rarer situation of noncontiguous spread could be taken if desired or if areas macroscopically suspicious for invasion are identified.

Another area of controversy is the necessity of sampling the vas deferens margins. These were sampled by some respondents to the survey with the rationale that these represent surgical margins. We are unaware of any study that addresses this issue specifically and the separation of this feature from seminal vesicle invasion. It would seem logical that positivity at this margin would be associated with contiguous invasion of the seminal vesicles, but sampling here appears to be unreported in most major series. In the survey, there was a significant split with 46% of pathologists reporting sampling the vas deferens margin; however, at the consensus conference, there was consensus that this should not be mandatory.

Seminal Vesicle Microscopic Assessment

Seminal vesicle invasion is defined by some groups as tumor infiltration of the muscular wall (Figure 1a),28 whereas other older definitions have included invasion of the periseminal vesicle adipose tissue (Figure 1b);29 thus, different series assess differing microscopic criteria. Furthermore, many studies do not specify which assessment method was used, and hence these data are difficult to analyze. If tumor infiltration of the adipose and connective tissue surrounding the seminal vesicle is to be defined as seminal vesicle invasion, then any distinction from extraprostatic invasion outside the seminal vesicles will be difficult and may have to be based on arbitrary measures. There was a consensus at 80% in the survey that only muscularis invasion constitutes seminal vesicle invasion, although a substantial minority would also accept peri-adipose tissue invasion around the seminal vesicle as indicative of tumor infiltration. There are large problems with defining exactly how near the adipose tissue invasion has to be to the seminal vesicles, which further complicates the distinction between extraprostatic extension and seminal vesicle invasion. The inclusion of invasion of adipose tissue that surrounds the seminal vesicles or invasion in the invaginated space as a feature of seminal vesicle invasion raises further problems of definition relating to how near to the seminal vesicles does the cancer need to be. On current evidence this would have to be an arbitrary distance, chosen with little rationale. For these reasons, muscularis invasion is recommended as the definition of invasion of the seminal vesicles.

Figure 1
figure 1

(a) Extraprostatic seminal vesicle invasion (pT3b). Tumor is seen infiltrating the muscle wall (left). The seminal vesicle epithelium is not infiltrated by tumor (right). (b) Extraprostatic seminal vesicle invasion with small cancer focus growing from outside (right) in the fibroadipose tissue. Seminal vesicle glands are seen in lower left corner. (c) Prostate cancer invading the intraprostatic part of the seminal vesicle (pT3a). Central zone of the prostate is seen to the left. Cancer is seen infiltrating the intraprostatic part of the seminal vesicle to the right. (d) Lymph node metastasis with arrow indicating how maximum diameter should be measured.

Full size image

A second more contentious problem is the distinction between intraprostatic and extraprostatic seminal vesicle invasion (Figure 1c). This problem was presented in the survey as a key issue of controversy. The seminal vesicles are not purely extraprostatic organs, but extend and connect with the prostate ejaculatory duct. As ejaculatory duct epithelium is difficult to distinguish from the seminal vesicles, and is present within the anatomical bounds of the prostate, differences in definition will lead to variations in the number of cases considered to have seminal vesicle invasion.

If intraprostatic seminal vesicle invasion is considered pT3b, then its differentiation from ejaculatory duct invasion status is of importance. This is not an issue if only extraprostatic seminal vesicle invasion is considered as pT3b. One interesting study that examined tumors showed invasion only of the intraprostatic portion of the seminal vesicle and ejaculatory duct, and showed that the survival rate was considerably more favorable than for those with extraprostatic seminal vesicle invasion.30

Prostatic carcinoma can invade the extraprostatic seminal vesicles by spreading along the ejaculatory duct (type 1), by direct invasion at the base of the prostate or by extending into peri-seminal vesicle soft tissue and then into the wall of the seminal vesicle (type 2) or, rarely, via discontinuous metastases (type 3).28, 31 Although these routes of infiltration are classified separately, it has been noted that types 1 and 2 invasion frequently occur together. The distinction of the main types of extraprostatic seminal vesicle invasion and their prognostic significance also has a scanty literature.32 The major findings appear to be that the discontinuous form is much rarer than direct spread. Although this earlier study suggested that those with discontinuous seminal vesicle invasion have a better outcome, this did not hold up on multivariate analysis when lymph node metastasis was also considered: unsurprisingly, considering their rarity. The risk of relapse has also been stratified by the amount of seminal vesicle involvement, but from the lack of specific comments from the survey, this is not yet routinely performed.33, 34

In the survey, a majority either did not offer an opinion or did not consider ejaculatory duct invasion to be extraprostatic extension. Additional to this, 89% of respondents did not considered ejaculatory duct invasion to be a feature of pT3b staging category (Table 1). As there is little evidence based on this issue, a satisfactory outcome would be the achievement of uniformity of practice between differing practitioners and institutions.

