SEARCH

SEARCH BY CITATION

Keywords:

  • Bladder;
  • Cancer;
  • Metastasis;
  • Staging;
  • Lymph node

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CURRENT STANDARDS
  5. DETECTION OF MICROMETASTATIC NODAL DISEASE
  6. FUTURE DIRECTIONS AND RESEARCH AGENDA
  7. CONCLUSION
  8. CONFLICT OF INTEREST
  9. REFERENCES

To review the current and newer, alternative methods for evaluating lymph nodes for tumor involvement in bladder cancer as relapse rates for organ-confined disease remain high despite improvements in surgical technique, suggesting the possibility of understaging. To propose a research agenda based on these findings. A PubMed literature search was performed to identify studies examining the prognostic implications of and outcomes associated with lymph node involvement in bladder cancer as well as those that utilized newer methodologies to identify the possibility for metastatic disease. Lymph node involvement remains one of the strongest predictors of clinical outcome in bladder cancer. Histologic and molecular techniques for identification of lymph node metastasis provide a sensitivity and specificity equal to if not higher than standard pathologic evaluation. Further research into this field would help to elucidate the potential utility of these techniques with regard to proper staging and potential relevance to clinical outcomes.


Abbreviations
PLND

pelvic lymph node dissection

H&E

haematoxylin and eosin

SEER

Surveillance Epidemiology and End Results

RFS

recurrence-free survival

IHC

immunohistochemical

RT-PCR

reverse transcriptase-PCR

qPCR

real-time quantitative PCR

CK19

cytokeratin 19

CK20

cytokeratin 20

MUC7

mucin 7

UPII

uroplakin II.

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CURRENT STANDARDS
  5. DETECTION OF MICROMETASTATIC NODAL DISEASE
  6. FUTURE DIRECTIONS AND RESEARCH AGENDA
  7. CONCLUSION
  8. CONFLICT OF INTEREST
  9. REFERENCES

Bladder cancer is the second most common genitourinary malignancy, with TCCs comprising ≈ 90% of primary bladder tumours. It is estimated that 70 530 individuals in the USA will be diagnosed with bladder cancer and that 14 680 individuals will die from this disease in 2010 [1]. Stage is fundamental to patient outcome and relative survival rates for bladder cancer are lower compared with other cancers, such as breast or prostate cancer [2]. Overall, the 5-year survival rate for bladder cancer is 80%, but, with stage progression, this drops dramatically from 92% to 97.2% for carcinoma in situ to a dismal 6–24% in cases of metastatic disease [2–7]. Although up to 86% of patients present with superficial or localized tumours, 20–40% present with, or progress to develop, invasive disease which carries a significant increase in the likelihood of having occult metastases.

Radical cystectomy with a pelvic lymph node dissection (PLND) remains the cornerstone of treatment for carcinoma invading bladder muscle. The aims of surgery are cure, local disease control and pathological staging. Advanced pathological stage may indicate the need for adjuvant therapies and so the goal for pathological outcome is organ-confined, node-negative (stage ≤ pT2N0) disease at the time of surgery. However, even in the setting of ‘favourable’ pathology, relapse rates remain high. Several studies have shown that 20–40% of patients with stage pT2N0 or lower tumours will ultimately recur, which strongly indicates the possibility of pathological understaging [3,8,9].

The presence of metastatic, or micrometastatic, nodal disease has been shown to be the most significant independent prognostic factor for mortality [3,9,10]. Given the importance of this finding, it is not surprising that there has been considerable research in recent years addressing ways to improve the detection and interpretation of nodal involvement. The present review presents an overview of the current findings with regard to lymph node metastasis in bladder cancer and how this relates to prognosis after cystectomy. In addition, the potential role of newer proposed methods for the detection of nodal micrometastatic disease is reviewed and a research agenda is put forward.

