Adjuvant Cisplatin Plus Methotrexate Versus Methotrexate, Vinblastine, Epirubicin, and Cisplatin in Locally Advanced Bladder Cancer: Results
http://www.100md.com
《临床肿瘤学》
the Trial Group AB 05/95 "Arbeitsgemeinschaft Urologische Onkologie" of the German Cancer Society
Division of Biostatistics, Central Institute of Mental Health Mannheim/University of Heidelberg, Heidelberg, Germany
ABSTRACT
PURPOSE: Radical cystectomy as standard treatment of muscle-invasive urothelial carcinoma of the urinary bladder cures less than 50% of patients with locally advanced bladder cancer. We compared two adjuvant combination chemotherapies in patients with stage pT3a-4a and/or pathologic node-positive transitional-cell carcinoma of the bladder after radical cystectomy.
PATIENTS AND METHODS: A total of 327 patients were randomly assigned to either adjuvant systemic chemotherapy with three cycles of cisplatin 70 mg/qm2 on day 1 and methotrexate 40 mg/qm2 on days 8 and 15 of a 21-day cycle (CM) or three cycles of methotrexate 30 mg/qm2 on days 1, 15, and 22, vinblastine 3 mg/qm2 on days 2, 15, and 22, epirubicin 45 mg/qm2 on day 2, and cisplatin 70 mg/qm2 on day 2 of a 28-day cycle (M-VEC).
RESULTS: The hazard ratio for progression-free survival as the primary end point was 1.13 (90% CI, 0.86 to 1.48) for 163 CM patients compared with 164 M-VEC patients whose right-hand limit remained below the upper bound compatible with the noninferiority hypothesis ( = .0403). The 5-year progression-free, tumor-specific, and overall survival rates (point estimates ± SE) for CM versus M-VEC were 46.3% ± 4.6% v 48.8% ± 4.5%, 52.0% ± 4.6% v 52.3% ± 4.8%, and 46.1% ± 4.3% v 45.1% ± 4.6%, respectively. WHO grade 3 and 4 leukopenia occurred in 7.0% of patients treated with CM and 22.2% of patients treated with M-VEC (P < .0001).
CONCLUSION: CM cannot be considered inferior to M-VEC with regard to progression-free survival of patients with locally advanced bladder cancer after radical cystectomy. Moreover, patients receiving adjuvant CM combination therapy experienced significantly less grade 3 and 4 leukopenia than patients treated with M-VEC.
INTRODUCTION
Radical cystectomy is recommended as curative treatment for locally advanced bladder cancer including tumor stages pT3 and pT4a and/or involvement of regional lymph nodes. However, more than half of these patients experience relapse, with distant metastasis being the predominant form of disease recurrence. Therefore, combination chemotherapy has been investigated as neoadjuvant or adjuvant adjunctive treatment in numerous trials. A recent meta-analysis on neoadjuvant chemotherapy for invasive bladder cancer described a modest but clear improvement in survival, encouraging the use of platinum-based combination therapy.1 For the adjuvant approach, three randomized trials have suggested a relapse-free survival improvement for patients receiving combination chemotherapy after radical cystectomy compared with patients undergoing surgery alone.2-4
The combination regimen of methotrexate, vinblastine, doxorubicin, and cisplatin (M-VAC), which was first reported as a palliative approach for advanced metastatic urothelial carcinoma in 1985,5 has succeeded as a gold standard treatment in randomized trials.6,7 The first randomized trial comparing adjuvant M-VAC or methotrexate, vinblastine, epirubicin, and cisplatin (M-VEC) to radical cystectomy alone found a significant progression-free survival improvement of 50% for the adjuvant group 40 months after surgery.3 Despite the efficacy of M-VAC or M-VEC combination therapy, severe hematologic and nonhematologic toxicity and a treatment-related mortality rate of up to 4% have caused concern.
The present study was initiated in 1994 as a noninferiority trial of a reduced chemotherapy regimen consisting of cisplatin plus methotrexate (CM) compared with conventional M-VEC, with the intention of finding a less toxic but effective regimen for the adjuvant treatment of patients with resected locally advanced bladder cancer with no substantial loss of efficacy.
PATIENTS AND METHODS
The study protocol was reviewed and approved by the committee of the "Arbeitsgemeinschaft Urologische Onkologie" of the German Cancer Society as well as local ethics committees of participating centers.
Eligibility
Patients eligible for the randomization process had to meet the following inclusion criteria. Disease status after radical cystectomy must be histologically confirmed stage pT3a-4a and/or pathologic node-positive transitional-cell carcinoma of the bladder (1997 TNM classification8). Squamous cell carcinoma and/or adenocarcinoma components were allowed if transitional-cell carcinoma was present. Radical cystectomy denoted the removal of the entire bladder, prostate, and seminal vesicles in men and removal of anterior pelvic organs in the female, including, if indicated, a portion of the anterior vagina. The method of urinary diversion was left to the discretion of the investigator. Bilateral pelvic lymph node dissection was a prerequisite for correct staging and included a full dissection of the lymph nodes bordered by the internal iliac arteries, external iliac arteries, and the pelvic floor bilaterally including obturator nodes. Any evidence of macroscopic or microscopic incomplete resection (tumor-positive margin of the specimen or grossly enlarged unresected lymph nodes) was considered an exclusion criterion. Distant metastasis was excluded by preoperative staging, including at least chest x-ray and abdominal ultrasound. Furthermore, patients between 18 and 70 years old were included if they had a Karnofsky performance score of at least 80 with adequate renal and liver function as well as bone marrow reserve (measured creatinine clearance 70 mL/min, serum creatinine 1.3 mg/dL, bilirubin < 1.5 mg/dL, ALT or AST < 60 U/L, WBC count 3.0 x 109/L, platelets 150 x 109/L, and hemoglobin 10 g/dL). Local intravesical adjuvant chemotherapy or immunotherapy before radical cystectomy was allowed. Furthermore, a written informed consent according to institutional and federal guidelines had to be signed by the patient. Patients were ineligible if the time interval between radical cystectomy and the first day of chemotherapy exceeded 4 months and if they had used any investigational agent in the months before enrollment onto the study.
Treatment Plan
The CM regimen consisted of cisplatin 70 mg per square meter of body-surface area (mg/qm2) on day 1 and methotrexate 40 mg/qm2 on days 8 and 15 every 21 days for a maximum of three cycles. M-VEC (methotrexate 30 mg/qm2 on days 1, 15, and 22; vinblastine 3 mg/qm2 on days 2, 15, and 22; epirubicin 45 mg/qm2 on day 2, and cisplatin 70 mg/qm2 on day 2) was administered every 28 days for a maximum of three cycles.
The following dose adjustments were planned. Methotrexate was omitted if the leukocyte count was less than 2.5 x 109/L or the thrombocyte count was less than 100 x 109/L. The beginning of a new cycle was delayed until the leukocyte count was greater than 3.0 x 109/L and the thrombocyte count was greater than 100 x 109/L. Doses of cisplatin and methotrexate were reduced by 50% if measured creatinine clearance was 50 to 70 mL/min, or the drugs were omitted if the creatinine clearance was less than 50 mL/min. Patients were to receive full supportive care. Use of growth factors was recommended only in case of leukopenic fever, dose omissions for methotrexate and vinblastine as a result of leukopenia, or delay of a consecutive cycle as a result of leukopenia. Hematologic and nonhematologic toxicity was graded according to the WHO grading system. Study drug therapy was discontinued if there was evidence of progressive disease under therapy, if the attending physician considered a change of therapy to be in the best interest of the patient, if the patient requested discontinuation, or if the drug exhibited unacceptable toxicity.
Follow-Up of Patients
Follow-up of both treatment groups occurred at 3-month intervals for 2 years, then at 6-month intervals for 3 years, and yearly thereafter. Follow-up consisted of physical examination, abdominal ultrasound and chest radiography, and computed tomography scanning or bone scanning as clinically indicated. The site and date of the first relapse, as well as the date and the reason in case of death, were recorded. Relapse was defined as the detection of at least one suspicious lesion that could not be identified by histologic, clinical, or laboratory criteria as an independent secondary malignancy. In case of relapse, it was left to the discretion of the patient and the attending physician whether further therapy, such as second-line chemotherapy, radiation, or surgical treatment, was initiated.
Statistical Methods
The primary basis for the statistical planning and analysis of the trial was the noninferiority version of the log-rank test for equivalence of two survivor functions9 satisfying an ordinary proportional hazards model in the sense of Cox.10 The test was applied with an equivalence margin of = .15, which gives the maximum tolerable difference of progression-free survival probabilities to be expected under both treatments for the same time since cystectomy. Practical implementation of the procedure was performed by estimating the regression coefficient ? of a proportional hazards model, with the indicator of treatment group CM as the only covariate, and comparing the right-hand limit of a 90% CI for the hazard ratio (HR) = e? to 1.5077 as a critical upper bound. For the purpose of checking the adequacy of the proportional hazards assumption, the procedure implemented by the S-plus (Statistical Sciences, Seattle, WA) function cox.zph was used (for a detailed description of the statistical rationale behind it, see Therneau and Grambsch11). Because the P value obtained from this test turned out to be as large as .45, there was no reason to question the appropriateness of the model.
