Outcome of surgical resection of pulmonary metastasis from urinary tract transitional cell carcinoma.

There is little information on pulmonary metastasectomy of urinary tract transitional cell carcinoma (TCC). In this study, we examined the long-term outcome and the factors associated with long-term survival after pulmonary metastasectomy of urinary tract TCC based on a 20-year single center experience. Between 1984 and 2005, 18 patients (12 men, six women) underwent pulmonary metastasectomy of the urinary tract TCC in our hospital. The clinicopathological and surgical data of these patients obtained from the medical records were analyzed in this retrospective study. The time interval between lung resection and death, or latest follow-up ranged from two to 200 months (median 52). Survival analysis was conducted by the Kaplan-Meier method and log-rank test. The cumulative three- and five-year survival rates were 59.8% and 46.5%, respectively. The number of resected metastatic tumors (solitary vs. multiple) was associated with long-term survival (P<0.05). The five-year survival rate of patients with solitary metastasis was 85.7% while that of patients with multiple metastases was 20.0%. Pulmonary metastasectomy of urinary tract TCC is associated with a favorable outcome, and solitary metastasis is associated with long-term survival. Aggressive management of solitary pulmonary metastasis from a urinary tract TCC is feasible in selected patients.


Introduction
Systemic chemotherapy remains the standard treatment for metastatic transitional cell carcinoma (TCC) of the urinary tract w1x. Modern cisplatin-based combination chemotherapy regimens introduced in the 1980s have been reported to be associated with overall response rates of 50-70% w2, 3x. Although cisplatin-based chemotherapy is highly effective initially, long-term survival is achieved in only a small proportion of patients. New treatment options, such as target therapies have been investigated w4x. Despite these efforts, metastatic TCC is still associated with poor prognosis.
Several groups have reported the surgical results of metastatic TCC including those in different locations in selected patients. In 1939, Barney and Churchill were the first to describe successful pulmonary metastasectomy w5x. It is now thought that surgical resection of pulmonary metastasis is an established treatment modality in patients with various types of metastatic diseases, such as colorectal cancer, osteosarcoma, and soft tissue sarcoma w6x. However, little is known about surgical resection of pulmonary metastasis from a urinary tract TCC. Because patients with pulmonary metastasis of urinary tract TCC have poor prognosis, surgical resection is rarely applied. To our knowledge, *Corresponding author. Tel.: q81-06-6972-1181; fax: q81-06-6981-8055. E-mail address: rkanzaki@tj8.so-net.ne.jp (R. Kanzaki).
there is little information on the significance and prognosis of patients who undergo resection of pulmonary metastatic lesions from urinary tract TCC.
In this retrospective study, we examined the long-term results, and factors associated with prolonged survival after pulmonary metastasectomy of urinary tract TCC based on a 20-year single center experience.

