Natural history of surgically treated bladder carcinoma with extravesical tumor extension

Authors

  • Marcus L. Quek M.D.,

    Corresponding author
    1. Department of Urology, Kenneth Norris Jr. Comprehensive Cancer Center, University of Southern California Keck School of Medicine, Los Angeles, California
    • Department of Urology, Kenneth Norris Jr. Comprehensive Cancer Center, University of Southern California Keck School of Medicine, MS#74, 1441 Eastlake Avenue, Suite 7416, Los Angeles, CA
    Search for more papers by this author
    • Fax: (323) 865-0120

  • John P. Stein M.D.,

    1. Department of Urology, Kenneth Norris Jr. Comprehensive Cancer Center, University of Southern California Keck School of Medicine, Los Angeles, California
    Search for more papers by this author
  • Peter E. Clark M.D.,

    1. Department of Urology, Kenneth Norris Jr. Comprehensive Cancer Center, University of Southern California Keck School of Medicine, Los Angeles, California
    Search for more papers by this author
  • Siamak Daneshmand M.D.,

    1. Department of Urology, Kenneth Norris Jr. Comprehensive Cancer Center, University of Southern California Keck School of Medicine, Los Angeles, California
    Search for more papers by this author
  • Gus Miranda,

    1. Department of Urology, Kenneth Norris Jr. Comprehensive Cancer Center, University of Southern California Keck School of Medicine, Los Angeles, California
    Search for more papers by this author
  • Jie Cai M.B., M.S.,

    1. Department of Preventive Medicine, Kenneth Norris Jr. Comprehensive Cancer Center, University of Southern California Keck School of Medicine, Los Angeles, California
    Search for more papers by this author
  • Susan Groshen Ph.D.,

    1. Department of Preventive Medicine, Kenneth Norris Jr. Comprehensive Cancer Center, University of Southern California Keck School of Medicine, Los Angeles, California
    Search for more papers by this author
  • Gary Lieskovsky M.D.,

    1. Department of Urology, Kenneth Norris Jr. Comprehensive Cancer Center, University of Southern California Keck School of Medicine, Los Angeles, California
    Search for more papers by this author
  • David I. Quinn M.D., Ph.D.,

    1. Department of Medicine, Kenneth Norris Jr. Comprehensive Cancer Center, University of Southern California Keck School of Medicine, Los Angeles, California
    Search for more papers by this author
  • Derek Raghavan M.D., Ph.D.,

    1. Department of Medicine, Kenneth Norris Jr. Comprehensive Cancer Center, University of Southern California Keck School of Medicine, Los Angeles, California
    Search for more papers by this author
  • Donald G. Skinner M.D.

    1. Department of Urology, Kenneth Norris Jr. Comprehensive Cancer Center, University of Southern California Keck School of Medicine, Los Angeles, California
    Search for more papers by this author

Abstract

BACKGROUND

The current TNM classification for bladder carcinoma stratifies extravesical extension into microscopic (pT3a) and macroscopic (pT3b) tumor involvement. The authors evaluated the outcomes of patients with pT3a and pT3b disease after radical cystectomy.

METHODS

Patients (n = 129) with transitional cell carcinoma of the bladder treated with radical cystectomy alone demonstrated pathologic extravesical tumor extension: 37 (29%) had pT3a disease and 92 (71%) had pT3b disease. No patient received any adjuvant therapy. With a median follow-up of 13.6 years, the presence of lymph node involvement, margin positivity, local (pelvic) and distant disease recurrence, and clinical outcomes were determined.