However, estimation of median progression-free survival time was based on a fully parametric model because both observed survivor functions dropped below 50% so late that the number of patients still at risk was too small for providing a sufficiently accurate estimate of the conditional probability of surviving the interval containing the observed median. Of the standard parametric models, the log-normal was the model that showed the best fit to the data. Accordingly, confidence limits for the median progression-free survival time were computed by transforming those for the expected value of the distribution of log T.
In addition to progression-free survival time, which had been chosen as the primary end point for the confirmatory analysis of this trial, the same techniques were also applied to the data obtained for the length of survival and time until death from the specific tumor. All eligible patients who received any adjuvant treatment were included in the analysis of possible toxic effects. The incidences of WHO grade 3 and 4 toxicities were compared using Fisher's exact test for homogeneity of two binomial distributions.
RESULTS
Between January 1, 1994, and September 30, 2000, 335 patients were registered onto the trial (Fig 1). Eight patients (2.4%) were deemed ineligible because inclusion criteria were not met (age, renal function, metastatic disease at time of radical cystectomy, secondary malignancy, and withdrawal of consent). The remaining 327 patients were equally assigned to adjuvant systemic chemotherapy with CM or M-VEC by randomization after having signed informed consent.
Demographic factors were similar between the two groups (Table 1). Median age was 60.5 years, and the ratio of males to females was almost 4:1. More than half of the patient group (56.6%) had tumor-positive regional lymph nodes. The median follow-up time for patients living free of relapse at the time of analysis was 42 months.
Treatment
The median time from surgery to chemotherapy was 56 days in both arms. Of the 163 patients assigned to receive adjuvant CM, 123 (75.5%) completed three cycles as planned (Table 2); 12.9% (21 of 163 patients) discontinued treatment early because of toxic effects or for other reasons. Nineteen patients (11.7%) declined treatment, and four patients (2.5%) died during the course of treatment. One patient died of cardiac failure (day 18 of a third cycle), two patients experienced severe bone marrow toxicity after two cycles, and one patient developed lethal pneumonia after one cycle.
Of the 164 patients assigned to the M-VEC arm, 113 (68.9%) completed three cycles; 21.4% (35 of 164 patients) stopped treatment early. Sixteen patients (9.8%) declined treatment, and one patient (0.6%) died during treatment with M-VEC as a result of cardiac dysfunction. All patients were analyzed as belonging to the treatment arm assigned by randomization, adopting the intent-to-treat principle.
Surgical Procedures
The only surgery-related requirements for eligibility were radical cystectomy with curative intent and en bloc resection of the tumor with negative margins, as well as absence of metastatic or unresected transitional-cell carcinoma. However, because patients were usually identified postoperatively, specific surgical and histopathologic procedures could not be required.
Toxicity
Hematologic toxic effects predominated. The most common hematologic toxic effect was leukopenia, with a significantly higher rate in patients treated with M-VEC compared with CM (22.2% v 7.0%, respectively). Except for a significant difference in grade 3 alopecia in favor of CM (1.9% in CM v 24.7% in M-VEC), no difference was observed for other hematologic or nonhematologic factors. Four patients (2.5%) died as a result of a toxic effect attributed to CM chemotherapy, and one patient (0.6%) died as a result of therapy in the M-VEC arm. This difference of therapy-related mortality was not statistically significant (Fisher's exact test, P = .371). The toxic effects classified as WHO grade 3 or higher that occurred among treated patients are listed in Table 3.
Progression-Free Survival
The median follow-up period for patients living free of progression was greater than 3 years. Tumor relapse was reported in 46.0% of the patients (75 of 163 patients) in the CM group and 44.5% of patients (73 of 164 patients) in the M-VEC group. The median duration of progression-free survival was 43.4 months in the CM arm and 49.7 months in the M-VEC arm by intent-to-treat analysis (Table 4). The 5-year progression-free survival rates were 46.3% in the CM treatment group and 48.8% in the M-VEC group (Fig 2). The HR for disease progression in the CM group, as compared with the M-VEC group, was 1.13 (90% CI, 0.86 to 1.48; P = .0403).
In the subgroup analyses of patients with lymph node–positive disease, the median progression-free survival was 36.2 months (90% CI, 27.5 to 47.7 months) in the CM group and 32.4 months (90% CI, 24.8 to 42.3 months) in the M-VEC group. The 5-year progression-free survival rates of patients with lymph node–positive disease were 41.9% in the CM group and 40.1% in the M-VEC group (Fig 3). The HR for disease progression in the CM group, as compared with the M-VEC group, was 0.95 (90% CI, 0.67 to 1.33; P = .0126).
In the analyses of patients with tumor extension beyond the bladder wall but without lymph node involvement, the median progression-free survival was 52.0 months (90% CI, 36.5 to 74.2 months) in the CM arm and 87.7 months (90% CI, 59.8 to 128.6 months) in the M-VEC arm. The 5-year progression-free survival rates were 51.5% in the CM group and 60.8% in the M-VEC group (Fig 4). The HR for disease progression in the CM group, as compared with the M-VEC group, was 1.46 (90% CI, 0.94 to 2.29; P = .4571).
Tumor-Specific and Overall Survival
Sixty-five patients in the CM treatment group (39.9%) and 62 patients in the M-VEC treatment group (37.8%) died of tumor disease during the follow-up period. The median duration of tumor-specific survival was 57.3 months in the CM group and 63.3 months in the M-VEC group (Fig 5). The HR for tumor-related death in the CM group, as compared with the M-VEC group, was 1.13 (90% CI, 0.84 to 1.51; P = .0528).
The 5-year rates of tumor-specific survival were 52.0% in the CM group and 52.3% in the M-VEC group (Table 4, Fig 5). In the CM and M-VEC arms, 14 patients (8.6%) and 15 patients (9.1%) died as a result of causes not related to tumor, respectively. The median duration of overall survival was 47.1 months in the CM group and 51.8 months in the M-VEC group (Table 4, Fig 6). The HR for death irrespective of cause in the CM group, as compared with the M-VEC group, was 1.10 (90% CI, 0.88 to 1.44; P = .0255). The 5-year rates of overall survival were 46.1% in the CM group and 45.1% in the M-VEC group.
Site of Tumor Relapse
Classification of first relapses with regard to site resulted as follows (Table 5). Local recurrence only occurred in 12 patients in the CM group (7.3%) and seven patients in the M-VEC group (4.3%). Lymphatic relapse only, which was typically encountered as retroperitoneal lymph node disease, was reported in 4.3% of CM patients (seven of 163 patients) and 6.1% of M-VEC patients (10 of 164 patients). Visceral metastases (without bone metastases) were found in 20.2% of patients in the CM group (33 of 163 patients) and 15.9% of patients in the M-VEC group (26 of 164 patients); at least 12.3% of CM patients (20 of 163 patients) and 15.2% of M-VEC patients (25 of 164 patients) developed bone metastases.
DISCUSSION
The frequent occurrence of tumor relapses after radical cystectomy for locally advanced bladder cancer, the favorable results of three former adjuvant trials,2-4 and the limited tolerability of adjuvant M-VEC or M-VAC combination chemotherapy provided the rationale for the present study. After introducing effective M-VAC combination chemotherapy in urothelial cancer in 1985,5 two successive, randomized, phase III trials demonstrated the superiority of the M-VAC regimen in advanced urothelial cancer compared with single-agent cisplatin7,12 or with cisplatin, cyclophosphamide, and doxorubicin combination therapy.6 Because of the pronounced hematologic and nonhematologic toxicity with M-VAC or M-VEC combination therapy, other potentially less toxic combination therapies, such as CM, have been investigated in advanced urothelial cancer.13,14
Neoadjuvant administration of chemotherapy has been exceedingly investigated compared with the adjuvant approach for patients with a high risk of relapse undergoing definite treatment by surgery or radiation. Although the majority of individual neoadjuvant studies was not able to demonstrate a significant survival advantage for pre-emptive chemotherapy compared with definitive treatment alone, a recent meta-analysis based on individual patient data from 10 randomized neoadjuvant trials was able to demonstrate a significant 13% reduction in the risk of death (HR = 0.87; 95% CI, 0.78 to 0.98; P = .016), which was equivalent to a 5% absolute overall survival benefit at 5 years from 45% to 50%.1 This meta-analysis also stated that a significant advantage exists using platinum-based combination chemotherapy compared with single-agent platinum. The most recent neoadjuvant study, which was not included in the former meta-analysis, achieved a survival advantage with borderline significance for 154 patients assigned to three cycles of neoadjuvant M-VAC combination therapy versus 153 patients randomly assigned to surgery alone (median survival, 77 months in the neoadjuvant arm v 46 months in the surgery-only arm; P = .06).15
Whether neoadjuvant or adjuvant application of systemic chemotherapy for locally advanced bladder cancer is the preferable treatment modality remains a matter of debate. To date, only one study has investigated the neoadjuvant and adjuvant treatment strategy directly.16 Similar to our experience, 40% of patients in this trial with histologically proven regional lymph node metastasis experienced long-term progression-free survival in both treatment arms.
