|
|
| |||||||||||||||||||||||||||||||||||||||
Background For patients with invasive bladder cancer the usual recommended treatment is radical cystectomy, although transurethral resection of the tumor, systemic chemotherapy, and radiotherapy are each effective in some patients. We sought to determine whether these treatments in combination might be as effective as radical cystectomy and thus might allow the bladder to be preserved and the cancer cured.
Methods We enrolled 53 consecutive patients with muscle-invading bladder cancer (stages T2 through T4, NXM0) in a trial of transurethral surgery, combination chemotherapy, and irradiation (4000 cGy) with concurrent cisplatin administration. Urologic evaluation of the tumor response directed further therapy: radical cystectomy in the 8 patients who had incomplete responses, additional chemotherapy and radiotherapy (6480 cGy) in the 34 patients who had complete responses or who were unsuited for cystectomy, and alternative care in the 11 patients who could not tolerate either irradiation or chemotherapy.
Results After a median follow-up of 48 months, 24 of the 53 patients (45 percent) were alive and free of detectable tumor. In 31 patients (58 percent) the bladder was free of invasive tumor and functioning well, even though in 9 (17 percent) a superficial tumor recurred and required further transurethral surgery and intravesical drug therapy. Of the 28 patients who had complete responses after initial treatment, 89 percent had functioning tumor-free bladders.
Conclusions Conservative combination treatment may be an acceptable alternative to immediate cystectomy in selected patients with bladder cancer, although a randomized clinical trial that included a group for simultaneous comparison would be required to produce definitive results.
We now report (after a median follow-up of 48 months) our results with the use of transurethral resection; systemic multidrug chemotherapy with cisplatin, methotrexate, and vinblastine6; and pelvic irradiation with concurrent administration of cisplatin. Our criteria for selecting patients whose bladders might be preserved included safeguards so that those selected for full chemotherapy and radiotherapy had the highest likelihood of local cure of invasive bladder cancer. Patients had to have a complete response of their primary tumor to the initial combination treatment with chemotherapy and radiotherapy to remain in the study, and only these patients were advised to continue receiving full doses of chemotherapeutic drugs and radiation. All patients who were medically fit for cystectomy and did not have a complete response to the initial therapy were advised to undergo radical cystectomy. This was an attempt to ensure that conservative treatment with bladder preservation would not compromise survival in patients who did not have an immediate complete response and to minimize the possible need for salvage cystectomy after full doses of radiation had been given.
Methods
Selection of Patients
We studied 53 consecutive patients 36 to 87 years old (mean age, 66) with biopsy-confirmed bladder cancer invading muscle (clinical stages T2 through T4, NXM0; see below) who were treated at the Massachusetts General Hospital Cancer Center. The study protocol combined a transurethral resection of the bladder tumor (a complete resection of visible tumor, if it could be safely accomplished); systemic multidrug chemotherapy with the combination methotrexate, vinblastine, and cisplatin; and pelvic external-beam radiation combined with two additional courses of cisplatin (Figure 1). Radical cystectomy was recommended for all patients who had less than a complete response to the initial two cycles of combination chemotherapy plus radiotherapy (4000 cGy) with two additional courses of intravenous cisplatin. Patients who responded completely were given consolidation treatment with additional radiation (2480 cGy) and also received one additional course of intravenous cisplatin.
|
The clinical stage of the primary tumor was T2 in 15 patients, T3 in 29 patients, and T4 in 9 patients. (The following are the clinical stages of bladder cancer: Ta, a papillary carcinoma that does not invade the lamina propria; Tis, carcinoma in situ -- a flat lesion that is not invasive, whose cells are shown to be of high grade on cytologic examination; T1, a tumor infiltrating subepithelial connective tissue [the lamina propria] but not beyond it; T2, a tumor infiltrating muscle, with no palpable mass or induration on bimanual examination after transurethral resection; T3, a tumor invading muscle, with a palpable mass or induration on bimanual examination after transurethral resection [T3a, a tumor invading muscle, but not beyond; T3b, a tumor documented as invading perivesical fat]; and T4, a deeply invasive tumor that is fixed to pelvic bone or that invades neighboring structures).
Pathological review, performed in all 53 patients, confirmed that all had tumors invading the muscularis propria. Fifty-two patients had transitional-cell carcinoma of the bladder, and one had a pure squamous-cell carcinoma.
