Phase II trial of gemcitabine and docetaxel in patients with advanced carcinoma of the urothelium

A trial of the Eastern Cooperative Oncology Group


  • The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute.

  • This study was conducted by the Eastern Cooperative Oncology Group (Robert L. Comis, M.D.)



Gemcitabine and docetaxel are active agents in advanced urothelial carcinoma. A Phase II trial of this combination was performed to determine the activity and toxicity of these agents in a multiinstitutional setting in patients previously treated with one prior chemotherapy regimen.


Twenty-nine eligible patients with advanced urothelial carcinoma were treated with docetaxel at a dose of 40 mg/m2 over 1 hour followed by gemcitabine, 800 mg/m2, over 30 minutes, both intravenously (i.v.) on Days 1 and 8. Cycles were repeated every 21 days until disease progression or a maximum of 6 cycles.


Five patients obtained an objective response for an overall response rate of 17% (90% confidence interval, 7–33%). One patient achieved a complete clinical response. The median overall survival of the group was 7.7 months. Toxicity was moderate with granulocytopenia, anorexia, and fatigue being the most commonly noted side effects.


Gemcitabine and docetaxel is an active second-line combination in patients with advanced urothelial carcinoma. Responses in visceral, lymph node, and soft tissues sites were observed. Granulocytopenia without fever, fatigue, and anorexia was common. Thromboembolic symptoms were reported and are of concern. The combination of gemcitabine and docetaxel has the potential to palliate a subset of previously treated patients with an adequate performance status. Cancer 2003;97:2743–7. © 2003 American Cancer Society.

DOI 10.1002/cncr.11413

Advanced transitional cell carcinoma is a moderately chemosensitive neoplasm. In the 1980s, cisplatin-based chemotherapy regimens demonstrated relatively high objective response rates with some evidence of an impact on the natural history of the disease. The M-VAC regimen (comprised of methotrexate, vinblastine, doxorubicin, and cisplatin) became the de facto standard of care, although the limitations of this regimen were highlighted in the update of the Intergroup trial, which compared cisplatin therapy with M-VAC in patients with advanced urothelial carcinoma. With a minimum follow-up of 6 years, only 3.7% of patients treated on the M-VAC arm were reported to be alive and free of disease, thus emphasizing the need to seek alternative therapeutic options.1

Over the last 10 years, a large number of new agents and combination regimens have been tested in patients with advanced urothelial carcinoma. Several of these agents, including paclitaxel, gemcitabine, and docetaxel, have exhibited significant activity against advanced transitional cell carcinoma.2–4 Many doublet and triplet combinations of these agents in various Phase II trials have reported overall response rates in the range of 50–80%, with some studies yielding intriguing, prolonged median survival rates compared with those obtained with the M-VAC regimen.5 A Phase I study of gemcitabine and docetaxel evaluating a Day 1 and Day 8 schedule reported responses in two patients with bladder carcinoma, and a small Phase II trial using a different dose and schedule similarly demonstrated activity in patients with urothelial carcinoma,6, 7 prompting interest in a formal evaluation of this doublet.

Over the past 20 years, a clinically evident stage migration8 has led to earlier appreciation of metastatic disease. Concomitantly, there has been an increased utilization of neoadjuvant and adjuvant chemotherapy, resulting in an increasing proportion of patients who are fit enough to be considered for second-line chemotherapy for metastatic disease. Recent Phase II studies evaluating second-line chemotherapy have demonstrated objective response rates in the range of 10–25% for a variety of single agents and combination regimens.9–11 We performed a Phase II trial of gemcitabine plus docetaxel in patients with advanced urothelial carcinoma whose disease had progressed after one prior systemic chemotherapy regimen.


Eligible patients had histologically confirmed transitional cell carcinoma (or mixed histologies containing a component of transitional cell carcinoma) of the urothelium with evidence of progressive, bidemensionally measurable regional or metastatic disease. Patients must have been disease-free from prior malignancies for at least 5 years, with an Eastern Cooperative Oncology Group (ECOG) performance status of 0–2 at the time of study entry. Patients must have demonstrated progressive disease after 1 and only 1 prior nongemcitabine/docetaxel-containing chemotherapy regimen (in either the adjuvant, neoadjuvant, or metastatic setting) and must have been at least 4 weeks past major surgery. Adequate renal and hepatic function were required with a serum creatinine level of ≤ 1.8 mg/dL, aspartate aminotransferase (AST) ≤ 2.5, and bilirubin ≤ 1.5 times the upper limit of normal. Adequate bone marrow reserve was mandated with a requirement for an absolute neutrophil count ≥ 1500 mm3 and a platelet count ≥ 75,000/μL at the time of study entry. Patients with significant congestive heart failure, severe ventricular arrhythmias, or ≥ Grade 2 peripheral neuropathy were excluded.