CURRENT STANDARDS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CURRENT STANDARDS
  5. DETECTION OF MICROMETASTATIC NODAL DISEASE
  6. FUTURE DIRECTIONS AND RESEARCH AGENDA
  7. CONCLUSION
  8. CONFLICT OF INTEREST
  9. REFERENCES

It has been proposed that a PLND performed at cystectomy serves both a diagnostic and therapeutic role [5,7,11–14]. The current approaches to PLND are based on the removal of lymph nodes most commonly harbouring metastatic disease, notably the external iliac, obturator and hypogastric lymph nodes. The boundaries for a standard PLND generally include the bifurcation of the common iliac vessels superiorly and the genitofemoral nerve laterally. The inferior and medial boundaries have more variable delineations, including the obturator nerve, bladder and internal iliac vessels medially and the endopelvic fascia, circumflex iliac veins, and node of Cloquet inferiorly [15–18]. After lymphadenectomy, the current standard for evaluation is formalin fixation with haematoxylin and eosin (H&E) staining of 5-µm thick sections of each node, followed by pathological assessment for the presence of tumour cells. By these standards, ≈ 25% of patients are found to have positive nodes at the time of cystectomy [19].

The Union Internationale Contre le Cancer recommends assessment of at least eight lymph nodes after lymphadenectomy, but results from the Surveillance Epidemiology and End Results (SEER) registry show that 40.3% of patients undergoing cystectomy do not even undergo a PLND. Furthermore, 67.9% of patients reviewed who underwent PLND had <6 nodes removed, suggesting suboptimal implementation of the current guidelines [20].

Many studies have found the pathological status of pelvic lymph nodes to be the strongest independent predictor of disease-specific mortality in patients with bladder cancer [3,8–10]. Quantitative measures of lymph node involvement, such as tumour burden (number of positive nodes) and lymph node density (number of positive nodes per total number of nodes removed), have been shown to be significant prognostic factors for stratification of patient outcomes (Table 1[20–31]). Consequently, the concept of an ‘extended’ nodal dissection has been proposed in order to increase the number removed and improve the detection of nodal involvement. Given the significant prognostic implications of pathological nodal status on outcome, the removal and accurate evaluation of nodes at cystectomy are essential for proper staging.

Table 1.  Clinical summary of selected studies evaluating the prognostic significance of various quantitative measures of pelvic lymph node pathological status
StudyTreatmentStudy designSample size, nResults
Extent of lymph node dissection (number of nodes removed)
Konety et al. (2003) [20]Cystectomy and PLNDRetrospective database review1923Review of the SEER database showed that removal of 10–14 lymph nodes had the largest positive effect on survival for all stages of bladder cancer, lending support for an extended PLND.
Leissner et al. (2004) [21]Cystectomy and extended PLNDProspective case series29043.7% of all positive nodes occurred outside the confines of a standard PLND. 3/29 patients with only one positive node occurred outside the confines of a standard PLND, indicating the necessity of an extended PLND for proper staging.
Koppie et al. (2006) [22]Cystectomy and standard PLNDRetrospective case series1121The probability of survival after cystectomy continued to increase with increasing numbers of lymph nodes removed at PLND.
Steven and Poulsen (2007) [23]Cystectomy and extended PLNDProspective case series33634.4% (22/69) of patients with nodal involvement had metastases above the bifurcation of the common iliac vessels, which would have been missed by a standard PLND.
Capitanio et al. (2008) [24]Cystectomy and standard PLNDRetrospective case series731The authors showed a higher probability of detecting nodal disease if 25 lymph nodes were removed vs 15 nodes (75% vs 10% probability).
Dhar et al. (2008) [17]Cystectomy and limited PLND (n= 336) vs cystectomy and extended PLND (n= 322)Retrospective case series658The 5-year RFS rate for patients who underwent limited PLND was lower than for those who underwent extended PLND (67% vs 77%).
Tumour burden (number of positive nodes)
Stockle et al. (1996) [25]Cystectomy alone (n= 86) vs cystectomy plus adjuvant chemotherapy (n= 80)Retrospective case series166Patient prognosis declined in a stepwise manner dependant on increasing lymph node involvement. Improved survival rates in the group receiving chemotherapy were demonstrated at each lymph node stage, significantly favouring those with only one lymph node involved.
Mills et al. (2001) [26]Cystectomy and standard PLNDProspective case series452The authors found a significantly longer median survival time in patients with <5 vs those with >5 positive nodes.
Herr et al. (2002) [27]Cystectomy and standard PLNDRetrospective case series322The authors found significantly longer 10-year overall survival rates among patients with <11 vs >11 positive nodes.
Stein et al. (2003) [28]Cystectomy and standard PLNDProspective case series1054The authors found significantly longer 10-year RFS rates among patients with <8, vs those with >8 positive nodes.
Lymph node density
Stein et al. (2003) [28]Cystectomy and standard PLNDProspective case series1054Patients with a lymph node density <20% had significantly higher 5-year RFS rates compared with those with a lymph node density >20% (43% vs 10%).
Herr (2003) [29]Cystectomy and standard PLNDProspective case series162Patients with a lymph node density <20% vs those with >20% had significantly higher 5-year overall survival rates (64% vs 8%). Lymph node density was also a significant predictor of disease-specific mortality on multivariate analysis.
Cheng et al. (2006) [30]Cystectomy and standard PLNDRetrospective case series133Patients with a lymph node density <20% had significantly higher 5-year DSS rates vs those with a lymph node density >20% (54% vs 0%).
Kassouf et al. (2008) [31]Cystectomy and standard PLNDRetrospective case series248Patients with a lymph node density <20% had significantly higher 5-year disease-specific survival rates vs those with a lymph node density >20% (54.6% vs 15.3%). Lymph node density was a significant predictor of disease-specific survival on Cox regression analysis.