In contrast to neoadjuvant administration of chemotherapy, proponents of adjuvant chemotherapy for locally advanced bladder cancer emphasize optimal patient selection based on exact assessment of histopathologic tumor stage. This allows the selection of patients with the highest risk for relapse, and only minimal occult tumor burden after radical cystectomy needs to be treated. Subsequently, improvement of recurrence-free survival by adjuvant combination chemotherapy was corroborated by three individual randomized controlled studies from Germany and the United States for the adjuvant administration of combination chemotherapy after radical cystectomy.2-4
Skinner et al2 randomly assigned 91 locally advanced bladder cancer patients to radical cystectomy plus four cycles of adjuvant chemotherapy (predominantly cisplatin, cyclophosphamide, and doxorubicin combination therapy) versus surgery alone. The trial was stopped because of an interim analysis that demonstrated a significant disease-free survival advantage for the adjuvant treatment arm 5 years after cystectomy (51% for adjuvant chemotherapy v 34% for surgery alone, P < .011). St?ckle et al3 terminated their study after an interim analysis of 49 patients demonstrated a significant disease-free survival difference of 50% at 3.5 years in favor of 26 patients receiving adjuvant treatment with M-VAC or M-VEC compared with the control group (63% for adjuvant chemotherapy v 13% for cystectomy alone; P = .0005).
A third randomized trial with a favorable result for adjuvant combination chemotherapy performed at Stanford University compared adjuvant cisplatin, methotrexate, and vinblastine combination chemotherapy to cystectomy alone.4 In contrast to the former two studies, deferred chemotherapy on progression was suggested by the protocol for patients assigned to the nonadjuvant arm. This study was also closed before accrual of the preplanned number of patients because of a significant difference in progression-free survival in favor of patients receiving adjuvant chemotherapy. Median time to progression for the 25 patients assigned to the adjuvant treatment arm was 37 months versus 12 months in the cystectomy-alone group (P < .01), whereas overall survival was not significantly different. Criticisms of all three former trials have been summarized by Sylvester and Sternberg17 and address small sample size, early stopping of patient entry and premature closure, statistical analyses, reporting of results, and drawing conclusions.
Adjuvant chemotherapy trials for invasive bladder cancer that did not find a significant difference for patients receiving adjuvant chemotherapy compared with a surgery-only group have been reported from Switzerland,18 Italy,19 and Germany.20 This may be attributed to the exclusion of lymph node–positive patients19 and to the administering of only single-agent cisplatin as adjuvant treatment to a predominantly pT2pN0 population.18
Only limited information on the degree of toxicity has been revealed from any of these adjuvant trials. Reports on palliative M-VAC or M-VEC combination chemotherapy causing grade 3 to 4 myelosuppression in more than half of the patients, a 25% incidence of nadir sepsis, and drug-related deaths in at least 3% of patients 21 prompted investigators to search for more tolerable but still active regimens. Efficacy of CM combination therapy as implemented for the current trial had been reported previously for 53 patients with advanced urothelial cancer.13 Although the majority of patients had dose omissions or delays, an overall response rate of 46% was noted. The standard M-VEC treatment arm of the current trial incorporated epirubicin instead of doxorubicin, as in the original M-VAC schema, because cardiotoxicity was of concern at the time of drafting the protocol.
Because of the nature of the current trial testing for noninferior efficacy of the reduced combination (CM) compared with the full combination (M-VEC), the results may only lend support to but not prove the superiority of adjuvant combination chemotherapy compared with surgery alone. Remarkably, the 5-year progression-free survival rate of greater than 40% for lymph node–positive disease after radical cystectomy plus adjuvant combination chemotherapy (CM or M-VEC) represents an average standard of care of 40 uro-oncologic centers in Germany. Retrospective studies of patients with lymph node–positive bladder cancer receiving adjuvant chemotherapy after radical cystectomy versus surgery alone reported significant 5-year overall survival advantages of 43% v 17% (P = .03)22 and 45% v 21% (P = .031)23 for the adjuvant group, respectively. A similar survival comparison from the Mayo Clinic demonstrated a 5-year tumor-specific survival advantage of 55% v 32% (P = .005) in favor of the adjuvant chemotherapy group compared with surgery alone, respectively,24 notwithstanding the limitation that any retrospective result may be biased by patient selection. Finally, the recurrence-free survival rate of 41% at 5 years for all lymph node–positive patients (n = 185) in our trial is slightly higher than 39% rate that was recently reported from another North American center of excellence administering adjuvant chemotherapy to 139 patients with lymph node–positive disease.25
When assessing the possible effects of baseline factors (eg, age, sex, tumor stage, blood cell counts, renal and liver function, and size of recruiting center) on the risk of progression using multivariate Cox regression analysis, the only significant predictor was the number of tumor-positive lymph nodes (data not shown). The current trial did not control for either the number of lymph nodes removed during radical cystectomy or the extent of pelvic lymphadenectomy because patients were usually identified postoperatively to be eligible for this trial. Therefore, a specific surgical and histopathologic protocol could not be required. Some authors have suggested a survival improvement for patients undergoing extensive lymphadenectomy with cystectomy in a retrospective survey,25-27 whereas others have seen a significant survival advantage with extensive lymphadenectomy only in patients with organ confined ( pT2) bladder cancer.28 Future trials on adjuvant chemotherapy for locally advanced bladder cancer may focus on this issue with more precision.
As previously reported for the palliative indication of combination chemotherapy in unresectable advanced urothelial cancer, the gold standard status of M-VAC therapy has declined.29 Results of a large, multinational, phase III trial comparing M-VAC with gemcitabine plus cisplatin combination therapy showed similar response rates and comparable overall survival in 405 patients. Although patient numbers and study design of this trial did not allow for the testing of significant noninferiority, gemcitabine plus cisplatin has become a well-established regimen for the treatment of advanced bladder cancer because of significantly less clinically relevant toxicity and better cost effectiveness.
The implications of our results must be assessed with a view to the equivalence margin of 0.15, which had been specified in the protocol for the difference between the respective survivor functions at any point of time. Thus, the major statistical hypothesis that could be established by means of our data does not rule out the possibility that the full combination therapy comprising all four drugs is more efficient than the reduced chemotherapy regimen. But we can be certain that this advantage, which is supported by the positive signs of the point estimates obtained from the total samples, fails to be of significant clinical relevance. For the adjuvant indication of combination chemotherapy with curative intent in resected locally advanced bladder cancer, the current report provides evidence that the efficacy of the less toxic regimen CM cannot be considered substantially inferior to the efficacy of M-VEC.
Appendix
The members of the Collaborative Group were as follows: Secretariat: M. St?ckle (principal investigator), S. Wellek (biostatistician), and J. Lehmann and M. Retz, C.Wiemers (data managers). Investigators (the number of patients enrolled onto the study is shown in parentheses): C. Wiemers, D. Sternberg, L. Franzaring, M. St?ckle, J. Thüroff, J. Beck, C. Huber, Johannes-Gutenberg-Universit?t, Mainz (n = 37); C. Weining, H.J. Piechota, L. Hertle, Universit?tsklinikum Münster, Münster (n = 32); G. Steiner, P. Albers, S.C. Müller, Universit?tsklinikum Bonn, Bonn (n = 28); C. Vierneisel, A. G?ll, M. Beitzinger, D. Frohneberg, St?dtisches Klinikum, Karlsruhe (n = 26); T. Becker, T. K?lble, H. Riedmiller, Philipps-Universit?t, Marburg (n = 22); H.J. Knopf, G. Engelhardt, P.-J. Funke, Evangelisches-Jung-Stilling Krankenhaus, Siegen (n = 21); J. Roloff, T. Sch?nfeld, P. Walz, Klinikum Lüdenscheid, Lüdenscheid (n = 14); C. Wiemers, M. Retz, J. Lehmann, A. Bannowsky, M. St?ckle, Christian-Albrechts-Universit?t, Kiel (n = 13); S. Langbein, F. Reiher, E.P. Allhoff, Otto-von-Guericke-Universit?t, Magdeburg (n = 13); M. Schiller, D. Sternberg, G. Hutschenreiter, Evangel. und Johanniter Klinikum, Oberhausen (n = 13); M. Staehler, M. Müller, K. Miller, Universit?tsklinikum Benjamin-Franklin, Berlin (n = 12); U. Gertenbach, U. Kaldenbach, J. Thüroff, S. Roth, Universit?t Witten-Herdecke, HELIOS Klinikum Wuppertal (n = 11); S. Tedsen, A. B?hle, D. Jocham, Universit?t Lübeck, Lübeck (n = 7); W. Kieser, H. Riedmiller, Julius-Maximilian-Universit?t, Würzburg (n = 6); G.M. Wacker-Backhaus, G. Ostendorf, C. Ehlert, H. Becker, Marienkrankenhaus, Hamburg (n = 6); L. Weissbach, Krankenhaus am Urban, Berlin (n = 6); C. Charvalakis, L. Lampante, St?dtisches Krankenhaus Kemper Hof, Koblenz (n = 6); V. Poulakis, E. Becht, S. Al-Batran, A. Knuth, E. Jaeger, Krankenhaus Nordwest, Frankfurt am Main (n = 6); R. Groh, R. Horsch, Kreiskrankenhaus Offenburg, Offenburg (n = 5); N. Vennemann, R.G. Anding, H. van Ahlen, Klinikum Osnabrück, Osnabrück (n = 5); T. K?lble, St?dtisches Klinikum Fulda, Fulda (n = 5); B. Buhmann, R. Hausmann, F. Pinkenburg, Kreiskrankenhaus Rendsburg, Rendsburg (n = 4); J. Gschwend, R. Hautmann, Universit?t Ulm, Ulm (n = 4); N.W. Fischer, D. Neisius, Krankenhaus der Barmherzigen Brüder, Trier (n = 3); K. Murr, M. Kühn, Johanniter-Krankenhaus, Stendal (n = 3); M.G. Friedrich, H. Huland, Universit?tsklinikum Eppendorf, Hamburg (n = 2); S. Siemer, U. Humke, Universit?t des Saarlandes, Homburg/Saar (n = 2); J. Steffens, St Antonius Hospital, Eschweiler (n = 2); D. Sternberg, M. Beer, Franziskus Krankenhaus, Berlin (n = 2); R. Nitze, R. Sintermann, Evangelisches Krankenhaus "Lutherhaus," Essen (n = 2); A. Kuczyk, U. Jonas, Medizinische Hochschule, Hannover (n = 2); S. Kiani, U. Tunn, St?dtische Kliniken, Offenbach (n = 2); G. Müller, E. T?lle, Herz-Jesu-Krankenhaus, Münster-Hiltrup (n = 2); L. Neubauer, St Franziskus Hospital, Lohne (n = 1); T. Drechsler, M. Hartmann, Hamburg Bundeswehr Krankenhaus (n = 1); U. Gertenbach, Allgemeines Krankenhaus, Hagen (n = 1); U. Engelmann, Universit?t K?ln (n = 1); D. Unverferth, Kreiskrankenhaus, Aurich (n = 1); E. Winter, P. Bub, Klinikum Schwerin (n = 1); G. Block, H.E. Reichert, Kreiskrankenhaus V?lklingen (n = 1); A. Nonnenmacher, Diakonie-Krankenhaus, Schw?bisch-Hall (n = 1); W. Rulf, W. Langhorst, Onkologische Praxis, Erkrath (n = 1); C.J. Güdemann, G. Staehler, Ruprecht-Karls-Universit?t, Heidelberg (n = 1).