Chemotherapy
Complete blood counts and blood chemical values, including measurements of creatinine, calcium, magnesium, bilirubin, aspartate aminotransferase, and alkaline phosphatase, were obtained before each dose of cisplatin was given. Complete blood counts were determined weekly during chemotherapy, as were the levels of blood urea nitrogen, creatinine, aspartate aminotransferase, and alkaline phosphatase. Methotrexate (30 mg per square meter of body-surface area) was given on days 0, 14, and 21 of a 28-day cycle, cisplatin (70 mg per square meter) on day 1, and vinblastine (3 mg per square meter) on days 1, 14, and 21. The doses of methotrexate and vinblastine were reduced if severe leukopenia or thrombocytopenia occurred or the serum bilirubin concentration rose above 2.0 mg per deciliter (34 µmol per liter). The doses of methotrexate and cisplatin were reduced if the serum creatinine concentration rose above 1.5 mg per deciliter (130 µmol per liter). The doses of cisplatin and vinblastine were reduced if neuropathy (including a lack of deep tendon reflexes), weakness, or severe paresthesias occurred. The doses of methotrexate were reduced for stomatitis.
Chemotherapy Combined with Radiotherapy
The bladder and pelvic lymphatic systems were treated according to a four-field box technique with carefully contoured fields; a total of 4500 cGy was given over a period of five weeks, in fractions of 180 cGy during each of five treatment sessions per week. Cisplatin (70 mg per square meter) was given the day before radiation therapy started, 21 days later, and again during the consolidation phase if consolidation treatment was administered. Patients selected for consolidation radiotherapy received radiation to whole pelvic fields, for a total dose of 4500 cGy. Therapy in all patients undergoing consolidation treatment was then simulated by introducing radiopaque material into the bladder and rectum to outline the fields to receive the final booster dose of 1980 cGy to the bladder tumor only, for a total dose of 6480 cGy. Diagrams provided by the urologic surgeon indicated the location of the initial tumor. These were used in combination with the initial CT scan in planning which target volumes should receive the booster doses, which in most patients did not include the entire bladder. The booster doses were generally given through small opposed lateral fields. All radiation was delivered by megavoltage beams from linear accelerators generating 10 to 25 MV16.
Criteria for Response and Follow-up Procedures
Urologic evaluation categorized the response of the primary tumor as complete if no tumor was visible on endoscopy or on biopsy of the tumor site and no tumorous cells were found on cytologic examination of the urine. Patients underwent cystoscopy, biopsy of the tumor site, manual examination under anesthesia, and urinary cytologic examination every three months for two years and every six months thereafter. Follow-up pelvic and abdominal CT scans were obtained after the initial, 4000-cGy dose of radiation, three months after the completion of treatment, and then every six months thereafter. At each follow-up evaluation, patients underwent assessment of hematologic indexes, measurement of serum creatinine, and liver-function studies.
Statistical Analysis
The times to the last follow-up evaluation, local recurrence of bladder cancer, appearance of distant metastases, and death were calculated from the date of starting treatment. Life-table probabilities of the control of local disease, survival without distant metastases, and overall survival were determined with the method of Kaplan and Meier, and statistical inferences on actuarial curves with the log-rank test19. Cox regression analysis was used to detect any simultaneous effect of several tumor factors considered to be possible predictors of local tumor control, distant metastases, and overall survival20. Models including two-factor interactions of all possible predictors were tested. The final models were chosen by stepwise elimination of nonsignificant predictor variables.
Results
Forty-two of the 53 patients (79 percent) entered in the study completed it with at most minor deviations from the protocol16. Eleven patients withdrew from the study: six patients could not tolerate the chemotherapy combined with radiotherapy (two had renal dysfunction, and one each had oral ulcers, intolerable incontinence, fatigue, and debilitation), two had cardiopulmonary failure (one had a myocardial infarction, and the other had congestive heart failure), and three refused treatment (two would allow only a partial cystectomy, and one refused any further therapy after undergoing chemotherapy and receiving the initial dose of radiation).