RESULTS

Of the 129 patients, 43 (33%) had lymph node tumor involvement: 13 of 37 patients with pT3a disease (35%) and 30 of 92 patients with pT3b disease (33%). The 10-year recurrence-free and overall survival for the entire group was 54% and 20%, respectively. No statistical difference between pT3a and pT3b disease was observed with regard to recurrence-free (P = 0.54) and overall (P = 0.66) survival. Lymph node involvement was predictive of a significantly worse 10-year recurrence-free survival (32%) compared with lymph node-negative disease (60%; P = 0.003). Local disease recurrence was reported to occur in 12 patients (9%), whereas 37 patients (29%) were reported to develop distant metastases. Among those who had disease recurrence, the type of disease recurrence (local or distant) was not found to be associated with tumor stage (pT3a vs, pT3b, P = 0.47).

CONCLUSIONS

This cohort of surgically managed patients provided insight into the long-term natural history of pathologically confirmed extravesical bladder carcinoma after radical cystectomy. There was no important difference in the incidence of lymph node involvement, survival rates, and disease recurrence rates between patients with microscopic and macroscopic extravesical extension. Adjuvant protocols should be undertaken for these high-risk patients to further improve on these clinical outcomes. Cancer 2003;98:955–61. © 2003 American Cancer Society.

DOI 10.1002/cncr.11569

Radical cystectomy is a standard treatment for high-grade invasive bladder carcinoma. The long-term clinical outcomes after cystectomy are related directly to pathologic stage and lymph node status.1 Organ-confined bladder carcinoma managed with radical cystectomy and pelvic lymphadenectomy is associated with better survival rates compared with extravesical disease, whereas lymph node involvement predicts the worst survival rates.1–4

The 1997 TNM staging system for bladder carcinoma stratifies extravesical involvement into microscopic (pT3a) and macroscopic (pT3b) extravesical tumor extension.5 The pathologic distinction between microscopic and macroscopic extension of the primary tumor and its clinical significance remains to be determined. In the current study, the natural history of 129 patients with bladder carcinoma treated uniformly with radical cystectomy and pelvic lymphadenectomy without adjuvant therapy was evaluated to determine the clinical significance of microscopic and macroscopic extravesical tumor extension.

MATERIALS AND METHODS

Between November 1971 and December 1997, 1054 patients with primary transitional cell carcinoma (TCC) of the bladder underwent a uniform radical cystectomy, bilateral pelvic iliac lymphadenectomy, and urinary diversion at the University of Southern California with the intent to cure. The outcomes of this cohort were described previously.1 Of this group, 247 patients (23%) demonstrated pathologic extravesical extension of the primary tumor. The medical records and pathology reports of these 247 patients were reviewed under an institutional review board-approved protocol. Patients were subdivided into either those with microscopic extravesical tumor involvement (pT3a) or macroscopic tumor extension into the perivesical tissue (pT3b) in accordance with the 1997 TNM staging system.5 Histologic grading was performed according to the method of Bergkvist et al.6 The detailed pathologic and medical records of 11 patients were either not available for review or were equivocal in their description of the extent of extravesical involvement. These patients therefore were excluded from analysis. Adjuvant therapies, including any radiation and/or chemotherapy, were administered to 107 patients and these patients also were excluded from analysis to avoid contaminating the natural history data set. The remaining 129 patients with extravesical disease managed with radical cystectomy alone comprise the sample in the current study.

The incidence of lymph node tumor involvement, margin positivity, clinical outcomes, and local and distant disease recurrences was evaluated. Data were analyzed according to the pathologic stage of the primary bladder tumor (pT3a vs. pT3b), lymph node status (positive vs. negative), and type of urinary diversion performed (orthotopic, continent cutaneous, incontinent conduit, or anal continent diversion). Clinical outcomes included recurrence-free and overall survival. Time to clinical disease recurrence, or recurrence-free survival, was calculated as the time from cystectomy to the date of the first documented clinical disease recurrence or until last follow-up if the patient had not experienced a clinical disease recurrence. Patients who died before clinical disease recurrence were censored at the time of death. Overall survival was calculated similarly as the time from cystectomy to the date of death. All deaths, regardless of the cause, were counted as events. Patients who were still alive were censored at the date of last contact.