EXTENT OF LYMPH NODE DISSECTION

Studies that recommend an extended PLND for the detection of microscopic metastatic disease date back to the early 1980s [13,14,16,17,20,21,23,27,28,32,33]. Compared with a ‘standard’ PLND, an ‘extended’ PLND is most commonly described as including the removal of lymph nodes cephalad to the bifurcation of the common iliac vessels to the level of the aortic bifurcation, sometimes including distal aortic and caval nodes up to the level of the inferior mesenteric artery, as well as presacral nodes. Most recently, Roth et al. [34] reported that 90% of nodes in the bladder’s primary drainage fields are removed by an extended PLND, including regions along the major pelvic vessels i.e. the internal iliac, external iliac, obturator, and common iliac regions up to the level at which the ureter crosses the common iliac vessels. As such, the extended PLND increases the number of lymph nodes removed [35,36]. Steven and Poulsen [23] evaluated the role of extended PLND in 336 patients and found that among those with positive nodes, 34.4% had involvement above the bifurcation of the common iliac vessels which would have been missed by standard PLND. Leissner et al. [21] reviewed the pathological nodal status of 290 patients who underwent an extended PLND, with an average of 43.1 lymph nodes removed per patient. They found that, of the 599 positive lymph nodes detected, 43.7% (262 nodes) occurred outside the confines of the standard PLND. Furthermore, in 3/29 patients the only positive node identified occurred outside the confines of the standard PLND, suggesting that an extended PLND was necessary for proper staging. These two studies show that 6–7% of patients would have been incorrectly identified as node-negative by standard PLND.

In an attempt to quantify the number of lymph nodes that need to be removed at lymphadenectomy, several investigators have found evidence to support a range from 10 to > 25 [20–22,24]. Capitanio et al. [24] reported that the probability of accurately detecting nodal disease increased from 10% to 75% with resection of 25 vs 15 nodes. Koppie et al. [22] sought to determine a threshold number of lymph nodes and, although they did not determine a minimum number needed, they found that the probability of survival after cystectomy continued to increase with increasing numbers of nodes removed.

Several other studies have reported that expanding the nodal dissection not only assists in the detection of nodal disease but may also potentially improve survival [17,20,21,27,32,37]. Dhar et al. [17] compared 336 patients who underwent limited PLND with 322 who underwent extended PLND. They reported that, in those with pT2N0 tumours, the 5-year recurrence-free survival (RFS) rate was higher if an extended PLND was performed as compared with a limited PLND (77% vs 67%, P= 0.12). However, when comparing outcomes for patients across all stages of pT2 and pT3 bladder cancers, the 5-year RFS rate for those patients who underwent extended PLND was significantly higher than for those who underwent limited PLND (35% vs 7%, P < 0.001). In addition, Poulsen et al. [16] reported that the 5-year RFS rate among patients with pT3aN0 or lower tumour stage (specimen-confined, node-negative, as defined in his paper) who underwent extended PLND was 90%, compared with 71% in those with pT3aN0 who underwent limited PLND. These studies both suggest that the removal of a larger quantity of lymph nodes increases not only the probability of detecting metastatic disease but also patient survival. Furthermore, Konety et al. [20] found that removal of 10–14 lymph nodes had the largest positive effect on survival in patients with all stages of bladder cancer, as well as specifically for organ-confined tumours, lending additional support for the importance of an extended PLND.