The acute reactions attributable to the chemotherapy included leukopenia (<3000 white cells per cubic millimeter) in 34 percent of the patients, oral ulcers in 24 percent, nausea and vomiting in 73 percent, and diarrhea in 10 percent. There were no deaths attributable to drug toxicity. The acute reactions attributable to chemotherapy combined with radiotherapy included bladder irritation in 32 percent of the patients, diarrhea in 26 percent, fatigue in 21 percent, and leukopenia in 8 percent. Among the 34 patients receiving chemotherapy and the complete dose of radiation (6480 cGy), 2 had transient hematuria; none of these 34 patients had incontinence, symptomatic urinary frequency, or severe injury to the intestine. When questioned at the follow-up examinations after the completion of treatment, all patients reported that their bladder function had remained satisfactory.
Radical cystectomy was performed in 15 patients (4 who could not tolerate the initial chemotherapy combined with radiotherapy, 8 who had incomplete responses after the initial chemotherapy combined with radiotherapy, and 3 who underwent the procedure as salvage therapy after full-dose chemotherapy combined with radiotherapy). Three patients had minor complications of the cystectomy, and one patient died of pelvic sepsis without tumor 13 months after the operation.
Of the 53 patients enrolled in the study, 28 (53 percent) were alive after a median follow-up of 48 months, 24 (45 percent) without evidence of tumor. Distant metastases were diagnosed in 22 patients (42 percent). The actuarial five-year overall survival was 48 percent among all 53 patients, 58 percent among the 42 completing the study, and 14 percent among the 11 who could not complete it (Figure 2). The overall survival of the 15 patients with stage T2 disease was 68 percent -- significantly better than that of patients in stages T3 and T4 (P = 0.02) (Figure 3). Among the 40 patients who did not have hydronephrosis at presentation, survival tended to be better (P = 0.10) and survival without metastatic disease was significantly better (P = 0.05) than among the 13 patients who presented with hydronephrosis (Figure 4). Among the 31 patients with complete responses, the estimated five-year survival was significantly better than among the 22 with incomplete responses to the initial chemotherapy combined with radiotherapy (63 percent vs. 27 percent, P = 0.004).
|
|
|
Among the 28 patients who were appropriately selected for bladder-preserving treatment (i.e., patients who responded completely after initial chemotherapy combined with radiotherapy and then received consolidation treatment with full doses of chemotherapeutic agents and radiation [6480 cGy]), 25 (89 percent) remained free of an invasive recurrence, but 3 (11 percent) required a radical cystectomy. Only 3 of the 13 patients with hydronephrosis at presentation had bladder preservation without a recurrence of an invasive tumor, whereas 28 of 40 patients without hydronephrosis at diagnosis had preservation without recurrence (23 percent vs. 70 percent, P = 0.002).
Proportional-Hazards Analysis
Regression analyses were used to provide quantitative estimates of the association of the following seven clinical and pathological tumor factors with overall survival and bladder preservation without an invasive tumor in the 42 patients completing treatment: clinical stage (T2 vs. T3 and T4), the presence or absence of (tumor-associated) hydronephrosis, DNA ploidy,21 tumor size on cystoscopic evaluation, morphologic configuration (papillary vs. solid; mixed tumors were grouped with papillary tumors), the presence of vascular invasion by tumor, and the presence of associated carcinoma in situ adjacent to the tumor or in selected mucosal specimens22. The possible influence of modification of the doses of chemotherapeutic agents was assessed for any correlation with clinical outcome, and none was found. Multivariate analysis of overall survival showed that the presence or absence of hydronephrosis as well as the clinical stage both approached statistical significance (Table 1). Univariate analysis of the rate of bladder preservation without recurrence of invasive tumor in relation to all studied tumor factors (Table 2) revealed a significant difference in only one comparison, that of patients who had hydronephrosis with those who did not (P = 0.003). Similarly, multivariate analysis demonstrated that the presence or absence of hydronephrosis was the only significant prognostic factor (P = 0.01).
|
|
Transurethral resection of invasive bladder cancer as thorough as possible, followed by multidrug chemotherapy and external-beam irradiation with concurrent administration of cisplatin, was fairly well tolerated by the patients we studied (mean age, 66 years). Fatigue, nephrotoxicity that is usually reversible,16 and hematologic toxicity have been the main adverse systemic effects. The value for creatinine clearance regarded as low enough to disqualify a patient from the study has been raised from 50 to 60 ml per minute to try to reduce the degree of both nephrotoxicity and leukopenia. The patients' pelvic tissues tolerated the combination therapy well. No patient who received complete treatment with chemotherapy and radiotherapy had a major complication involving the bladder or the rectum, although two had intermittent hematuria, which was controlled by conservative measures. Our results and those of the Radiation Therapy Oncology Group23 suggest that combining cisplatin with radiation may not have any important synergistic toxic effects on the bladder or bowel.