Bladder carcinoma recurrences were classified as local (pelvic) or distant (metastatic). Local disease recurrences were defined as any disease occurring within the soft tissue field of exenteration. Distant disease recurrences were defined as any disease occurring outside the pelvis. Patients with concomitant local and distant disease were classified as having distant disease recurrences.

Kaplan-Meier plots7 were generated to estimate the 5-year and 10-year recurrence-free and overall survival rates. Standard errors were based on the Greenwood formula.8 To estimate the chance of local or distant disease recurrence, the method of competing risks was used with local and distant disease recurrence or death as the included competing failures.9 The log-rank test (overall8 and stratified10) was used to compare survival and time to disease recurrence among subgroups of patients. The Pearson chi-square test for associations or the Mantel-Haenszel test for trend11 was used to evaluate the association between clinical variables. All P values are two sided.

The specific form of urinary diversion after radical cystectomy depended primarily on the era in which the urinary reconstruction was performed. Factors such as manual dexterity, mental competency, renal function, and associated comorbidities also influenced the type of urinary diversion constructed. The form of urinary diversion stratified by pathologic stage is presented in Table 1.

Table 1. Demographics of 129 Patients with Extravesical Extension of Bladder Carcinoma Treated with Radical Cystectomy Between 1971 and 1997
Pathologic stagepT3a (%)pT3b (%) All pT3 (%)
No. of patients3792 129
  • a

    Statistical comparison between pT3a and pT3b.

Age at surgery (yrs)  P = 0.28a 
 Median70 y68 y 68 y
 Range47–8631–85 31–86
Gender  P = 0.87a 
 Male26 (70)66 (72)  92 (71)
 Female11 (30)26 (28)  37 (29)
Median follow-up  P = 0.59a 
 9.4 years13.7 years 13.6 years
Histologic grade  P = 0.86a 
 Low 1 (3) 2 (2)   3 (2)
 High36 (97)90 (98) 126 (98)
Lymph node status  P = 0.78a 
 Negative24 (65)62 (67)  86 (67)
 Positive13 (35)30 (33)  43 (33)
Surgical margin status  P = 0.86a 
 Negative36 (97)90 (98) 126 (98)
 Positive 1 (3) 2 (2)   3 (2)
Type of diversion  P = 0.19a 
 Orthotopic20 (54)33 (36)  53 (41)
  Kock ileal reservoir1930 49
  T-pouch03 3
  Studer ileal reservoir10 1
 Continent cutaneous 8 (22)41 (45)  49 (38)
  Kock pouch839 47
  Double T-pouch01 1
  W-ileal reservoir01 1
 Conduit (incontinent) 8 (22)14 (15)  22 (17)
  Ileal conduit814 22
 Anal continent 1 (3) 4 (4)   5 (4)
  Kock rectum01 1
  Ureterosigmoidostomy13 4

RESULTS

A total of 129 patients (92 males [71%] and 37 females) had pathologically confirmed TCC of the bladder with extravesical tumor extension (pT3 disease; Table 1). The median age of the cohort was 68 years (range, 31–86 years). Of the 129 patients, 37 (29%) had microscopic tumor involvement of the perivesical tissues (pT3a) and 92 (71%) had macroscopic extravesical disease (pT3b). Nearly all patients had high grade disease (98%). Overall, 43 patients (33%) had lymph node tumor involvement, including 13 (35%) with pT3a and 30 (33%) with pT3b tumors. No significant difference in the incidence of lymph node positivity was observed when patients were stratified by stage (pT3a vs. pT3b; P = 0.78).

Pathologic evaluation of the cystectomy specimen revealed involvement of the surgical margin with tumor in three patients (2%). Therefore, no difference was noted between patients with pT3a and pT3b tumors in terms of surgical margin status.