These studies do not advocate a specific quantity for the ideal number of lymph nodes removed. In addition, a variety of factors can influence the number of nodes detected after PLND, including pathological handling, but it is clear that the meticulous dissection and removal of all tissue within the confines of the lymph node dissection boundaries is essential to allow for the detection of metastatic disease and potentially improved patient survival.

NUMBER OF POSITIVE NODES

The extent of nodal involvement, or tumour burden, has also been reported as an independent prognostic factor for survival in patients with bladder cancer [3,8,10,26–28]. An early study evaluating the efficacy of adjuvant MVAC/MVEC chemotherapy found that patient prognosis declined in a stepwise manner dependent on the increasing extent of lymph node involvement determined at the time of cystectomy [25]. Furthermore, improved progression-free survival rates in the chemotherapy group compared with the non-chemotherapy group were found at each lymph node stage, significantly favouring those with only one node involved vs 2–5 and > 5 nodes. This has held true in the follow-up analysis 16 years after closure of the initial study [38].

Several recent studies have also shown the prognostic influence of the degree of lymph node positivity on survival rates [10,27,28]. Mills et al. [26] reviewed the results of 452 patients and found 83 to have pathologically positive nodes. Of these 83 patients, 26 had a single node, 57 had <5, and 26 had >5 nodes involved. The investigators found a significantly prolonged median survival time (27 vs 15 months, P= 0.0027) in the group of patients who had <5 compared with those with >5 positive nodes. Herr et al. [27] determined the impact of nodal involvement among 64 of 322 patients with positive nodes at time of cystectomy. Of these patients, the 10-year overall survival rates for those with >11 and <11 nodes involved were 20% and 44%, respectively (P= 0.004). Stein et al. [28] also reported improved 10-year RFS among 244 of 1054 patients studied with positive nodes. Those patients with <8 positive nodes had a 10-year RFS rate of 40%, whereas in those with >8 the RFS was 10% (P < 0.001).

LYMPH NODE DENSITY

Lymph node density, or the ratio of pathologically positive nodes to the total number of nodes removed, has also been shown to be an independent prognostic factor for patient outcomes in bladder cancer. Lymph node density, therefore, reflects both the tumour burden as well as the extent of lymphadenectomy, or pathological examination of lymph nodes. Two initial studies showed significantly longer 5-year disease-specific survival and RFS rates in patients with a lymph node density <20%. The first study evaluated predictive factors for survival in 1054 patients. The investigators reported 5-year RFS rates of 43% and 10% in those with a lymph node density <20% or >20%, respectively (P < 0.001) [28]. The second study, which evaluated outcomes in 162 patients with node-positive bladder cancer, showed an even larger discrepancy between the 5-year overall survival rates in those with a lymph node density >20% or <20%, at 8% and 64% (P < 0.001), respectively [29]. Lymph node density was also a significant predictor of disease-specific mortality on multivariate analysis (P= 0.002).

A more recent study found that among 248 patients with positive nodes, the 5-year disease-specific survival for patients with a lymph node density >20% was significantly lower compared with those with a density <20%, at 54.6% and 15.3%, respectively (P < 0.01) [31]. Furthermore, on Cox regression analysis a lymph node density <20% remained a significant predictor of disease-specific survival, even in the setting of adjuvant chemotherapy. A larger discrepancy between 5-year disease-specific survival rates was shown by Cheng et al. [30] Of 133 patients undergoing cystectomy, those with a lymph node density >20% had a 5-year disease-specific survival rate of 0%, whereas those with a lymph node density <20% had a 5-year disease-specific survival of 54% (P < 0.001) [30]. Similar findings, supporting the negative prognostic implications associated with a higher pelvic lymph node density, have also been reported in other studies, with cut-off values ranging from 18 to 25%[20,39–43].

