SEARCH

SEARCH BY CITATION

Keywords:

  • bladder cancer;
  • metastasis;
  • radical cystectomy;
  • survival

Abstract

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

OBJECTIVES

To assess the clinical outcome of patients with bladder cancer who underwent attempted radical cystectomy (RC) with curative intent, but whose procedures were aborted due to intraoperative findings of metastatic disease, as the presence of metastatic disease at RC is associated with a poor prognosis and there are no data on the optimum management strategy in this situation.

PATIENTS AND METHODS

In all, 248 consecutive patients with bladder cancer had attempted RC at one academic institution between 1994 and 2003. We retrospectively reviewed the records of 35 patients who had an aborted RC due to intraoperative findings of metastatic disease. The pathological characteristics, time to recurrence, overall survival, disease-specific survival, and suitability for adjuvant or salvage therapies were examined.

RESULTS

Of the 35 patients who had an aborted RC for metastatic disease, 21 (60%) died from the disease within the study period (median time to cancer-specific death 26.4 months), 11 (31%) are alive with evidence of persistent disease or progression, and three (9%) are alive with no evidence of recurrence or progression. Seven patients had a salvage RC after successful adjuvant treatment, of whom three died from recurrent disease (at a mean of 46.5 months after initial exploration, 31.9 months after salvage RC), one is alive with bladder cancer recurrence to the rectum 10 months after salvage cystectomy, and three have no evidence of disease progression at a mean of 10 months after salvage RC.

CONCLUSIONS

The prognosis of patients who undergo an aborted attempt at curative RC due to intraoperative findings of metastatic disease is poor. Although a few patients might subsequently have salvage RC, many of these patients still have poor outcomes even if adjuvant treatments are used. When metastatic disease is discovered at RC, completing the cystectomy should be considered, although further studies are needed to show a clinical benefit.


Abbreviations
RC

radical cystectomy

(P)LND

(pelvic) lymph node dissection

CIS

carcinoma in situ

MVAC

methotrexate, vinblastine, doxorubicin, and cisplatin.

INTRODUCTION

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

Invasive bladder cancer continues to be a significant cause of death in the USA; in the 2007 alone, ≈67 160 new cases of bladder cancer will be diagnosed and ≈13 750 people will die from the disease in the USA [1]. Radical cystectomy (RC) and pelvic lymph node dissection (PLND) is the standard treatment for patients with muscle-invasive bladder cancer. There are well-defined predictors of bladder cancer-specific survival after RC, including pathological stage and lymph node status [2–7].

A significant proportion of patients with clinically localized disease will be upstaged to locally advanced and metastatic disease at the time of RC [8]. An even smaller proportion of patients will have unanticipated findings of bulky metastatic disease at the time of a planned curative RC. The clinical outcome and optimum management of patients with the intraoperative finding of metastases at RC is unclear. Thus we conducted a single-institution, retrospective analysis to evaluate the clinical outcomes of patients with bladder cancer who had an aborted attempt at curative RC due to intraoperative findings of metastasis, to further define optimum management strategies in this subset of patients.

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

We retrospectively reviewed the clinical and pathological data on 248 consecutive patients who had an attempted RC and bilateral PLND with curative intent at one institution between July 1994 and August 2003. The clinical indication for RC in all patients was either muscle-invasive bladder cancer or high-grade T1 or carcinoma in situ (CIS) refractory to intravesical therapy. The preoperative evaluation for clinical staging included serum chemistry tests, excretory urogram, chest X-ray, CT of the abdomen and pelvis and/or MRI. Bone scans were taken if patients had abnormal liver function studies or elevated alkaline phosphatase levels. All patients had had cystoscopy and examination under anaesthesia. We retrospectively reviewed the records of 35 patients who had an aborted attempt at RC due to intraoperative findings of unexpected metastatic disease. Clinical and pathological information was obtained from individual patient charts, medical records and approved cancer registries. No patient had known metastatic disease at the time of RC. Institutional Review Board approval was obtained before starting the study.

All RC specimens were examined by a dedicated genitourinary pathologist; the 2002 TNM classification system was used for pathological staging, and the WHO classification for pathological grading. For the salvage RC specimens, multiple sections from the tumour, the bladder wall, and mucosa adjacent to and distant from the tumour, along with the ureters and regional lymph nodes, were evaluated.