At the time of analysis, all patients who were still alive had been followed for at least 2 years. With a median follow-up of 13.6 years (range, 2–17 years), the 10-year recurrence-free survival rate for the 129 patients was 54% (Table 2, Fig. 1A). The overall survival rate for the entire cohort was 20% at 10 years (Table 2, Fig. 1B). When categorized by microscopic (pT3a) and macroscopic (pT3b) extravesical extension, the 10-year recurrence-free survival rates were 56% and 53% for pT3a and pT3b disease, respectively (P = 0.54; Table 2, Fig. 2A). In the current study, all disease recurrences occurred within 4 years after cystectomy. Furthermore, the overall survival at 10 years was not found to be significantly different between pT3a and pT3b tumors, although the absence of difference may simply reflect the relatively small numbers in these subsets of data (Fig. 2B; also see Figs. 5B,6B).

Table 2. Overall and Recurrence-Free 5 and 10-Year Survival of Patients with Extravesical Extension of Bladder Caccinoma treated with Radical Cystectomya
FactorsNo. of patientsProbability of surviving and remaining recurrence-free (P ± SE)
Overall survival (yrs)Recurrence-free survival (yrs)
510 510 
  • LN: lymph node.

  • a

    Probabilities of overall survival and recurrence-free survival were based on Kaplan–Meier estimates. Plus-minus values are estimates of the standard error (SE) calculated using the Greenwood formula. Two-sided P values were based on the stratified log-rank test.

Stage       
 All pT31290.34 ± 0.040.20 ± 0.04 0.54 ± 0.050.54 ± 0.05 
 pT3a370.31 ± 0.080.26 ± 0.08 0.56 ± 0.090.56 ± 0.09 
    P = 0.66  P = 0.54
 pT3b920.36 ± 0.050.18 ± 0.05 0.53 ± 0.060.53 ± 0.06 
LN status       
 All LN (−)860.47 ± 0.050.31 ± 0.05 0.60 ± 0.050.60 ± 0.05 
    P = 0.002  P = 0.003
 All LN (+)430.23 ± 0.040.10 ± 0.04 0.32 ± 0.050.32 ± 0.05 
 pT3a/LN (−)240.40 ± 0.100.32 ± 0.11 0.72 ± 0.100.72 ± 0.10 
    P = 0.82  P = 0.30
 pT3b/LN (−)620.42 ± 0.060.24 ± 0.06 0.60 ± 0.070.60 ± 0.07 
 pT3a/LN (+)130.15 ± 0.100.15 ± 0.10 0.24 ± 0.140.24 ± 0.014 
    P = 0.68  P = 0.98
 pT3b/LN (+)300.22 ± 0.080.05 ± 0.05 0.38 ± 0.110.38 ± 0.11 
   StratifiedP = 0.67 StratifiedP = 0.46
Type of urinary diversion       
 pT3a/orthotopic200.32 ± 0.110.32 ± 0.11 0.61 ± 0.120.61 ± 0.12 
 pT3b/orthotopic330.33 ± 0.080.16 ± 0.08 0.58 ± 0.100.58 ± 0.10 
 pT3a/cutaneous90.33 ± 0.160.33 ± 0.16 0.57 ± 0.190.57 ± 0.19 
 pT3b/cutaneous450.44 ± 0.080.25 ± 0.07 0.49 ± 0.080.49 ± 0.08 
 pT3a/incontinent80.25 ± 0.150.13 ± 0.12 0.42 ± 0.210.42 ± 0.21 
 pT3b/incontinent140.29 ± 0.120.07 ± 0.07 0.54 ± 0.160.54 ± 0.16 
   StratifiedP = 0.45 StratifiedP = 0.65
Figure 1.

(A) Recurrence-free survival of patients with pT3 bladder carcinoma who were treated with radical cystectomy. (B) Overall survival of patients with pT3 bladder carcinoma who were treated with radical cystectomy.

Figure 2.

(A) Recurrence-free survival of patients with pT3a and pT3b bladder carcinoma who were treated with radical cystectomy. (B) Overall survival of patients with pT3a and pT3b bladder carcinoma who were treated with radical cystectomy.