It is evident from the present paper that removal of an adequate number of lymph nodes, and more importantly accurate detection of positive lymph nodes, is critical to properly stage and manage patients with advanced bladder cancer. However, some authors have suggested that the association between the removal of more nodes at the time of cystectomy and improved survival is confounded by the ‘Will Rogers Phenomenon’[16,18,27,44]. Specifically, the removal of more nodes results in identification of nodal involvement in patients that would otherwise be considered node-negative, thus resulting in clinical upstaging. As a result, patients in the higher stage category, who have a larger tumour burden than those who were upstaged, experience an overall improvement in survival because of the inclusion of patients with a smaller tumour burden. In addition, removal of patients from the lower stage owing to upstaging, results in an improved survival for those in the former group because of removal of patients with newly determined nodal involvement. To counter the argument of this potential effect, many authors have instead suggested that removal of pathologically negative nodes may increase the removal of micrometastatic disease, thus decreasing the amount of residual disease and effectively increasing survival [16,17,19,20,27,37,45]. Regardless of the impact, all of these studies are reliant on proper pathological assessment to determine the presence or absence of disease in the nodal tissue.

DETECTION OF MICROMETASTATIC NODAL DISEASE

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CURRENT STANDARDS
  5. DETECTION OF MICROMETASTATIC NODAL DISEASE
  6. FUTURE DIRECTIONS AND RESEARCH AGENDA
  7. CONCLUSION
  8. CONFLICT OF INTEREST
  9. REFERENCES

Standard pathological evaluation of lymph nodes provides a limited assessment, with estimates that <0.1% of each lymph node is visually inspected for the involvement of tumour cells [46]. Given the well supported prognostic, and perhaps therapeutic, role of removing and identifying positive lymph nodes, various mechanisms for improving the detection of metastatic disease have been the focus of recent research. Detection of nodal disease depends on several factors, including the pathologist, the volume of tumour burden in the node, and the section examined. It follows that as the volume of metastatic disease decreases to the point of being micrometastatic, or <2 mm in diameter, the likelihood of detecting small clusters or even single metastatic cells in the background of normal lymphoid follicular tissue also decreases. While it has been argued that the sensitivity of H&E staining could be increased by examining a greater number of sections per lymph node, this would be a laborious and cumbersome process for the pathologist.

Several histological and molecular techniques have been explored to detect nodal micrometastases in other cancers. Immunohistochemical (IHC) staining, which can detect one tumour cell among 10 000–100 000 normal cells, more clearly delineates target tumour cells from surrounding lymphoid follicular cells than does H&E staining, thus facilitating the pathologist’s task of detecting tumour cells [47]. These techniques have been used to detect nodal disease in breast, cervical and prostate cancers, showing sensitivities greater than or equal to standard pathological evaluation [48–53]. A modification of reverse transcriptase-PCR (RT-PCR), real-time quantitative PCR (qPCR), has replaced PCR to serve as the basic technique used by several investigators for detecting nodal disease [54,55]. With this method, Gibson et al. [54] reported a sensitivity that enables the detection of 10–100 pg of RNA from a target gene. Furthermore, this method has been shown to be highly precise, with intra- and interassay coefficients of variability at <2% and <3%, respectively. Recently, IHC analysis and qPCR have been developed and studied for the detection of nodal micrometastases in bladder cancer with promising results (Table 2[56–62]). Potential tumour markers that have been evaluated with these techniques include cytokeratin AE 1/3, cytokeratin 19 (CK19) and 20 (CK20), mucin 7 (MUC7), and uroplakin II (UPII), a urothelial-specific marker. The findings of these studies will be discussed in the context of their potential clinical application with the goal of encouraging further research into this field to improve accurate diagnostic staging for patients with bladder cancer.