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank the patients as well as the clinicians and practitioners who participated in this trial. We also thank the staff at the central trial offices in Homburg/Saar, Germany (Amelie Teuchert, Irene Lehmann, and Vera Mathei-Mayer), and in Kiel, Germany (Urte Fischer-Nürnberg, Astrid Hallen, Cora Prien, and Birgit Küppers).
NOTES
Supported in part by the Gunther-Voges Gesellschaft and Stiftung Berliner Bank, Germany (neither sponsor had any influence on study design or on the collection, analysis, or interpretation of data; and neither had any influence in the writing of the report or the decision to submit the manuscript for publication).
Presented at the following meetings: 54th Congress of the German Society of Urology in Wiesbaden, Germany, September 18-21, 2002; XVIIIth Congress of the European Association of Urology in Madrid, Spain, March 12-15, 2003; 98th Annual Meeting of the American Urological Association Chicago, IL, April 26-May 1, 2003; and the 39th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, May 31-June 3, 2003.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Advanced Bladder Cancer Meta-Analysis Collaboration: Neoadjuvant chemotherapy in invasive bladder cancer: A systematic review and meta-analysis.Lancet 361: 1927–1934, 2003
Skinner DG, Daniels JR, Russell CA, et al: The role of adjuvant chemotherapy following cystectomy for invasive bladder cancer: A prospective comparative trial. J Urol 145: 459-464, 1991
St?ckle M, Meyenburg W, Wellek S, et al: Advanced bladder cancer (stages pT3b, pT4a, pN1 and pN2): Improved survival after radical cystectomy and 3 adjuvant cycles of chemotherapy—Results of a controlled prospective study. J Urol 148: 302-306, 1992
Freiha F, Reese J, Torti FM: A randomized trial of radical cystectomy versus radical cystectomy plus cisplatin, vinblastine and methotrexate chemotherapy for muscle invasive bladder cancer. J Urol 155: 495-499, 1996
Sternberg CN, Yagoda A, Scher HI, et al: Preliminary results of M-VAC (methotrexate, vinblastine, doxorubicin and cisplatin) for transitional cell carcinoma of the urothelium. J Urol 133: 403-407, 1985
Logothetis CJ, Dexeus FH, Finn L, et al: A prospective randomized trial comparing MVAC and CISCA chemotherapy for patients with metastatic urothelial tumors. J Clin Oncol 8: 1050-1055, 1990
Loehrer PJ Sr, Einhorn LH, Elson PJ, et al: A randomized comparison of cisplatin alone or in combination with methotrexate, vinblastine, and doxorubicin in patients with metastatic urothelial carcinoma: A cooperative group study. J Clin Oncol 10: 1066-1073, 1992
Sobin LH, Wittekind CH: UICC TNM Classification of Malignant Tumors (ed 4). New York, NY, John Wiley & Sons, Inc, 1997
Wellek S: Testing Statistical Hypotheses of Equivalence. Boca Raton, FL, Chapman & Hall/CRC, 2003
Cox DR: Regression models and life tables with discussion. J R Stat Soc B 34: 187-220, 1972
Therneau T, Grambsch P: Modeling Survival Data: Extending the Cox Model. Berlin, Germany, Springer, 2000
Saxman SB, Propert KJ, Einhorn LH, et al: Long-term follow-up of a phase III intergroup study of cisplatin alone or in combination with methotrexate, vinblastine, and doxorubicin in patients with metastatic urothelial carcinoma: A cooperative group study. J Clin Oncol 15: 2564-2569, 1997
Stoter G, Splinter TA, Child JA, et al: Combination chemotherapy with cisplatin and methotrexate in advanced transitional cell cancer of the bladder. J Urol 137: 663-667, 1987
Hillcoat BL, Raghavan D, Matthews J, et al: A randomized trial of cisplatin versus cisplatin plus methotrexate in advanced cancer of the urothelial tract. J Clin Oncol 7: 706-709, 1989
Grossman HB, Natale RB, Tangen CM, et al: Neoadjuvant chemotherapy plus cystectomy compared with cystectomy alone for locally advanced bladder cancer. N Engl J Med 349: 859-866, 2003
Millikan R, Dinney C, Swanson D, et al: Integrated therapy for locally advanced bladder cancer: Final report of a randomized trial of cystectomy plus adjuvant M-VAC versus cystectomy with both preoperative and postoperative M-VAC. J Clin Oncol 19: 4005-4013, 2001
Sylvester R, Sternberg C: The role of adjuvant combination chemotherapy after cystectomy in locally advanced bladder cancer: What we do not know and why. Ann Oncol 11: 851-856, 2000
Studer UE, Bacchi M, Biedermann C, et al: Adjuvant cisplatin chemotherapy following cystectomy for bladder cancer: Results of a prospective randomized trial. J Urol 152: 81-84, 1994
Pianezza O, Meneguelo M, Merlo F, et al: Adjuvant chemotherapy in locally advanced bladder cancer: Long-term follow-up of a multicenter study. J Urol 169:337, 2003 (abstr 1305)
Otto T, B?rgermann C, Krege S, et al: Adjuvant chemotherapy in locally advanced bladder cancer (pT3/pT4a,pN1-2,M0): A phase III study. Eur Urol 39:147, 2001 (abstr 577)
Sternberg CN, Yagoda A, Scher HI, et al: Methotrexate, vinblastine, doxorubicin, and cisplatin for advanced transitional cell carcinoma of the urothelium: Efficacy and patterns of response and relapse. Cancer 64: 2448-2458, 1989
Takahashi A, Tsukamoto T, Tobisu K, et al: Radical cystectomy for invasive bladder cancer: Results of multi-institutional pooled analysis. Jpn J Clin Oncol 34: 14-19, 2004
Mills RD, Turner WH, Fleischmann A, et al: Pelvic lymph node metastases from bladder cancer: Outcome in 83 patients after radical cystectomy and pelvic lymphadenectomy. J Urol 166: 19-23, 2001
Frank I, Cheville JC, Blute ML, et al: Transitional cell carcinoma of the urinary bladder with regional lymph node involvement treated by cystectomy: Clinicopathologic features associated with outcome. Cancer 97: 2425-2431, 2003
Stein JP, Cai J, Groshen S, et al: Risk factors for patients with pelvic lymph node metastases following radical cystectomy with en bloc pelvic lymphadenectomy: Concept of lymph node density. J Urol 170: 35-41, 2003
Leissner J, Hohenfellner R, Thüroff JW, et al: Lymphadenectomy in patients with transitional cell carcinoma of the urinary bladder; significance for staging and prognosis. BJU Int 85: 817-823, 2000
Herr HW: Extent of surgery and pathology evaluation has an impact on bladder cancer outcomes after radical cystectomy. Urology 61: 105-108, 2003
Poulsen AL, Horn T, Steven K: Radical cystectomy: Extending the limits of pelvic lymph node dissection improves survival for patients with bladder cancer confined to the bladder wall. J Urol 160: 2015-2019, 1998
von der Maase H, Hansen SW, Roberts JT, et al: Gemcitabine and cisplatin versus methotrexate, vinblastine, doxorubicin, and cisplatin in advanced or metastatic bladder cancer: Results of a large, randomized, multinational, multicenter, phase III study. J Clin Oncol 18: 3068-3077, 2000(Jan Lehmann, Margitta Ret)
Division of Biostatistics, Central Institute of Mental Health Mannheim/University of Heidelberg, Heidelberg, Germany
ABSTRACT
PURPOSE: Radical cystectomy as standard treatment of muscle-invasive urothelial carcinoma of the urinary bladder cures less than 50% of patients with locally advanced bladder cancer. We compared two adjuvant combination chemotherapies in patients with stage pT3a-4a and/or pathologic node-positive transitional-cell carcinoma of the bladder after radical cystectomy.