Figure 3.

(A) Recurrence-free survival of patients with microscopic extravesical extension (pT3a) bladder carcinoma who were treated with radical cystectomy, stratified by lymph node status. (B) Overall survival of patients with microscopic extravesical extension (pT3a) bladder carcinoma who were treated with radical cystectomy, stratified by lymph node status.

Figure 4.

(A) Recurrence-free survival of patients with macroscopic extravesical extension (pT3b) bladder carcinoma who were treated with radical cystectomy, stratified by lymph node status. (B) Overall survival of patients with macroscopic extravesical extension (pT3b) bladder carcinoma who were treated with radical cystectomy, stratified by lymph node status.

Although no difference was observed in the incidence of lymph node involvement when patients were stratified by microscopic or macroscopic extravesical extension, lymph node status was found to be a significant prognostic factor (Table 2). The recurrence-free survival for patients with lymph node tumor involvement and either pT3a or pT3b disease was significantly lower compared with patients without positive lymph nodes (Figs. 3A,4A). No significant difference in recurrence-free or overall survival rates was noted when comparing patients with pT3a and pT3b disease and controlling for lymph node status (Figs. 5,6).

Figure 5.

(A) Recurrence-free survival of bladder carcinoma patients with lymph node-negative extravesical disease stratified by the degree of extravesical extension (pT3a vs. pT3b). (B) Overall survival of bladder carcinoma patients with lymph node-negative extravesical disease stratified by the degree of extravesical extension (pT3a vs. pT3b).

Figure 6.

(A) Recurrence-free survival of bladder carcinoma patients with lymph node-positive extravesical disease stratified by the degree of extravesical extension (pT3a vs. pT3b). (B) Overall survival of bladder carcinoma patients with lymph node-positive extravesical disease stratified by the degree of extravesical extension (pT3a vs. pT3b).

Of the 129 patients, 49 (38%) developed recurrent disease during follow-up (Table 3). Local (pelvic) disease recurrence alone occurred in 12 patients and distant metastatic disease developed in 37 patients. Of the 37 patients with pT3a disease, 14 developed disease recurrence (2 locally and 12 at a distant site). Similarly, 35 of the 92 patients with pT3b tumor experienced disease recurrence (10 locally and 25 distally) (Table 3). When the type of disease recurrence was evaluated in only those patients who experienced clinical disease recurrences, two patterns were suggested, although neither was statistically significant. There were more local disease recurrences in patients with pT3b tumors compared with patients with pT3a tumors and there were more distant disease recurrences with lymph node-positive disease.

Table 3. Local and Distant Recurrences after Radical Cystectomy for Bladder Carcinoma with Extravesical Tumor Extension
StageNo. of patientsNo. of patients with local recurrence (%)No. of patients with distant recurrence (%)P valuea
  • LN: lymph node.

  • a

    P values are based on the Fisher exact test and include only patients (n = 49) who had disease recurrence.

All pT34912 (24)37 (76) 
pT3a14 2 (14)12 (86) 
    0.47
pT3b3510 (29)25 (71) 
LN Status    
 All pT3/LN(−)27 8 (30)19 (70) 
    0.51
 All pT3/LN(+)22 4 (18)18 (82) 
 pT3a/LN(−)6 1 (17) 5 (83) 
    0.63
 pT3b/LN(−)21 7 (33)14 (67) 
 pT3a/LN(+)8 1 (12) 7 (88) 
    1.00
 pT3b/LN(+)14 3 (21)11 (79) 

The 5-year and 10-year overall and recurrence-free survival rates are reported for patients with pT3a and pT3b disease stratified by the form of urinary diversion (Table 2). The impact of the urinary reconstruction was difficult to interpret given the nonrandomized nature of selecting a patient for surgery. Adjustment for the form of urinary diversion revealed no significant difference between the outcomes for those patients with pT3a versus those with pT3b disease with regard to recurrence-free survival (P = 0.65) or overall survival (P = 0.45).