Table 2.  Comparison of studies investigating the detection of nodal micrometastases in bladder cancer by novel IHC staining and RT-PCR techniques
StudyTechniqueBiomarkerResults
Cytokeratin
Yang et al. (1999) [56]IHC analysisKeratinKeratin detected in one lymph node that was negative by H&E
Jiang et al. (2001) [57]IHC analysisCK20100% of lymph nodes positive by H&E were positive for CK20.
Watts et al. (unpublished)IHC analysisCytokeratin AE1/3Cytokeratin AE1/3 detected in 8/20 lymph nodes in 3/13 cases that were negative by H&E
Kurahashi et al. (2005) [58]RT-PCRCK19CK19 detected in 49 lymph nodes from 10 patients vs 29 lymph nodes from six patients determined positive by H&E. There were 4/8 recurrences positive for CK19, two of which were negative by H&E
MUC7
Retz et al. (2004) [59]RT-PCRMUC7MUC7 detected in 5/5 nodes determined positive by H&E. MUC7 detected in 46/160 nodes determined negative by H&E.
UP II
Seraj et al. (2001) [60]RT-PCRUPIIUPII detected in 4/16 lymph nodes that were negative by H&E
Wu et al. (2005) [61]RT-PCRUPIIUPII detected in 6/66 LNs that were negative by H&E. A total of 4/8 patients with recurrence had positivity for UPII
Kurahashi et al. (2005) [58]RT-PCRUPIIUPII detected in 58/633 LNs from 10 patients that were negative by H&E. A total of 4/8 recurrences were positive for CK19, two of which were negative by H&E
Copp et al. (2006) [62]RT-PCRUPIIUPII detected in 24/46 lymph nodes vs13/46 nodes positive by H&E. A total of 11/33 LNs negative by H&E had positivity for UPII. A total of 11/24 LNs with positivity for UPII were negative by H&E. Overall recurrence rate at 1.84 years for patients with UPII- positive, H&E-negative LNs was 91% vs 5% in patients with UPII-negative LNs

CK20 AND CYTOKERATIN AE1/3

Cytokeratin is a non-specific epithelial cell marker. Both IHC analysis and RT-PCR for cytokeratin have been used to identify micrometastases in lymph nodes initially read as pathologically negative in other cancers [53,58]. In addition IHC staining for cytokeratin is routinely performed for identification of large volume metastasis, or to confirm the primary origin of a tumour as urothelial. Yang et al. [56] first looked at the use of cytokeratin IHC analysis to detect micrometastasis in nodes previously read as negative by standard methods in 19 patients with high grade carcinoma invading bladder muscle [56]. They found evidence of nodal involvement in one patient that was missed by routine pathological evaluation. Jiang et al. [57] evaluated pathologically positive lymph nodes from 26 patients for the expression of CK20. They found 100% concordance between pathological and IHC CK20 expression in the lymph nodes sampled. They went on to investigate the utility of IHC analysis for cytokeratin AE1/3, a pan marker of epithelial cells, in 13 patients who underwent radical cystectomy. A total of 121 lymph nodes were removed and determined to be pathologically negative. Three of 13 patients were found to harbour a total of eight positive nodes by IHC analysis (Fig. 1). This shows that IHC evaluation of lymph nodes by for cytokeratin AE1/3 may prove helpful in the identification of micrometastases missed by standard evaluation.

image

Figure 1. Metastatic urothelial cells identified by cytokeratin AE1/3 immunohistochemical analysis of lymph nodes as a cluster or single cell (inset).

Download figure to PowerPoint

MUC7

Mucins, high molecular weight glycoproteins, serve as a protective barrier between urothelial cells and the contents of the bladder lumen. MUC7, which is normally restricted to human salivary glands, has been expressed by carcinoma in situ and invasive carcinomas of bladder urothelial cells [59,63,64]. Retz et al. [59] examined 240 lymph nodes from 25 radical cystectomy samples by H&E and RT-PCR for MUC7. They found 100% concordance for MUC7 RNA in 5/5 lymph nodes positive by H&E. In addition, they found an additional 46/160 nodes positive for MUC7 that were pathologically negative. Among patients with superficial disease, 7/39 (18%) pathologically negative nodes were positive for MUC7, indicating potentially undiagnosed metastatic disease.