After recovering from the laparotomy patients were referred to a medical oncologist to discuss potential systemic treatment options. After surgery, patients were scheduled to have a regular physical examination, routine serum chemistry tests and imaging studies to monitor for disease progression. Bone scintigraphy, CT and other imaging studies were used when clinically indicated. When patients died the cause of death was determined by the treating physician, chart review corroborated by death certificates, or by death certificates alone. Most patients who were identified as having died from bladder cancer had progressive, widely disseminated, and often highly symptomatic metastases at the time of death.

Postoperative chemotherapy was given to 30 (86%) of the patients who had an aborted attempt at RC; the status of chemotherapy was unknown in four (11%). One patient is known to have not received adjuvant chemotherapy before death from bladder cancer. Chemotherapy regimens included: carboplatin, methotrexate and vinblastine, gemcitabine/cisplatin, methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC), and taxol/carboplatin. Six of these patients (17%) also received radiation therapy after surgery, four (11%) did not, whereas the radiotherapy status was uncertain in 25 (72%).

Fisher’s exact test and the chi-square test were used to evaluate the association between categorical variables. Differences in variables with a continuous distribution between dichotomous categories were assessed using the Mann–Whitney U-test. Differences in variables with a continuous distribution across ranked categories were assessed using the Kruskal–Wallis nonparametric analysis of variance. The Kaplan-Meier method was used to estimate the survival distribution, and differences were assessed with the log-rank statistic.

RESULTS

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

The clinical and pathological characteristics of the 35 patients who had an aborted attempt at curative RC are shown in Table 1. All patients had invasive, high-grade or BCG-refractory superficial urothelial carcinoma by transurethral resection, and had pathological confirmation of metastasis at surgery. One patient also had elements of primary adenocarcinoma and another had a component of small cell carcinoma in the transurethral resection specimen, with pathological confirmation of these elements in the LND at the time of the aborted RC. Five patients had had neoadjuvant chemotherapy and two of these had also had preoperative radiotherapy for locally advanced disease. Three patients who had had neoadjuvant chemotherapy eventually died from bladder cancer with a mean (range) survival of 7.2 (4.8–9.7) months. The mean time from diagnosis of muscle invasion to RC in the 25 patients with clinical muscle-invasive disease was 4.3 months. The mean time from aborted RC to administration of adjuvant chemotherapy was 2.1 (0.5–6.8) months.

Table 1.  Clinical and pathological information for 35 patients with bladder cancer who had an aborted RC between 1989 and 2004 for intraoperative findings of metastasis
VariableMean (range) or n (%)
  • *

    18 patients had elements of CIS on biopsy, 11 of these had intravesical treatment;

  • †two patients also had preoperative radiotherapy;

  • ‡31 patients had PLND; LND was deferred in four due to local invasion or obvious metastases in other sites.

Age, years65 (43–86)
Follow-up, months18.5 (1.4–106)
Male27 (77)
Female8 (23)
Race
 White31 (88)
 Black2 (6)
 Other2 (6)
Clinical stage
 Tis-only1 (3)*
 T19 (26)
 T225 (71)
Neoadjuvant chemotherapy5 (14)
Lymph node involvement
 Grossly positive32 (91)
 Only site of metastasis23 (66)
T4 disease during RC8 (23)
Adjuvant chemotherapy30
Subsequent salvage RC7 (20)
Died from disease21 (60)
Alive with persistent disease11 (31)
Alive with no evidence of disease3 (9)

At attempted RC, gross metastases were discovered in all patients and confirmed by biopsy or LND. Gross lymph node involvement was present in 32 (91%) patients and this was the only finding of metastasis in 23 (66%). The remaining patients had evidence of metastases in other locations, including liver, peritoneum, mesentery and retroperitoneum. Eight patients (23%) had evidence of T4 disease during surgery, with areas of involvement including the pelvic side-wall, pubis, rectum, and obturator foramen; LND was deferred in four of these patients. For the 31 patients who had PLND, the mean (range) number of lymph nodes retrieved was 9.7 (1–36), with a mean number of positive nodes of 2.9 (1–12), and a mean percentage of positive lymph nodes of 45.7 (6.7–100)%.

Of the 35 patients who had aborted RC for metastatic disease, 21 (60%) died from disease within the study period (median time to cancer-specific death 26.4 months), 11 (31%) are alive with evidence of persistent disease or progression, and three (9%) are alive with no evidence of recurrence or progression. Of these three patients, two had salvage RC for pT3a and pTis disease, with survival after initial laparotomy of 13.3 and 15.2 months, respectively, and survival after salvage RC of 5.2 and 9.1 months, respectively. The third patient had clinical T1 disease with grossly positive nodes and had LND at the time of initial laparotomy with subsequent chemotherapy; he is still alive with no evidence of disease at 3.5 years after surgery.