If intraprostatic seminal vesicle infiltration by tumor is to be indicative of stage T3b disease, then distinction from ejaculatory duct invasion becomes a critical issue. If not, then pathologists will have to accept the fact that there is no distinct cutoff between extra- and intraprostatic portions of the seminal vesicles. Amputation of the seminal vesicles close to their base would minimize this issue, but by no means eradicate it.

After consideration of the literature, the conference achieved consensus, voting that only invasion of cancer into the extraprostatic portion of the seminal vesicles should be considered to be seminal vesicle invasion (Table 1). It was suggested that when blocks of seminal vesicle are taken, care should be given to ensure that the seminal vesicles are not surrounded by prostatic tissue.

Survey and Consensus Conference Results

The majority of respondents to the survey do not distinguish the type of seminal vesicle invasion, when present, although it should be straightforward to distinguish between the different forms of invasion if required. As mentioned above, there is only limited evidence that type of spread is of prognostic significance once other prognostic factors are accounted for. In view of this, there was consensus at the conference that such distinctions are not necessary in routine practice.

The rare situation was considered that if a tumor is present within endothelial-lined spaces in the muscular wall of the seminal vesicles, should this be categorized as pT3b? In the survey, 55% thought that vascular invasion was the equivalent of stromal invasion, whereas 45% did not. This situation was mirrored in the conference where the split was 56 and 44% (Table 1). Although the TNM classification advises that where there is doubt, the lower stage should be used, there is clearly no consensus at present in relation to examination of the prostate gland. Fortunately this situation occurs very infrequently, as vascular invasion by tumor in the prostate is itself uncommon.

Lymph node metastasis

Lymph Node Macroscopic Assessment

As radical prostatectomy is generally reserved for men at low risk of metastatic disease, the rate of lymph node involvement is generally low. Possible special techniques to examine the lymph nodes associated with surgical resection of the prostate include frozen sectioning of pelvic lymph nodes and sentinel lymph node sampling. Opinions as to the utility of frozen section diagnosis in assessment of pelvic lymph nodes at the time of radical prostatectomy have varied over the last 20 years. In 1986, Epstein et al35 reviewed 310 patients and found that frozen sections detected 67% of positive lymph nodes that were grossly uninvolved by tumor, and 100% of the patients with grossly involved lymph nodes. In view of this detection of occult tumor deposits in macroscopically normal glands, the authors considered frozen section to be a valuable adjunct to radical prostatectomy. Young et al36 in 1999 examined the cost and accuracy of frozen sections before radical prostatectomy and found a false-negative rate of 33%. These authors estimated the cost of metastatic cancer detection to be over $10 000 and this, coupled with their demonstrated negative rate, led them to conclude that frozen section was not warranted as a routine procedure. Beissner et al37 in 2002 found an even higher false-negative rate of 70% but, by stratifying patients into low, intermediate and high risk based on nomograms, the sensitivity of lymph node metastases was improved. They concluded that low-risk patients (stage T2 or less, prostate specific-antigen (PSA) of ≤10 ng/ml and Gleason score of ≤6) gain no benefit from intraoperative frozen section, whereas the intermediate group (stage T2 or less and Gleason score 7 and/or PSA between 10.1 and 20 ng/ml) gained minimal benefit. The high-risk groups who have opted for surgery do gain benefit as they have a high risk of positive lymph nodes, and the detection of lymph node metastases would spare the patient the morbidity associated with an unnecessary radical prostatectomy. In parallel, the refinement of predictive tables for the risk of lymph node metastasis relative to biopsy Gleason score and presenting PSA has reduced the necessity for pre- or peri-operative nodal examination.38 These findings suggest that the current evidence is that intraoperative frozen section is mainly helpful in high-risk patients to prevent unnecessary major surgery. Another intraoperative sampling technique, sentinel node surgery, has only recently been used in prostate pathology, and is usually used in a research setting.39

Staging of prostate cancer may be improved through the recovery of greater numbers of lymph nodes for examination purposes.40 Xylene and other clearance techniques have been used by some to increase lymph node retrieval.41, 42 Fat clearance has been shown to increase lymph node harvest in other tumors such as carcinoma of the colon,43 but the lymph nodes in colonic mesentery are well defined and lack the degree of fat infiltration seen in the pelvic nodes. Wawroschek et al42 showed that by examining lymph nodes at several levels, combined with the use of immunohistochemistry, the node-positive rate in low-risk patients increased from 5 to 11%, but that there was a smaller increase, from 34 to 37%, in intermediate-risk patients. The cautionary note regarding the use of immunohistochemistry to detect micrometastasis is that we are uncertain of the prognostic significance of any tumor so detected, as there is a lack of studies in this area. Certainly, in some other comparable tumors, such as breast cancer, metastases <2 mm are classified separately to those of greater dimension. Considering the rarity of lymph node metastases in prostate cancer when compared with cancers from other organs, it appears unlikely that such stratification will be feasible in the foreseeable future.