UPII

Uroplakin II is one of four (UP Ia, Ib, II, III) integral membrane proteins that form the unit membrane of human urothelium and is the most specific of the uroplakin group to urothelium. Its expression has been preserved in cells from TCC, including poorly differentiated tumours where protein expression often decreases [58,60–62,65–67]. As such, it has been proposed to serve as a molecular marker for detecting bladder cancer cells.

Seraj et al. [60] published the earliest study, which included 19 patients with stage pTa-T3N0M0 disease. Pelvic nodes and perivesical tissue were evaluated by RT-PCR for the presence of UPII mRNA. Among the patients with pathologically organ-confined tumours, 5/17 showed positivity for UPII in the perivesical tissue, and 1/7 showed positivity in pelvic lymph nodes. In addition, of patients with pathologically negative nodes, 13/24 were positive for UPII in the perivesical tissue and 4/16 in the pelvic lymph nodes. Furthermore, all patients with pathologically positive nodes were also positive for UPII by RT-PCR, suggesting that the sensitivity of this assay is at least equivalent to that of pathological evaluation. However, these findings probably underestimate the ability of this technique to identify positive lymph nodes because only one node per patient, deemed to be the largest or ‘most suspicious’ of harbouring metastatic disease, was evaluated. If a less subjective approach had been used, it is possible that other nodes might have been identified that contained metastases.

Wu et al. [61] also reported on the use of RT-PCR for detecting UPII in a study that assessed 82 pelvic lymph nodes from 21 patients with bladder cancer. UPII was detected in 15/16 nodes with pathologically confirmed tumour involvement. UPII was also noted in six nodes (10%) that were pathologically negative. In contrast to the study by Seraj et al. [60] all lymph nodes were evaluated for the presence of tumour cells, thus possibly limiting the number of false negative results.

A more recent paper by Copp et al. [62] studied 46 patients with stage cTa-T4N0M0 bladder cancer who underwent cystectomy and pelvic node dissection. The lymph nodes were first evaluated by standard pathological evaluation (H&E), followed by RT-PCR for the presence of UPII. They reported a 100% concordance between UPII and pathological assessment for node-positive and -negative disease. However, 33% of pathologically negative nodes were found to be positive for UPII, and 46% of nodes positive by RT-PCR had been designated pathologically negative. Furthermore, at a median follow-up time of 1.84 years, the overall recurrence rate for local, nodal or metastatic disease in patients with UPII RT-PCR positivity was 79% (19 patients) compared with only 5% in the one patient with UPII RT-PCR negative nodes (P < 0.001). On multivariate analysis, UPII RT-PCR positivity was found to be a significant predictor of recurrence (P < 0.001). Finally, with regard to nodal status, only one lymph node per patient was evaluated, half by RT-PCR and half by H&E, examining a 1 mm-wide cross-section of each half. As such, the sensitivity and specificity of each technique in reference to the other may not be as reliable.

These results suggest that the sensitivity and specificity for RT-PCR detection of UPII as an indicator of nodal disease are at least comparable to standard H&E. In addition, they show that RT-PCR may have a stronger positive predictive power for recurrence than standard pathological evaluation.

CK19 AND UPII

Kurahashi et al. [58] used qPCR for the detection of both CK19 and UP II in 40 patients with carcinoma invading the bladder muscle at cystectomy with a median (range) follow-up of 22 (7–36) months. A total of 760 lymph nodes were examined histologically and by qPCR. The authors used the mean relative mRNA value + 2 sd as a threshold for determination of nodal CK19 or UPII mRNA positivity. Twenty-nine nodes from six patients were positive by pathological assessment. Of these 29 nodes, CK19 and UPII expression were confirmed in 27 nodes from five patients and 29 nodes from all six patients, respectively. However, an additional seven and 11 histologically negative nodes were positive for CK19 and UPII, respectively. Among the remaining 34 patients with 633 pathologically negative nodes, positivity for CK19 RNA and UPII mRNA was detected in 13 nodes from five patients and in 58 nodes from 11 patients, respectively. Combining these latter findings, 12 patients were found to have micrometastatic disease that was missed by standard technique.