PATIENTS AND METHODS: A total of 327 patients were randomly assigned to either adjuvant systemic chemotherapy with three cycles of cisplatin 70 mg/qm2 on day 1 and methotrexate 40 mg/qm2 on days 8 and 15 of a 21-day cycle (CM) or three cycles of methotrexate 30 mg/qm2 on days 1, 15, and 22, vinblastine 3 mg/qm2 on days 2, 15, and 22, epirubicin 45 mg/qm2 on day 2, and cisplatin 70 mg/qm2 on day 2 of a 28-day cycle (M-VEC).
RESULTS: The hazard ratio for progression-free survival as the primary end point was 1.13 (90% CI, 0.86 to 1.48) for 163 CM patients compared with 164 M-VEC patients whose right-hand limit remained below the upper bound compatible with the noninferiority hypothesis ( = .0403). The 5-year progression-free, tumor-specific, and overall survival rates (point estimates ± SE) for CM versus M-VEC were 46.3% ± 4.6% v 48.8% ± 4.5%, 52.0% ± 4.6% v 52.3% ± 4.8%, and 46.1% ± 4.3% v 45.1% ± 4.6%, respectively. WHO grade 3 and 4 leukopenia occurred in 7.0% of patients treated with CM and 22.2% of patients treated with M-VEC (P < .0001).
CONCLUSION: CM cannot be considered inferior to M-VEC with regard to progression-free survival of patients with locally advanced bladder cancer after radical cystectomy. Moreover, patients receiving adjuvant CM combination therapy experienced significantly less grade 3 and 4 leukopenia than patients treated with M-VEC.
INTRODUCTION
Radical cystectomy is recommended as curative treatment for locally advanced bladder cancer including tumor stages pT3 and pT4a and/or involvement of regional lymph nodes. However, more than half of these patients experience relapse, with distant metastasis being the predominant form of disease recurrence. Therefore, combination chemotherapy has been investigated as neoadjuvant or adjuvant adjunctive treatment in numerous trials. A recent meta-analysis on neoadjuvant chemotherapy for invasive bladder cancer described a modest but clear improvement in survival, encouraging the use of platinum-based combination therapy.1 For the adjuvant approach, three randomized trials have suggested a relapse-free survival improvement for patients receiving combination chemotherapy after radical cystectomy compared with patients undergoing surgery alone.2-4
The combination regimen of methotrexate, vinblastine, doxorubicin, and cisplatin (M-VAC), which was first reported as a palliative approach for advanced metastatic urothelial carcinoma in 1985,5 has succeeded as a gold standard treatment in randomized trials.6,7 The first randomized trial comparing adjuvant M-VAC or methotrexate, vinblastine, epirubicin, and cisplatin (M-VEC) to radical cystectomy alone found a significant progression-free survival improvement of 50% for the adjuvant group 40 months after surgery.3 Despite the efficacy of M-VAC or M-VEC combination therapy, severe hematologic and nonhematologic toxicity and a treatment-related mortality rate of up to 4% have caused concern.
The present study was initiated in 1994 as a noninferiority trial of a reduced chemotherapy regimen consisting of cisplatin plus methotrexate (CM) compared with conventional M-VEC, with the intention of finding a less toxic but effective regimen for the adjuvant treatment of patients with resected locally advanced bladder cancer with no substantial loss of efficacy.
PATIENTS AND METHODS
The study protocol was reviewed and approved by the committee of the "Arbeitsgemeinschaft Urologische Onkologie" of the German Cancer Society as well as local ethics committees of participating centers.
Eligibility
Patients eligible for the randomization process had to meet the following inclusion criteria. Disease status after radical cystectomy must be histologically confirmed stage pT3a-4a and/or pathologic node-positive transitional-cell carcinoma of the bladder (1997 TNM classification8). Squamous cell carcinoma and/or adenocarcinoma components were allowed if transitional-cell carcinoma was present. Radical cystectomy denoted the removal of the entire bladder, prostate, and seminal vesicles in men and removal of anterior pelvic organs in the female, including, if indicated, a portion of the anterior vagina. The method of urinary diversion was left to the discretion of the investigator. Bilateral pelvic lymph node dissection was a prerequisite for correct staging and included a full dissection of the lymph nodes bordered by the internal iliac arteries, external iliac arteries, and the pelvic floor bilaterally including obturator nodes. Any evidence of macroscopic or microscopic incomplete resection (tumor-positive margin of the specimen or grossly enlarged unresected lymph nodes) was considered an exclusion criterion. Distant metastasis was excluded by preoperative staging, including at least chest x-ray and abdominal ultrasound. Furthermore, patients between 18 and 70 years old were included if they had a Karnofsky performance score of at least 80 with adequate renal and liver function as well as bone marrow reserve (measured creatinine clearance 70 mL/min, serum creatinine 1.3 mg/dL, bilirubin < 1.5 mg/dL, ALT or AST < 60 U/L, WBC count 3.0 x 109/L, platelets 150 x 109/L, and hemoglobin 10 g/dL). Local intravesical adjuvant chemotherapy or immunotherapy before radical cystectomy was allowed. Furthermore, a written informed consent according to institutional and federal guidelines had to be signed by the patient. Patients were ineligible if the time interval between radical cystectomy and the first day of chemotherapy exceeded 4 months and if they had used any investigational agent in the months before enrollment onto the study.
Treatment Plan
The CM regimen consisted of cisplatin 70 mg per square meter of body-surface area (mg/qm2) on day 1 and methotrexate 40 mg/qm2 on days 8 and 15 every 21 days for a maximum of three cycles. M-VEC (methotrexate 30 mg/qm2 on days 1, 15, and 22; vinblastine 3 mg/qm2 on days 2, 15, and 22; epirubicin 45 mg/qm2 on day 2, and cisplatin 70 mg/qm2 on day 2) was administered every 28 days for a maximum of three cycles.
The following dose adjustments were planned. Methotrexate was omitted if the leukocyte count was less than 2.5 x 109/L or the thrombocyte count was less than 100 x 109/L. The beginning of a new cycle was delayed until the leukocyte count was greater than 3.0 x 109/L and the thrombocyte count was greater than 100 x 109/L. Doses of cisplatin and methotrexate were reduced by 50% if measured creatinine clearance was 50 to 70 mL/min, or the drugs were omitted if the creatinine clearance was less than 50 mL/min. Patients were to receive full supportive care. Use of growth factors was recommended only in case of leukopenic fever, dose omissions for methotrexate and vinblastine as a result of leukopenia, or delay of a consecutive cycle as a result of leukopenia. Hematologic and nonhematologic toxicity was graded according to the WHO grading system. Study drug therapy was discontinued if there was evidence of progressive disease under therapy, if the attending physician considered a change of therapy to be in the best interest of the patient, if the patient requested discontinuation, or if the drug exhibited unacceptable toxicity.
Follow-Up of Patients
Follow-up of both treatment groups occurred at 3-month intervals for 2 years, then at 6-month intervals for 3 years, and yearly thereafter. Follow-up consisted of physical examination, abdominal ultrasound and chest radiography, and computed tomography scanning or bone scanning as clinically indicated. The site and date of the first relapse, as well as the date and the reason in case of death, were recorded. Relapse was defined as the detection of at least one suspicious lesion that could not be identified by histologic, clinical, or laboratory criteria as an independent secondary malignancy. In case of relapse, it was left to the discretion of the patient and the attending physician whether further therapy, such as second-line chemotherapy, radiation, or surgical treatment, was initiated.
Statistical Methods
The primary basis for the statistical planning and analysis of the trial was the noninferiority version of the log-rank test for equivalence of two survivor functions9 satisfying an ordinary proportional hazards model in the sense of Cox.10 The test was applied with an equivalence margin of = .15, which gives the maximum tolerable difference of progression-free survival probabilities to be expected under both treatments for the same time since cystectomy. Practical implementation of the procedure was performed by estimating the regression coefficient ? of a proportional hazards model, with the indicator of treatment group CM as the only covariate, and comparing the right-hand limit of a 90% CI for the hazard ratio (HR) = e? to 1.5077 as a critical upper bound. For the purpose of checking the adequacy of the proportional hazards assumption, the procedure implemented by the S-plus (Statistical Sciences, Seattle, WA) function cox.zph was used (for a detailed description of the statistical rationale behind it, see Therneau and Grambsch11). Because the P value obtained from this test turned out to be as large as .45, there was no reason to question the appropriateness of the model.