Patients and methods
In our hospital, lung resection is considered for patients with pulmonary metastasis of urinary tract TCC when the patient meets the following criteria. (1) Pulmonary metastases are deemed completely respectable and mediastinal or hilar lymph nodes are not involved by preoperative radiological studies. Generally, mediastinal or lymph node involvement is assessed by contrast-enhanced CT. Lymph nodes were interpreted as positive if )1 cm across the short-axis diameter.
(2) Metastatic disease is limited to the lungs.
(3) Locoregional control of the primary cancer was achieved or achievable. (4) Good overall general conditions and adequate respiratory function to tolerate lung resection.
According to the eligibility criteria described above, 18 patients (12 men, six women) underwent pulmonary metastasectomy of the urinary tract TCC in our hospital between 1984 and 2005. A retrospective analysis of these 18 patients was performed. Histopathological evaluation of the resected specimens of the lung confirmed urinary tract TCC metastases in all patients. Clinical information was obtained from the medical records of our hospital.
The primary sites were the bladder in nine patients, the upper urinary tract in six patients, and multiple tumors in both the bladder and upper urinary tract in three patients. Of nine patients with bladder cancer, one patient underwent transurethral resection of the bladder tumor (TUR-Bt), and eight patients underwent cystectomy. All six patients with renal pelvis or ureter cancer underwent nephroureterectomy. Of three patients with multiple tumors in both the bladder and renal pelvis or ureter, one patient underwent nephroureterectomy and cystectomy while two patients underwent nephroureterectomy and TUR-Bt. Seventeen patients underwent surgery alone as initial therapy for the primary tumor. One patient underwent simultaneous resection of the primary cancer and pulmonary metastasis. The mean and median time interval between resection of the primary urinary tract TCC and lung resection was 37 and 24 months, respectively (range: 0-126 months). The mean patient age at the time of metastasectomy was 66 years (range: 44-84 years). None of the patients had a history of resection of extrapulmonary metastases. Eight of the 18 patients including one patient who underwent simultaneous resection of the primary cancer and pulmonary metastasis received preoperative or postoperative cisplatin-based chemotherapy or both. Five patients received methotrexate, epirubicin, and cisplatin (MEC) chemotherapy preoperatively, and partial response (PR) was observed in four and no response in one. One patient received preoperative MEC and postoperative gemcitabine, cisplatin (GC), while another received preoperative methotrexate, vinblastine, adriamycin and cisplatin (M-VAC), and PR was observed in these two patients. One patient received postoperative MEC chemotherapy without any preoperative chemotherapy. The response to chemotherapy was assessed using the criteria described in a previous study w7x. Table 1 summarizes the clinicopathological characteristics of the patients.
The disease-free interval (DFI) was defined as the time period from surgical resection of primary cancer to the initial diagnosis of pulmonary metastasis. The criteria used for complete resection were the following: lack of other extrapulmonary metastases, no locoregional recurrence, and no macroscopic tumor tissue is left behind at lung resection.
Follow-up was generally based on chest X-ray or chest and abdominal CT, physical examination and laboratory blood tests performed every six to 12 months after lung resection. The follow-up information was obtained from the hospital medical records, the corresponding urological departments, letters from the general practitioners, or from the death certificates of the registry office. Patients or their families were contacted by phone if necessary. Overall survival was the main endpoint defined as the time interval between the date of lung resection and death, or the last follow-up for living patients. The time interval between lung resection and death, or latest follow-up in the present series ranged from two to 200 months (median 52).
All statistical analyses were conducted using StatView 5.0 software (SAS Institute, Berkley, CA, USA). Overall survival was analyzed by the Kaplan-Meier method using the date of pulmonary resection as the starting point. The significance of differences between groups was analyzed by the log-rank test. A P-0.05 was considered statistically significant.

Results
Complete resection was achieved in 16 patients (89%). There was no operative mortality. Two patients (11%) developed postoperative complications, including wound infection in one patient and pyothorax in another. Table 2 provides details on the metastatic tumors and lung resection. We excluded the patients with apparent hilar or mediastinal lymph node metastases determined by preoperative radiological examinations from candidate for pulmonary resection, however, two patients revealed to have hilar or mediastinal lymph node metastases pathologically after the operation. Three patients underwent repeat lung   Two patients who developed recurrences after lung resection were lost to follow-up. Eleven patients died of the disease, two patients are still alive with the disease, and three patients are alive with no evidence of disease (NED). All three patients who are alive with NED are long survivors, )48 months. Details of these three patients are listed in Table 3. The sites of recurrences after lung resection were the lung in five patients, bone in five patients, brain in two patients, urinary system in two patients, and unknown in one patient. The cumulative three-and five-year survival rates were 59.8% and 46.5%, respectively (Fig. 1).
The following factors were assessed by univariate analysis for their association with long-term survival; sex, age, primary site, pathological stage of the primary urinary tract TCC according to TNM classification, histological grading of the primary urinary tract TCC, perioperative chemotherapy beforeyafter lung resection, DFI, number of lung metastatic tumors, diameter of the largest resected lung metastases, type of resection (lobectomy or sublobar resection), presence of hilar or mediastinal lymph node metastasis, and completeness of resection. The results of univariate analysis are shown in Table 4. The number of resected metastatic tumors (solitary vs. multiple) was significantly associated with long-term survival (P-0.05). The five-year survival rate of patients with solitary metastasis was 85.7% while that of patients with multiple metastases was 20.0%.