Follow-up data revealed that half (4/8) of the patients with a subsequent cancer recurrence had positive nodes by qPCR, two of whom had been found to be pathologically negative by H&E. Analysis of recurrence rates showed that those patients with evidence of nodal involvement – either pathological or by qPCR analysis – had significantly lower cause-specific survival rates than those with no evidence of lymph node involvement (P < 0.005). However, the presence of nodal micrometastases by qPCR was not an independent predictor for cause-specific survival on multivariate analysis.

FUTURE DIRECTIONS AND RESEARCH AGENDA

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CURRENT STANDARDS
  5. DETECTION OF MICROMETASTATIC NODAL DISEASE
  6. FUTURE DIRECTIONS AND RESEARCH AGENDA
  7. CONCLUSION
  8. CONFLICT OF INTEREST
  9. REFERENCES

High grade bladder cancer remains a lethal disease. Little progress has been made over the past decades towards a significant impact on patient survival. It is becoming clear that this disease may best be tackled using multiple approaches. Currently, the decision to use adjunct therapies remains dependent on the proper identification of disease stage. Considerable research has shown that nodal involvement is the strongest independent prognostic variable for patient outcomes, meaning that identification of this is of paramount importance. Indeed many studies have advocated the use of an extended PLND to enhance the detection and removal of nodal metastatic disease. The current standard for lymph node dissection and pathological interpretation by formalin fixation and H&E staining clearly understages a significant number of patients, as suggested by the high recurrence rates for those with organ-confined, node-negative disease. Furthermore, solid tumour recurrence is almost uniformly fatal.

In the present review, the sensitivity of IHC analysis and RT-PCR analyses for all tumour markers across the studies ranged from 93.1 to 100%. Specifically, the sensitivity of IHC analysis for CK20 was 100%[57]. The sensitivity of PCR for UPII ranged from 93.8 [60] to 100%[58,60,62], and from 93.1 to 100% for CK19 and MUC7, respectively [58,59]. As such, these techniques show sensitivities equal to or greater than that of H&E evaluation of lymph nodes for tumour involvement.

There is a range of potential clinical applications for IHC or PCR analysis in enhancing the detection of nodal metastases. At a basic level, they could be used in conjunction with standard H&E evaluation in order to enhance sensitivity for detecting nodal disease. Given the similar, if not higher, sensitivity of both techniques compared with H&E, it seems possible that they could replace H&E staining as the standard method of postoperative lymph node evaluation. Several studies in the present review revealed that IHC analysis and qPCR detected nodal metastases missed by H&E, but only a few studies included follow-up data [58,61,62]. Larger studies using these techniques in a prospective manner with longer follow-up are needed to determine their efficacy on patient outcomes. While it has been historically difficult to accrue patients in large bladder cancer studies, these are necessary to advance the management of this disease. Multicentre studies using standardized techniques would hold the most promise in elucidating the value of these methods.

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CURRENT STANDARDS
  5. DETECTION OF MICROMETASTATIC NODAL DISEASE
  6. FUTURE DIRECTIONS AND RESEARCH AGENDA
  7. CONCLUSION
  8. CONFLICT OF INTEREST
  9. REFERENCES

In summary, bladder cancer is a common malignancy, representing the fourth and eighth most frequent solid tumour in men and women, respectively. Proper identification and pathological examination of lymph nodes is essential for the prognosis and staging of patients with this disease. Many studies have advocated an extended PLND to improve the detection of metastatic disease and improve patient survival. However, an exact template for the most efficacious lymph node dissection remains to be defined. Based on current standards, understaging continues to be a problem and newer techniques for enhancing the detection of nodal metastases have therefore been investigated. The use of IHC and PCR analyses for detecting nodal involvement have yielded promising results regarding the potential to stage more accurately patients with bladder cancer. Additional research is needed to determine the ideal template for the extent of lymph node dissection at cystectomy but equally, if not more importantly, to identify the most sensitive tumour marker for the detection of nodal disease, as well as its ability to predict outcomes with regard to need for adjunct therapy, tumour recurrence and ultimately patient survival.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. CURRENT STANDARDS
  5. DETECTION OF MICROMETASTATIC NODAL DISEASE
  6. FUTURE DIRECTIONS AND RESEARCH AGENDA
  7. CONCLUSION
  8. CONFLICT OF INTEREST
  9. REFERENCES