Kaplan-Meier plots for overall survival from the time of the aborted attempt at RC are shown in Fig. 1a; the median survival for all patients was 18 months. The percentage of patients surviving at 12, 24 and 36 months was 67%, 43%, and 18%, respectively. Figure 1b shows that patients who responded to chemotherapy and subsequently had salvage RC had better survival than those who did not have salvage RC (median survival 38.5 vs 16.2 months, log-rank P = 0.04).

image

Figure 1. Overall mortality rates after an aborted attempt at RC due to intraoperative findings of metastases: a, from the time of laparotomy; and b, stratified by whether or not patients had salvage RC.

Download figure to PowerPoint

Seven patients had salvage RC after successful adjuvant treatments, with a mean (range) follow-up of 30.5 (13.3–67.7) months and a mean time from initial laparotomy to salvage RC of 11.1 (4.8–26.4) months. No patients who had salvage RC had received chemo- or radiotherapy before the initial aborted laparotomy. The demographics and outcomes of these seven patients are shown in Table 2; three who had salvage RC died from recurrent disease, with a mean time after initial exploration of 46.5 (38.3–62.7) months and a mean time after salvage RC of 31.9 (12.2–57.9) months. Of the three patients who died, one had cT2 disease, microscopic lymph node involvement, and an abdominal mass at initial aborted surgery, with resolution of the abdominal mass after chemo-radiation, recurrence of the abdominal mass 1 year after salvage RC, with urinary obstruction requiring a nephrostomy tube. The second patient had cT2 disease, gross lymph node involvement, and urinary obstruction requiring a nephrostomy tube after initial aborted surgery with a good response to chemotherapy facilitating salvage cystectomy and subsequent development of an abdominal metastasis ≈1 year later. The third patient had cT1 disease and gross lymph node involvement that was upstaged to pT3b at salvage RC, and who died almost 5 years later from widespread bladder cancer metastases. All salvage RC patients who died were men and had T3 disease in the RC specimen. One patient is alive with bladder cancer recurrence to the rectum necessitating a colostomy at 10 months after salvage RC, despite a pathological stage of pTIS-only in the salvage RC specimen, with no positive lymph nodes. Three of the seven patients have no evidence of disease progression after salvage RC, with a mean survival since laparotomy of 18.1 (13.3–25.8) months and a mean survival since salvage RC of 9.9 (5.2–15.4) months. Figure 2 shows a Kaplan-Meier survival plot from the time of salvage RC for these seven patients.

Table 2.  The demographics and outcomes of seven patients who had salvage RC after a response to systemic chemotherapy for metastatic bladder cancer
VariableMean (range) or n
Age, years64 (48–77)
Follow-up, months30.5 (13.3–67.7)
Months from initial laparotomy to salvage RC11.1 (4.8–26.4)
Disease-specific mortality3 (43)
Months to death after initial laparotomy46.5 (38.3–62.7)
Months to death after salvage RC31.9 (12.2–57.9)
Alive with disease recurrence1
Alive with no evidence of disease3
Male5
Female2
Clinical stage
 Tis-only1
 T12
 T24
Salvage RC pathology
 Tis1
 T3a5
 T3b1
image

Figure 2. Overall mortality in the seven patients who had an aborted attempt at RC due to intraoperative findings of metastases, followed by subsequent salvage RC after a positive response to chemotherapy.

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 optimum treatment for patients found to have gross evidence of metastatic bladder cancer at the time of RC is not well defined. Should the urologist proceed with surgery when grossly positive nodes are found at RC? Recent series reported improved outcomes in some patients after RC, particularly when the primary tumour is organ-confined and lymph node involvement is small [9–13]. As outcomes for advanced disease might be poor regardless of treatment, many surgeons will choose the most conservative treatment.

Herr and Donat [14] attempted to address this question in a study on 84 patients with bulky node-positive (N2-3) bladder cancer found at RC and who had not had neoadjuvant chemotherapy. The patients had RC and extended PLND, and were followed for up to 10 years. The median survival time for all patients was 19 months. Disease-free survival was most pronounced among those patients whose primary tumours were pathologically confined compared to those with extravesical extension. This study supports an aggressive surgical approach for patients with grossly positive nodes found during RC, especially in those who might not survive non-operative treatments. Although no patients received chemotherapy unless there was tumour relapse, the authors did not report whether responders or long-term survivors were more likely to receive chemotherapy. No definitive conclusion can be drawn about the effect of chemotherapy on survival in these patients.