However, estimation of median progression-free survival time was based on a fully parametric model because both observed survivor functions dropped below 50% so late that the number of patients still at risk was too small for providing a sufficiently accurate estimate of the conditional probability of surviving the interval containing the observed median. Of the standard parametric models, the log-normal was the model that showed the best fit to the data. Accordingly, confidence limits for the median progression-free survival time were computed by transforming those for the expected value of the distribution of log T.
In addition to progression-free survival time, which had been chosen as the primary end point for the confirmatory analysis of this trial, the same techniques were also applied to the data obtained for the length of survival and time until death from the specific tumor. All eligible patients who received any adjuvant treatment were included in the analysis of possible toxic effects. The incidences of WHO grade 3 and 4 toxicities were compared using Fisher's exact test for homogeneity of two binomial distributions.
RESULTS
Between January 1, 1994, and September 30, 2000, 335 patients were registered onto the trial (Fig 1). Eight patients (2.4%) were deemed ineligible because inclusion criteria were not met (age, renal function, metastatic disease at time of radical cystectomy, secondary malignancy, and withdrawal of consent). The remaining 327 patients were equally assigned to adjuvant systemic chemotherapy with CM or M-VEC by randomization after having signed informed consent.
Demographic factors were similar between the two groups (Table 1). Median age was 60.5 years, and the ratio of males to females was almost 4:1. More than half of the patient group (56.6%) had tumor-positive regional lymph nodes. The median follow-up time for patients living free of relapse at the time of analysis was 42 months.
Treatment
The median time from surgery to chemotherapy was 56 days in both arms. Of the 163 patients assigned to receive adjuvant CM, 123 (75.5%) completed three cycles as planned (Table 2); 12.9% (21 of 163 patients) discontinued treatment early because of toxic effects or for other reasons. Nineteen patients (11.7%) declined treatment, and four patients (2.5%) died during the course of treatment. One patient died of cardiac failure (day 18 of a third cycle), two patients experienced severe bone marrow toxicity after two cycles, and one patient developed lethal pneumonia after one cycle.
Of the 164 patients assigned to the M-VEC arm, 113 (68.9%) completed three cycles; 21.4% (35 of 164 patients) stopped treatment early. Sixteen patients (9.8%) declined treatment, and one patient (0.6%) died during treatment with M-VEC as a result of cardiac dysfunction. All patients were analyzed as belonging to the treatment arm assigned by randomization, adopting the intent-to-treat principle.
Surgical Procedures
The only surgery-related requirements for eligibility were radical cystectomy with curative intent and en bloc resection of the tumor with negative margins, as well as absence of metastatic or unresected transitional-cell carcinoma. However, because patients were usually identified postoperatively, specific surgical and histopathologic procedures could not be required.
Toxicity
Hematologic toxic effects predominated. The most common hematologic toxic effect was leukopenia, with a significantly higher rate in patients treated with M-VEC compared with CM (22.2% v 7.0%, respectively). Except for a significant difference in grade 3 alopecia in favor of CM (1.9% in CM v 24.7% in M-VEC), no difference was observed for other hematologic or nonhematologic factors. Four patients (2.5%) died as a result of a toxic effect attributed to CM chemotherapy, and one patient (0.6%) died as a result of therapy in the M-VEC arm. This difference of therapy-related mortality was not statistically significant (Fisher's exact test, P = .371). The toxic effects classified as WHO grade 3 or higher that occurred among treated patients are listed in Table 3.
Progression-Free Survival
The median follow-up period for patients living free of progression was greater than 3 years. Tumor relapse was reported in 46.0% of the patients (75 of 163 patients) in the CM group and 44.5% of patients (73 of 164 patients) in the M-VEC group. The median duration of progression-free survival was 43.4 months in the CM arm and 49.7 months in the M-VEC arm by intent-to-treat analysis (Table 4). The 5-year progression-free survival rates were 46.3% in the CM treatment group and 48.8% in the M-VEC group (Fig 2). The HR for disease progression in the CM group, as compared with the M-VEC group, was 1.13 (90% CI, 0.86 to 1.48; P = .0403).
In the subgroup analyses of patients with lymph node–positive disease, the median progression-free survival was 36.2 months (90% CI, 27.5 to 47.7 months) in the CM group and 32.4 months (90% CI, 24.8 to 42.3 months) in the M-VEC group. The 5-year progression-free survival rates of patients with lymph node–positive disease were 41.9% in the CM group and 40.1% in the M-VEC group (Fig 3). The HR for disease progression in the CM group, as compared with the M-VEC group, was 0.95 (90% CI, 0.67 to 1.33; P = .0126).
In the analyses of patients with tumor extension beyond the bladder wall but without lymph node involvement, the median progression-free survival was 52.0 months (90% CI, 36.5 to 74.2 months) in the CM arm and 87.7 months (90% CI, 59.8 to 128.6 months) in the M-VEC arm. The 5-year progression-free survival rates were 51.5% in the CM group and 60.8% in the M-VEC group (Fig 4). The HR for disease progression in the CM group, as compared with the M-VEC group, was 1.46 (90% CI, 0.94 to 2.29; P = .4571).
Tumor-Specific and Overall Survival
Sixty-five patients in the CM treatment group (39.9%) and 62 patients in the M-VEC treatment group (37.8%) died of tumor disease during the follow-up period. The median duration of tumor-specific survival was 57.3 months in the CM group and 63.3 months in the M-VEC group (Fig 5). The HR for tumor-related death in the CM group, as compared with the M-VEC group, was 1.13 (90% CI, 0.84 to 1.51; P = .0528).
The 5-year rates of tumor-specific survival were 52.0% in the CM group and 52.3% in the M-VEC group (Table 4, Fig 5). In the CM and M-VEC arms, 14 patients (8.6%) and 15 patients (9.1%) died as a result of causes not related to tumor, respectively. The median duration of overall survival was 47.1 months in the CM group and 51.8 months in the M-VEC group (Table 4, Fig 6). The HR for death irrespective of cause in the CM group, as compared with the M-VEC group, was 1.10 (90% CI, 0.88 to 1.44; P = .0255). The 5-year rates of overall survival were 46.1% in the CM group and 45.1% in the M-VEC group.
Site of Tumor Relapse
Classification of first relapses with regard to site resulted as follows (Table 5). Local recurrence only occurred in 12 patients in the CM group (7.3%) and seven patients in the M-VEC group (4.3%). Lymphatic relapse only, which was typically encountered as retroperitoneal lymph node disease, was reported in 4.3% of CM patients (seven of 163 patients) and 6.1% of M-VEC patients (10 of 164 patients). Visceral metastases (without bone metastases) were found in 20.2% of patients in the CM group (33 of 163 patients) and 15.9% of patients in the M-VEC group (26 of 164 patients); at least 12.3% of CM patients (20 of 163 patients) and 15.2% of M-VEC patients (25 of 164 patients) developed bone metastases.
DISCUSSION
The frequent occurrence of tumor relapses after radical cystectomy for locally advanced bladder cancer, the favorable results of three former adjuvant trials,2-4 and the limited tolerability of adjuvant M-VEC or M-VAC combination chemotherapy provided the rationale for the present study. After introducing effective M-VAC combination chemotherapy in urothelial cancer in 1985,5 two successive, randomized, phase III trials demonstrated the superiority of the M-VAC regimen in advanced urothelial cancer compared with single-agent cisplatin7,12 or with cisplatin, cyclophosphamide, and doxorubicin combination therapy.6 Because of the pronounced hematologic and nonhematologic toxicity with M-VAC or M-VEC combination therapy, other potentially less toxic combination therapies, such as CM, have been investigated in advanced urothelial cancer.13,14
Neoadjuvant administration of chemotherapy has been exceedingly investigated compared with the adjuvant approach for patients with a high risk of relapse undergoing definite treatment by surgery or radiation. Although the majority of individual neoadjuvant studies was not able to demonstrate a significant survival advantage for pre-emptive chemotherapy compared with definitive treatment alone, a recent meta-analysis based on individual patient data from 10 randomized neoadjuvant trials was able to demonstrate a significant 13% reduction in the risk of death (HR = 0.87; 95% CI, 0.78 to 0.98; P = .016), which was equivalent to a 5% absolute overall survival benefit at 5 years from 45% to 50%.1 This meta-analysis also stated that a significant advantage exists using platinum-based combination chemotherapy compared with single-agent platinum. The most recent neoadjuvant study, which was not included in the former meta-analysis, achieved a survival advantage with borderline significance for 154 patients assigned to three cycles of neoadjuvant M-VAC combination therapy versus 153 patients randomly assigned to surgery alone (median survival, 77 months in the neoadjuvant arm v 46 months in the surgery-only arm; P = .06).15
Whether neoadjuvant or adjuvant application of systemic chemotherapy for locally advanced bladder cancer is the preferable treatment modality remains a matter of debate. To date, only one study has investigated the neoadjuvant and adjuvant treatment strategy directly.16 Similar to our experience, 40% of patients in this trial with histologically proven regional lymph node metastasis experienced long-term progression-free survival in both treatment arms.