Discussion
In our hospital, a total of 780 pulmonary metastasectomies have been performed for various diseases, such as colorectal cancer, soft tissue sarcoma, hepatocellular carcinoma according to the general eligibility criteria described in Patients and methods w8x. In this study, we analyzed pulmonary metastasectomy of the urinary tract TCC, with a special emphasis on the long-term outcome and the factors associated with prognosis.
According to the guidelines, systemic chemotherapy is the standard treatment for metastatic urinary tract TCC w1, 9x. When the tumor responds well to the chemotherapy, aggressive surgical approach is encouraged. However, when the disease is refractory to chemotherapy, the indication for metastasectomy is limited to symptomatic patients. Metastasectomy without chemotherapy is not usually recommended, however, there is no clear evidence against metastasectomy without chemotherapy.
In 1982, Cowles and colleagues w10x were the first group to report the surgical outcome of pulmonary metastasis of urinary tract TCC. They reported a median survival of five years for six patients following surgical resection for solitary pulmonary metastasis of urinary tract TCC without chemotherapy. A decade after that report, based on the development of modern cisplatin-based combination chemotherapy regimens w2x, aggressive surgical resection of metastatic TCC that responds well to chemotherapy is encouraged by some authors w7, 11x. However, the majority of patients in the above studies had locally advanced disease or metastatic pelvic lymph nodes, while only a few had pulmonary metastasis. Otto et al. w12x examined surgical results of metastatic TCC refractory to chemotherapy and concluded that the indication for metastasectomy was limited to symptomatic patients because no survival benefit was confirmed in this cohort. Siefker-Radtke et al. w13x reported a heterogeneous group of 31 patients with distant metastatic TCC. Their series included 24 patients with pulmonary metastasis, which is the largest number of cases treated by pulmonary metastasectomy. Nine of their 31 patients underwent metastasectomy without chemotherapy. The median overall survival time was 23 months with a five-year survival rate of 33%. Recently, Abe and co-workers w14x and Lehmann et al. w15x reported the surgical results of two heterogeneous groups of patients with distant metastatic TCC with a median overall survival of 28 months and 42 months, respectively.
In the present study, the cumulative three-and five-year overall survival rates for pulmonary metastasectomy were 59.8% and 46.5%, respectively. It is difficult to compare our results with those of previous studies with a five-year overall survival rate ranging from 27 to 33% w13, 15x, because of the heterogeneous metastatic sites of the patients reported in the previous studies. One possibility that could explain this difference is that pulmonary metastasectomy contributes to better outcome than metastasectomy of other sites, such as retroperitoneal lymph nodes, distant lymph nodes, brain, skin, and bone.
Neither the site of the primary tumor nor the surgical mode for the primary tumor affected survival in the present study. Because only one patient received perioperative chemotherapy as an initial treatment for the primary tumor (the patient underwent simultaneous resection of the primary cancer and pulmonary metastasis received preoperative MEC chemotherapy), the effect of chemotherapy as initial treatment for primary tumor cannot be assessed. In the present study, similar to the report of Siefker-Radtke et al. w13x, neither chemotherapy nor the response status to such therapy affected survival. In fact, two of the three long-term survivors without disease in our series underwent surgery alone. Lehmann et al. w15x also reported two longterm survivors who never received chemotherapy. Initial treatment for the primary tumor also does not affect survival in the present study. Based on these data, we conclude that surgical resection is one of the useful treatment for a subgroup of patients with pulmonary metastasis of urinary tract TCC. The present study identified solitary metastasis as a factor associated with prolonged survival after pulmonary metastasectomy. To our knowledge, this is the first study that identified this factor to be associated with prolonged survival after pulmonary metastasectomy of the urinary tract TCC. Siefker-Radtke et al. w13x reported that age, gender, race, primary site, duration of chemotherapy after the diagnosis of metastases before resection and pathology of the primary tumor or metastases did not influence survival in their series, although the number of metastatic tumors was not assessed in their report. In the present study, the survival of patients with solitary metastasis was quite good. These data are in agreement with the first report on the surgical results of pulmonary metastasis of the urinary tract TCC by Cowles et al. w10x.
The limitation of this study is the small number of patients. Accordingly, it is difficult to define the significant prognostic factors using multivariate analysis. Furthermore, a longer follow-up period is needed to determine the true recurrence rate in the long-term survivors.

Conclusions
Surgical resection of pulmonary metastasis of urinary tract TCC provides favorable long-term outcome in selected patients. Solitary metastasis was associated with long-term survival after pulmonary metastasectomy. Aggressive management of patients with solitary pulmonary metastasis from urinary tract TCC is feasible in selected patients.