Several investigators reported improved outcomes with adjuvant chemotherapy in selected patients with high-risk bladder cancer [15–18]. Studies also showed a beneficial role for neoadjuvant chemotherapy in selected cases of bladder cancer [19,20]. Grossman et al.[20] reported that neoadjuvant MVAC therapy increased the likelihood of eliminating residual tumour, and was associated with improved survival among patients with locally advanced bladder cancer compared with RC alone, although that study did not include patients with nodal metastases. One possible mechanism for the survival benefit of neoadjuvant chemotherapy is tumour downstaging, which might enable subsequent resection of the primary cancer.

Surgery after chemotherapy might also have a beneficial role in selected cases of initially unresectable or metastatic bladder cancer [21–25]. Dodd et al.[21] reported their retrospective experience with 50 of 203 patients with initially unresectable or metastatic disease who had surgery for suspected or known residual disease after MVAC chemotherapy; 17 (34%) had no tumour at surgery and 30 (60%) had complete resection of residual tumour, indicating a complete response to chemotherapy plus surgery. Of the 30 patients, 10 (33%) were alive at 5 years. Sweeney et al.[25] reported a series of 11 patients with biopsy-confirmed TCC in the retroperitoneal lymph nodes who had surgery after a response to chemotherapy. Seven patients were treated with RC, positive retroperitoneal lymph nodes were found in nine, and the median survival was 14 months. Tumour involvement in fewer than two lymph nodes corresponded with an increased disease-specific survival. These results show a potential benefit from surgery after chemotherapy in selected patients with metastatic TCC confined to sub-diaphragmatic nodal fields.

The preferred sequence of chemotherapy (before or after surgery) was investigated by Millikan et al.[26] in a study of 140 patients with locally advanced urothelial cancer who were randomly assigned to receive MVAC in a neoadjuvant or adjuvant setting; that study included patients with node-positive disease. Although there was no significant difference in disease-specific survival between chemotherapy before or after surgery, there was a significant survival benefit with chemotherapy over surgery alone. A significant percentage of patients in the node-positive group could be cured by the combination of multi-agent chemotherapy and surgery; 81 (58%) remained disease-free with a median follow-up of 6.8 years. Patients with no residual muscle-invasive disease at RC after neoadjuvant chemotherapy were more likely to be cured.

It is interesting to consider what might have been the outcome for the present patients if surgery had been attempted rather than aborting the RC and waiting to assess the results of systemic chemotherapy. The prognosis appears to be poor for patients in whom plans for curative RC are abandoned due to intraoperative findings of metastases, despite subsequent systemic chemotherapy. A large proportion of the present patients never became candidates for salvage RC. This could be viewed as a possible justification to proceed with surgical extirpation in selected cases when intraoperative metastases are found and surgical resection is deemed feasible. Other factors, such as quality of life, will undoubtedly influence the decision of the surgeon. Proceeding with surgical extirpation could provide a palliative benefit in patients with advanced disease. However, subsequent chemotherapy would be less tolerable in these patients after a major operation. Given the poor prognosis and desire to do no harm, many surgeons will select a conservative treatment plan. Although the present study appears to show an overall survival advantage for patients who had salvage RC, these were patients who had responded to chemotherapy and were good surgical candidates, thus creating a selection bias for patients in better health.

Potential limitations of the present study include the small sample size, retrospective nature, and limited follow-up for some patients. We had insufficient patients to stratify outcomes for various clinicopathological factors. Our results will need to be validated in larger studies with a longer follow-up. Clearly, there is a need for improved clinical staging to prevent the surgeon from being faced with the dilemma of whether to proceed with RC after discovering metastases during surgery. The high rate of clinical understaging was reported by other investigators [27,28].

In conclusion, the prognosis is poor for patients with bladder cancer who undergo an aborted RC due to the finding of grossly positive lymph nodes or metastatic disease at the time of surgery. However, a subset of patients will respond to subsequent chemotherapy and might eventually become candidates for salvage RC. Larger, randomized studies are needed to define the optimum management for patients with grossly positive disease encountered at the time of RC.

REFERENCES

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