DISCUSSION

Radical cystectomy remains a standard treatment for invasive bladder carcinoma.1 Over a 26-year period, 129 patients uniformly treated with radical cystectomy and en bloc bilateral pelvic iliac lymphadenectomy alone were found to have a primary bladder tumor involving the perivesical tissue (pT3 disease). This group of patients did not receive adjuvant therapies and, therefore, they represent a cohort that may provide a better understanding of the natural history of surgically treated extravesical TCC. Although the most recent TNM staging system classifies extravesical tumor involvement as either microscopic (pT3a) or macroscopic (pT3b) disease, the analysis of these 129 patients suggests that this distinction may not be necessary. In the current series, no significant differences in recurrence-free and overall survival between microscopic (pT3a) and macroscopic (pT3b) extravesical tumor extension were observed. As expected, lymph node involvement was associated with worse recurrence-free and overall survival for patients with extravesical extension.

The analysis of 129 patients that were treated uniformly with surgery alone suggests little difference between patients with microscopic or macroscopic extravesical tumor extension overall and when controlling for lymph node status. The incidence of lymph node involvement was nearly identical for patients with pT3a and pT3b disease. Patients with pT3aN0 tumors had recurrence-free and overall survival rates that were statistically similar to those for patients with pT3bN0 tumors. In addition, patients with pT3a/lymph node-positive disease had similar clinical outcomes as patients with pT3b/lymph node-positive disease.

The overall 5-year survival rates of 31% and 36% observed for patients with pT3a and pT3b disease, respectively, compare favorably with other contemporary series.2–4 Dalbagni et al.2 reported a 5-year overall survival rate of 26% in 100 patients with pT3 disease. They demonstrated similar 5-year survival rates between the 32 patients with pT3a disease (23%) and the 69 patients with pT3b disease (28%). Pagano et al.4 reported a 5-year survival rate of 15% for patients with extravesical extension. Cheng et al.3 reviewed the Mayo Clinic experience in which 42 patients with pT3a disease and 14 patients with pT3b disease were treated with radical cystectomy and bilateral pelvic lymphadenectomy. Patients were grouped into a single pT3 category with a reported 10-year cancer-specific survival rate of 50% and an overall 10-year survival rate of 28%. These findings are similar to the findings of the current study, namely, a recurrence-free survival rate of 54% and an overall survival rate of 20% for all pT3 patients at 10 years.

The current study has several weaknesses. First, this is a retrospective analysis with a small sample of patients that lacks prospective validation. Second, although this is a so-called “pure” group of patients who have not received any adjuvant therapy, selection bias may exist. We do not believe that there were major differences in baseline characteristics (age, gender, histologic grade, lymph node status, or surgical margin status) between patients with pT3 tumors who received adjuvant therapy and patients who did not receive adjuvant therapy. However, we cannot eliminate the possibility that other unarticulated characteristics were used to preselect good prognosis patients with pT3b disease who did not receive adjuvant therapy or to preselect poor prognosis patients with pT3a disease who did receive adjuvant therapy. This preselection could either attenuate a real difference between pT3a and pT3b tumors or it could create an apparent difference when none exists. Third, the precise definition of microscopic and macroscopic tumor extension may be difficult to determine in a retrospective analysis. Despite our best efforts, 11 patients had to be excluded from analysis because this distinction could not be made.

This analysis confirms that radical cystectomy provides excellent local control even in patients with locally extensive extravesical disease. Because the clinical outcomes for patients with pT3a and pT3b disease were similar, the current study suggests that even if some differences did exist, the distinction between microscopic and macroscopic extravesical extension may not be clinically important. In examining this select cohort of surgically managed patients who did not receive any form of adjuvant therapy, we also provide insight into the long-term treated natural history of pathologically confirmed extravesical bladder carcinoma. Adjuvant protocols for these high-risk patients may be considered to further improve clinical outcomes.

Ancillary