The rate of complete responses to our initial program of combined therapy was 58 percent among the 53 patients enrolled and 68 percent among the 40 who did not have tumor-associated hydronephrosis at presentation. During a median follow-up of four years, 3 of 28 patients (11 percent) with complete responses to the full courses of chemotherapy and radiotherapy had local recurrence of an invasive tumor. Nine patients (17 percent of the total) had recurrence of a superficial bladder cancer (carcinoma in situ in eight of these nine patients), successfully managed in eight by further transurethral surgery and intravesical drug therapy. Future recurrences are of concern but are less of a worry now, since we have followed the patients for a median of four years and all patients have given up smoking. No invasive bladder tumor has recurred in 31 patients, or 58 percent of those studied -- a rate higher than that achieved with transurethral surgery alone,10,12 radiation therapy alone,1,9,14 or systemic chemotherapy alone24. Although it is not possible to attribute these results to one of these treatments rather than to another, the combination of all three is fairly well tolerated. Preliminary results of the use of a similar therapeutic strategy by a multi-institutional national group in 91 patients also indicate a high initial response rate and low morbidity23.
According to multivariate analysis of tumor factors possibly influencing survival after our combination treatment, only the presence or absence of hydronephrosis and the tumor stage independently approached statistical significance (P = 0.07 and P = 0.09, respectively). The actuarial overall survival rate of 48 percent at five years in our series is not inferior to that reported in other trials, randomized7,14,15 or not,5,11,12,13 most of which included radical cystectomy for all patients. Multivariate analysis also showed that the only independent prognostic factor indicating that the bladder would remain free from invasive tumor was the absence of hydronephrosis at the time of diagnosis (P = 0.01). Patients with hydronephrosis at the time of presentation should not be treated with the combination treatment used in our study, because their chance of remaining free of invasive tumor is only about 1 in 5. Such patients should be considered for cystectomy. The success rate of bladder preservation was 89 percent among the patients with complete responses to initial treatment, and this finding supports the strategy of combination therapy and bladder preservation in selected patients.
Our results are equivalent to those reported for radical cystectomy, but our conclusions are weakened by the absence of a simultaneous, randomized control group. Without a randomized comparison of conventional cystectomy with chemotherapy, radiotherapy, and bladder-preserving treatment combined, our findings about overall survival may have been influenced by an unknown sample bias. Prospective, randomized trials will have to be conducted to evaluate whether selective bladder preservation relying on chemotherapy and radiotherapy is an effective alternative to radical cystectomy. It may be difficult, however, to recruit patients for a trial comparing cystectomy and bladder-preserving treatment because of patients' strong preference for bladder preservation.
Supported in part by a grant (CA-56381-01) from the National Cancer Institute to the Departments of Urology and Radiation Oncology, Massachusetts General Hospital.
We are indebted to Dr. George R. Prout, Jr., for his leadership in clinical research during his tenure as chief of urology at Massachusetts General Hospital (1971-1991) and chairman of the National Bladder Cancer Group (1973-1986), and for his signal contributions to the design and implementation of this study.
Source Information
From the Departments of Medical Oncology (D.S.K.), Radiation Oncology (W.U.S., J.T.E.), and Urology (P.P.G., N.M.H., A.F.A.), Massachusetts General Hospital and Harvard Medical School, Boston.
Address reprint requests to Dr. Kaufman at MGH Cancer Center, Massachusetts General Hospital, Fruit St., Boston, MA 02114.
References
| |||||||||||||||||||||||||||||||||||||||
Related Letters:
Bladder Preservation after Treatment of Invasive Bladder Cancer
Lytton B., Berman D. J., Kaufman D. S., Shipley W. U., Heney N. M., Althausen A. F.
Extract |
Full Text
N Engl J Med 1994;
330:867-868, Mar 24, 1994.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