In contrast to neoadjuvant administration of chemotherapy, proponents of adjuvant chemotherapy for locally advanced bladder cancer emphasize optimal patient selection based on exact assessment of histopathologic tumor stage. This allows the selection of patients with the highest risk for relapse, and only minimal occult tumor burden after radical cystectomy needs to be treated. Subsequently, improvement of recurrence-free survival by adjuvant combination chemotherapy was corroborated by three individual randomized controlled studies from Germany and the United States for the adjuvant administration of combination chemotherapy after radical cystectomy.2-4
Skinner et al2 randomly assigned 91 locally advanced bladder cancer patients to radical cystectomy plus four cycles of adjuvant chemotherapy (predominantly cisplatin, cyclophosphamide, and doxorubicin combination therapy) versus surgery alone. The trial was stopped because of an interim analysis that demonstrated a significant disease-free survival advantage for the adjuvant treatment arm 5 years after cystectomy (51% for adjuvant chemotherapy v 34% for surgery alone, P < .011). St?ckle et al3 terminated their study after an interim analysis of 49 patients demonstrated a significant disease-free survival difference of 50% at 3.5 years in favor of 26 patients receiving adjuvant treatment with M-VAC or M-VEC compared with the control group (63% for adjuvant chemotherapy v 13% for cystectomy alone; P = .0005).
A third randomized trial with a favorable result for adjuvant combination chemotherapy performed at Stanford University compared adjuvant cisplatin, methotrexate, and vinblastine combination chemotherapy to cystectomy alone.4 In contrast to the former two studies, deferred chemotherapy on progression was suggested by the protocol for patients assigned to the nonadjuvant arm. This study was also closed before accrual of the preplanned number of patients because of a significant difference in progression-free survival in favor of patients receiving adjuvant chemotherapy. Median time to progression for the 25 patients assigned to the adjuvant treatment arm was 37 months versus 12 months in the cystectomy-alone group (P < .01), whereas overall survival was not significantly different. Criticisms of all three former trials have been summarized by Sylvester and Sternberg17 and address small sample size, early stopping of patient entry and premature closure, statistical analyses, reporting of results, and drawing conclusions.
Adjuvant chemotherapy trials for invasive bladder cancer that did not find a significant difference for patients receiving adjuvant chemotherapy compared with a surgery-only group have been reported from Switzerland,18 Italy,19 and Germany.20 This may be attributed to the exclusion of lymph node–positive patients19 and to the administering of only single-agent cisplatin as adjuvant treatment to a predominantly pT2pN0 population.18
Only limited information on the degree of toxicity has been revealed from any of these adjuvant trials. Reports on palliative M-VAC or M-VEC combination chemotherapy causing grade 3 to 4 myelosuppression in more than half of the patients, a 25% incidence of nadir sepsis, and drug-related deaths in at least 3% of patients 21 prompted investigators to search for more tolerable but still active regimens. Efficacy of CM combination therapy as implemented for the current trial had been reported previously for 53 patients with advanced urothelial cancer.13 Although the majority of patients had dose omissions or delays, an overall response rate of 46% was noted. The standard M-VEC treatment arm of the current trial incorporated epirubicin instead of doxorubicin, as in the original M-VAC schema, because cardiotoxicity was of concern at the time of drafting the protocol.
Because of the nature of the current trial testing for noninferior efficacy of the reduced combination (CM) compared with the full combination (M-VEC), the results may only lend support to but not prove the superiority of adjuvant combination chemotherapy compared with surgery alone. Remarkably, the 5-year progression-free survival rate of greater than 40% for lymph node–positive disease after radical cystectomy plus adjuvant combination chemotherapy (CM or M-VEC) represents an average standard of care of 40 uro-oncologic centers in Germany. Retrospective studies of patients with lymph node–positive bladder cancer receiving adjuvant chemotherapy after radical cystectomy versus surgery alone reported significant 5-year overall survival advantages of 43% v 17% (P = .03)22 and 45% v 21% (P = .031)23 for the adjuvant group, respectively. A similar survival comparison from the Mayo Clinic demonstrated a 5-year tumor-specific survival advantage of 55% v 32% (P = .005) in favor of the adjuvant chemotherapy group compared with surgery alone, respectively,24 notwithstanding the limitation that any retrospective result may be biased by patient selection. Finally, the recurrence-free survival rate of 41% at 5 years for all lymph node–positive patients (n = 185) in our trial is slightly higher than 39% rate that was recently reported from another North American center of excellence administering adjuvant chemotherapy to 139 patients with lymph node–positive disease.25
When assessing the possible effects of baseline factors (eg, age, sex, tumor stage, blood cell counts, renal and liver function, and size of recruiting center) on the risk of progression using multivariate Cox regression analysis, the only significant predictor was the number of tumor-positive lymph nodes (data not shown). The current trial did not control for either the number of lymph nodes removed during radical cystectomy or the extent of pelvic lymphadenectomy because patients were usually identified postoperatively to be eligible for this trial. Therefore, a specific surgical and histopathologic protocol could not be required. Some authors have suggested a survival improvement for patients undergoing extensive lymphadenectomy with cystectomy in a retrospective survey,25-27 whereas others have seen a significant survival advantage with extensive lymphadenectomy only in patients with organ confined ( pT2) bladder cancer.28 Future trials on adjuvant chemotherapy for locally advanced bladder cancer may focus on this issue with more precision.
As previously reported for the palliative indication of combination chemotherapy in unresectable advanced urothelial cancer, the gold standard status of M-VAC therapy has declined.29 Results of a large, multinational, phase III trial comparing M-VAC with gemcitabine plus cisplatin combination therapy showed similar response rates and comparable overall survival in 405 patients. Although patient numbers and study design of this trial did not allow for the testing of significant noninferiority, gemcitabine plus cisplatin has become a well-established regimen for the treatment of advanced bladder cancer because of significantly less clinically relevant toxicity and better cost effectiveness.
The implications of our results must be assessed with a view to the equivalence margin of 0.15, which had been specified in the protocol for the difference between the respective survivor functions at any point of time. Thus, the major statistical hypothesis that could be established by means of our data does not rule out the possibility that the full combination therapy comprising all four drugs is more efficient than the reduced chemotherapy regimen. But we can be certain that this advantage, which is supported by the positive signs of the point estimates obtained from the total samples, fails to be of significant clinical relevance. For the adjuvant indication of combination chemotherapy with curative intent in resected locally advanced bladder cancer, the current report provides evidence that the efficacy of the less toxic regimen CM cannot be considered substantially inferior to the efficacy of M-VEC.
Appendix
The members of the Collaborative Group were as follows: Secretariat: M. St?ckle (principal investigator), S. Wellek (biostatistician), and J. Lehmann and M. Retz, C.Wiemers (data managers). Investigators (the number of patients enrolled onto the study is shown in parentheses): C. Wiemers, D. Sternberg, L. Franzaring, M. St?ckle, J. Thüroff, J. Beck, C. Huber, Johannes-Gutenberg-Universit?t, Mainz (n = 37); C. Weining, H.J. Piechota, L. Hertle, Universit?tsklinikum Münster, Münster (n = 32); G. Steiner, P. Albers, S.C. Müller, Universit?tsklinikum Bonn, Bonn (n = 28); C. Vierneisel, A. G?ll, M. Beitzinger, D. Frohneberg, St?dtisches Klinikum, Karlsruhe (n = 26); T. Becker, T. K?lble, H. Riedmiller, Philipps-Universit?t, Marburg (n = 22); H.J. Knopf, G. Engelhardt, P.-J. Funke, Evangelisches-Jung-Stilling Krankenhaus, Siegen (n = 21); J. Roloff, T. Sch?nfeld, P. Walz, Klinikum Lüdenscheid, Lüdenscheid (n = 14); C. Wiemers, M. Retz, J. Lehmann, A. Bannowsky, M. St?ckle, Christian-Albrechts-Universit?t, Kiel (n = 13); S. Langbein, F. Reiher, E.P. Allhoff, Otto-von-Guericke-Universit?t, Magdeburg (n = 13); M. Schiller, D. Sternberg, G. Hutschenreiter, Evangel. und Johanniter Klinikum, Oberhausen (n = 13); M. Staehler, M. Müller, K. Miller, Universit?tsklinikum Benjamin-Franklin, Berlin (n = 12); U. Gertenbach, U. Kaldenbach, J. Thüroff, S. Roth, Universit?t Witten-Herdecke, HELIOS Klinikum Wuppertal (n = 11); S. Tedsen, A. B?hle, D. Jocham, Universit?t Lübeck, Lübeck (n = 7); W. Kieser, H. Riedmiller, Julius-Maximilian-Universit?t, Würzburg (n = 6); G.M. Wacker-Backhaus, G. Ostendorf, C. Ehlert, H. Becker, Marienkrankenhaus, Hamburg (n = 6); L. Weissbach, Krankenhaus am Urban, Berlin (n = 6); C. Charvalakis, L. Lampante, St?dtisches Krankenhaus Kemper Hof, Koblenz (n = 6); V. Poulakis, E. Becht, S. Al-Batran, A. Knuth, E. Jaeger, Krankenhaus Nordwest, Frankfurt am Main (n = 6); R. Groh, R. Horsch, Kreiskrankenhaus Offenburg, Offenburg (n = 5); N. Vennemann, R.G. Anding, H. van Ahlen, Klinikum Osnabrück, Osnabrück (n = 5); T. K?lble, St?dtisches Klinikum Fulda, Fulda (n = 5); B. Buhmann, R. Hausmann, F. Pinkenburg, Kreiskrankenhaus Rendsburg, Rendsburg (n = 4); J. Gschwend, R. Hautmann, Universit?t Ulm, Ulm (n = 4); N.W. Fischer, D. Neisius, Krankenhaus der Barmherzigen Brüder, Trier (n = 3); K. Murr, M. Kühn, Johanniter-Krankenhaus, Stendal (n = 3); M.G. Friedrich, H. Huland, Universit?tsklinikum Eppendorf, Hamburg (n = 2); S. Siemer, U. Humke, Universit?t des Saarlandes, Homburg/Saar (n = 2); J. Steffens, St Antonius Hospital, Eschweiler (n = 2); D. Sternberg, M. Beer, Franziskus Krankenhaus, Berlin (n = 2); R. Nitze, R. Sintermann, Evangelisches Krankenhaus "Lutherhaus," Essen (n = 2); A. Kuczyk, U. Jonas, Medizinische Hochschule, Hannover (n = 2); S. Kiani, U. Tunn, St?dtische Kliniken, Offenbach (n = 2); G. Müller, E. T?lle, Herz-Jesu-Krankenhaus, Münster-Hiltrup (n = 2); L. Neubauer, St Franziskus Hospital, Lohne (n = 1); T. Drechsler, M. Hartmann, Hamburg Bundeswehr Krankenhaus (n = 1); U. Gertenbach, Allgemeines Krankenhaus, Hagen (n = 1); U. Engelmann, Universit?t K?ln (n = 1); D. Unverferth, Kreiskrankenhaus, Aurich (n = 1); E. Winter, P. Bub, Klinikum Schwerin (n = 1); G. Block, H.E. Reichert, Kreiskrankenhaus V?lklingen (n = 1); A. Nonnenmacher, Diakonie-Krankenhaus, Schw?bisch-Hall (n = 1); W. Rulf, W. Langhorst, Onkologische Praxis, Erkrath (n = 1); C.J. Güdemann, G. Staehler, Ruprecht-Karls-Universit?t, Heidelberg (n = 1).