Survey and Consensus Conference Results

These conclusions were reflected in the survey results, where 22% reported never receiving an intraoperative frozen section, 57% received a frozen section in <10% of cases and only 21% of respondents received a frozen section with >10% of radical prostatectomy cases. From this it is clear that frozen section is used in a minority of cases, possibly being those of high risk where metastatic deposits to the lymph nodes are more likely. Only 2% of respondents reported ever receiving sentinel node samples.

Lymph node excision is not inevitably performed as part of a prostatectomy, and of the surveyed centers, 21% reported receiving lymph nodes with >90% of the specimens, 21% with 51–90% of specimens, 31% with 10–50% specimens and 26% with 1–10% of specimens. It was noted that one center never received lymph nodes as part of a radical prostatectomy procedure.

There was similar variability in how lymph nodes were submitted. Of the survey respondents, 2.5% reported receiving the lymph nodes, as left and right pelvic, in one container, 64% received them as left and right pelvic nodes in separate containers and 31% received specifically designated lymph node groups. There appears to be little uniformity among urologists as to how lymph nodes are submitted.

There was no consensus as to the optimal method for sampling the lymph nodes. In the survey, 27% reported that they sampled all the tissues submitted by the surgeon, 60% submitted only macroscopically identifiable nodal tissue and 8% processed only a sample of each lymph node identified. This was identified as a key area for the consensus conference, which also failed to reach consensus, although there was consensus that sampling of only a portion of each lymph node was inadequate. In the conference, 41% considered that it was necessary to submit all fat and nodal tissues, whereas 59% of participants considered the submission of purely nodal tissue sufficient.

Lymph node microscopic assessment

There are little published data relating to the pathological examination of pelvic lymphadenectomies in patients undergoing radical prostatectomy. The number of lymph nodes obtained in a lymphadenectomy dissection varies widely among centers, which is a function of surgical technique as well as pathological practice. It was further recognized at the consensus conference that methods of nodal identification of lymph nodes by microscopy vary between pathologists. One study reported that a median of 16 nodes (range 5–40) could be detected for each patient, and that the rate of cancer detection increased with the number of nodes present, suggesting that a minimum of 13 nodes was required.40 Such high yields, as reported in the survey, are not the norm and there was clear consensus that the number present should be recorded. This would be in concordance with lymphadenectomy assessment of other major cancers; however, it is recognized that counting of pelvic lymph nodes presents more of a challenge for pelvic lymph nodes than at many other sites, because of the large number showing replacement by adipose tissue.44 It has also been suggested that lymph nodes in patients with significant prostate cancers often show reactive change, making them easier to identify.41

Apart from assessment of the number of lymph nodes and their involvement by tumor, a number of studies have examined the characteristics of the deposits and their association with outcome. The diameter of the largest metastasis (Figure 1d) appears to be more predictive of cancer-specific survival than the number of positive nodes alone,45, 46, 47 whereas the presence of extranodal extension has been shown not to be predictive on multivariate analysis.46

Survey and Consensus Conference Results

Quantitation of the number of lymph nodes seen on microscopy was considered necessary information for a pathology report by the survey respondents, with 95% of pathologists reporting that they recorded this in the final report. However, it was noted that there is great variability in the number of nodes that are usually identified. Thus, 46% of respondents reported finding on average <5 lymph nodes per radical prostatectomy specimen, 40% found 5–10 lymph nodes, 8% found 11–15 lymph nodes whereas 3% found >15 lymph nodes.

In spite of this evidence, in the survey, 45% of respondents stated that they did not report on the size of the largest metastasis in lymph nodes, whereas 30% did report this. In all, 8% reported the size of the largest lymph node with a metastasis, and 17% reported both the size of the largest metastasis and the size of the largest involved lymph node. Of interest, 88% of the survey respondents reported extracapsular extension in spite of the fact that this was considered to be a feature of less prognostic significance. At the conference, there was a consensus of 81% that the diameter of the largest lymph node metastasis should be included in the final pathology report (Table 1).