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank the patients as well as the clinicians and practitioners who participated in this trial. We also thank the staff at the central trial offices in Homburg/Saar, Germany (Amelie Teuchert, Irene Lehmann, and Vera Mathei-Mayer), and in Kiel, Germany (Urte Fischer-Nürnberg, Astrid Hallen, Cora Prien, and Birgit Küppers).
NOTES
Supported in part by the Gunther-Voges Gesellschaft and Stiftung Berliner Bank, Germany (neither sponsor had any influence on study design or on the collection, analysis, or interpretation of data; and neither had any influence in the writing of the report or the decision to submit the manuscript for publication).
Presented at the following meetings: 54th Congress of the German Society of Urology in Wiesbaden, Germany, September 18-21, 2002; XVIIIth Congress of the European Association of Urology in Madrid, Spain, March 12-15, 2003; 98th Annual Meeting of the American Urological Association Chicago, IL, April 26-May 1, 2003; and the 39th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, May 31-June 3, 2003.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Advanced Bladder Cancer Meta-Analysis Collaboration: Neoadjuvant chemotherapy in invasive bladder cancer: A systematic review and meta-analysis.Lancet 361: 1927–1934, 2003
Skinner DG, Daniels JR, Russell CA, et al: The role of adjuvant chemotherapy following cystectomy for invasive bladder cancer: A prospective comparative trial. J Urol 145: 459-464, 1991
St?ckle M, Meyenburg W, Wellek S, et al: Advanced bladder cancer (stages pT3b, pT4a, pN1 and pN2): Improved survival after radical cystectomy and 3 adjuvant cycles of chemotherapy—Results of a controlled prospective study. J Urol 148: 302-306, 1992
Freiha F, Reese J, Torti FM: A randomized trial of radical cystectomy versus radical cystectomy plus cisplatin, vinblastine and methotrexate chemotherapy for muscle invasive bladder cancer. J Urol 155: 495-499, 1996
Sternberg CN, Yagoda A, Scher HI, et al: Preliminary results of M-VAC (methotrexate, vinblastine, doxorubicin and cisplatin) for transitional cell carcinoma of the urothelium. J Urol 133: 403-407, 1985
Logothetis CJ, Dexeus FH, Finn L, et al: A prospective randomized trial comparing MVAC and CISCA chemotherapy for patients with metastatic urothelial tumors. J Clin Oncol 8: 1050-1055, 1990
Loehrer PJ Sr, Einhorn LH, Elson PJ, et al: A randomized comparison of cisplatin alone or in combination with methotrexate, vinblastine, and doxorubicin in patients with metastatic urothelial carcinoma: A cooperative group study. J Clin Oncol 10: 1066-1073, 1992
Sobin LH, Wittekind CH: UICC TNM Classification of Malignant Tumors (ed 4). New York, NY, John Wiley & Sons, Inc, 1997
Wellek S: Testing Statistical Hypotheses of Equivalence. Boca Raton, FL, Chapman & Hall/CRC, 2003
Cox DR: Regression models and life tables with discussion. J R Stat Soc B 34: 187-220, 1972
Therneau T, Grambsch P: Modeling Survival Data: Extending the Cox Model. Berlin, Germany, Springer, 2000
Saxman SB, Propert KJ, Einhorn LH, et al: Long-term follow-up of a phase III intergroup study of cisplatin alone or in combination with methotrexate, vinblastine, and doxorubicin in patients with metastatic urothelial carcinoma: A cooperative group study. J Clin Oncol 15: 2564-2569, 1997
Stoter G, Splinter TA, Child JA, et al: Combination chemotherapy with cisplatin and methotrexate in advanced transitional cell cancer of the bladder. J Urol 137: 663-667, 1987
Hillcoat BL, Raghavan D, Matthews J, et al: A randomized trial of cisplatin versus cisplatin plus methotrexate in advanced cancer of the urothelial tract. J Clin Oncol 7: 706-709, 1989
Grossman HB, Natale RB, Tangen CM, et al: Neoadjuvant chemotherapy plus cystectomy compared with cystectomy alone for locally advanced bladder cancer. N Engl J Med 349: 859-866, 2003
Millikan R, Dinney C, Swanson D, et al: Integrated therapy for locally advanced bladder cancer: Final report of a randomized trial of cystectomy plus adjuvant M-VAC versus cystectomy with both preoperative and postoperative M-VAC. J Clin Oncol 19: 4005-4013, 2001
Sylvester R, Sternberg C: The role of adjuvant combination chemotherapy after cystectomy in locally advanced bladder cancer: What we do not know and why. Ann Oncol 11: 851-856, 2000
Studer UE, Bacchi M, Biedermann C, et al: Adjuvant cisplatin chemotherapy following cystectomy for bladder cancer: Results of a prospective randomized trial. J Urol 152: 81-84, 1994
Pianezza O, Meneguelo M, Merlo F, et al: Adjuvant chemotherapy in locally advanced bladder cancer: Long-term follow-up of a multicenter study. J Urol 169:337, 2003 (abstr 1305)
Otto T, B?rgermann C, Krege S, et al: Adjuvant chemotherapy in locally advanced bladder cancer (pT3/pT4a,pN1-2,M0): A phase III study. Eur Urol 39:147, 2001 (abstr 577)
Sternberg CN, Yagoda A, Scher HI, et al: Methotrexate, vinblastine, doxorubicin, and cisplatin for advanced transitional cell carcinoma of the urothelium: Efficacy and patterns of response and relapse. Cancer 64: 2448-2458, 1989
Takahashi A, Tsukamoto T, Tobisu K, et al: Radical cystectomy for invasive bladder cancer: Results of multi-institutional pooled analysis. Jpn J Clin Oncol 34: 14-19, 2004
Mills RD, Turner WH, Fleischmann A, et al: Pelvic lymph node metastases from bladder cancer: Outcome in 83 patients after radical cystectomy and pelvic lymphadenectomy. J Urol 166: 19-23, 2001
Frank I, Cheville JC, Blute ML, et al: Transitional cell carcinoma of the urinary bladder with regional lymph node involvement treated by cystectomy: Clinicopathologic features associated with outcome. Cancer 97: 2425-2431, 2003
Stein JP, Cai J, Groshen S, et al: Risk factors for patients with pelvic lymph node metastases following radical cystectomy with en bloc pelvic lymphadenectomy: Concept of lymph node density. J Urol 170: 35-41, 2003
Leissner J, Hohenfellner R, Thüroff JW, et al: Lymphadenectomy in patients with transitional cell carcinoma of the urinary bladder; significance for staging and prognosis. BJU Int 85: 817-823, 2000
Herr HW: Extent of surgery and pathology evaluation has an impact on bladder cancer outcomes after radical cystectomy. Urology 61: 105-108, 2003
Poulsen AL, Horn T, Steven K: Radical cystectomy: Extending the limits of pelvic lymph node dissection improves survival for patients with bladder cancer confined to the bladder wall. J Urol 160: 2015-2019, 1998
von der Maase H, Hansen SW, Roberts JT, et al: Gemcitabine and cisplatin versus methotrexate, vinblastine, doxorubicin, and cisplatin in advanced or metastatic bladder cancer: Results of a large, randomized, multinational, multicenter, phase III study. J Clin Oncol 18: 3068-3077, 2000(Jan Lehmann, Margitta Ret)