Docetaxel was administered intravenously (i.v.) over 1 hour at a dose of 40 mg/m2 on Days 1 and 8, followed by gemcitabine, 800 mg/m2 i.v. over 30 minutes. Dexamethasone at a dose of 8 mg was given orally 12 hours prior and 12 hours and 24 hours after each dose of docetaxel. Courses of docetaxel and gemcitabine were scheduled to be repeated every 21 days, but were not initiated until the absolute neutrophil count was ≥ 1500/μL, the platelet count was ≥ 75,000 / μL, and the serum creatinine was ≤ 1.8 mg/dL. Dose modifications of up to 50% of both agents were mandated for febrile neutropenia, or a bleeding episode with a platelet count ≤ 40,000 /μL, any platelet nadir ≤ 20,000 /μL, or if Day 8 therapy could not be delivered secondary to neutropenia or thrombocytopenia. Growth factor support usage was according to the American Society of Clinical Oncology guidelines, and growth factor support was not used to maintain dose intensity.12

Prior to study entry, all patients underwent physical examination and had their ECOG performance status and weight documented. There also was a pretherapy complete blood count (CBC) and differential, serum creatinine, blood urea nitrogen, AST, bilirubin, and computed tomography (CT) scans of the chest and or abdomen. A CBC and the chemistries mentioned earlier were repeated on the first day of each subsequent therapy cycle. Tumor measurements were performed with each cycle of therapy if determined by physical examination or every other cycle if evaluation required chest radiographs or CT scans.

Each site was required to have approval of this clinical trial by a local Human Investigations Committee in accord with an assurance filed with and approved by the Department of Health and Human Services. All patients provided written informed consent prior to registration onto this study.

Tumor responses were analyzed using ECOG criteria.13 National Cancer Institute (NCI) common toxicity criteria (Version 2) were used to analyze toxicity. Patients with objective responses or stable disease continued to receive therapy for a maximum of six cycles. At the time the study was designed, a response rate of ≥ 25% was considered to represent activity that would be of interest. The study was designed with two stages of accrual. Twenty-two eligible patients were planned for the first stage of accrual. If ≥ 3 responses were observed among the first 22 evaluable patients, the study was planned to accrue up to an additional 18 patients. Survival was determined from the date of registration until the date of death. The duration of response was calculated from the time the response criteria were met until disease progression or the date the patient was last known to be in disease remission (if response was continuing). Simple medians, ranges, and proportions were used to describe patient characteristics. Exact binomial confidence intervals were computed for response rates.


Between December 1999 and December 2001, 31 patients from 13 ECOG institutions were entered to the study and were evaluable for toxicity. Two patients were deemed ineligible, 1 because of having received parenteral antibiotic therapy within 1 week of registration and the second because there was no prior failure of an initial chemotherapy regimen. All 29 eligible patients had tumors that were histologically pure or predominantly transitional cell carcinoma. The bladder was the primary site of disease in 25 patients. There were multiple sites involved in 13 patients, with 6 patients having lesions of the renal pelvis, 6 patients having ureteral disease, and 1 patient having urethral disease. Ten patients had received prior radiotherapy. Eighteen patients had been treated previously with carboplatin and paclitaxel; 8 patients had received previous treatment with the M-VAC regimen (comprised of methotrexate, vinblastine, doxorubicin, and cisplatin); 1 patient had received treatment with cisplatin plus 5-fluorouracil; 1 patient had received treatment with cisplatin alone; and 1 patient had been treated with carboplatin, doxorubicin, vinblastine and methotrexate. The median number of cisplatin and paclitaxel cycles administered for the entire group was 4 (range, 1–6 cycles); 38% of the patients completed 6 cycles of therapy. Additional patient characteristics are listed in Table 1.

Table 1. Patient Characteristics
No. of patients%
  • ECOG: Eastern Cooperative Oncology Group.

  • a

    Adapted from Bajorin DF, Dodd PM, Mazumdar M, et al. Long-term survival in metastatic transitional cell carcinoma and prognostic factors predicting outcome of therapy. J Clin Oncol. 1999;17:3173–3181.14

Age (yrs), median (range)64 (48–79)
 Male25 (86)
 Female 4 (14)
ECOG performance status 
 0 6 (21)
 119 (66)
 2 4 (14)
Sites of metastases 
 Liver 6 (21)
 Lung16 (55)
 Bone 3 (10)
 Soft tissue/lymph nodes16 (55)
Prognostic factors (Bajorin et al.)a 
 0 risk factors 8 (28)
 1 risk factors18 (62)
 2 risk factors 3 (10)


Clinically significant treatment-related hematologic toxicity manifested primarily as granulocytopenia. Ten patients experienced Grade 3 or 4 granulocytopenia; however, only 2 patients experienced neutropenic fever. Four patients had Grade 3 thrombocytopenia and eight patients developed Grade 3 or 4 anemia. Clinically significant (Grade 3 and 4) nonhematologic toxicities included two patients with thromboembolic events, one patient with Grade 4 renal insufficiency, one patient with Grade 4 hyperglycemia, and one patient with Grade 4 edema. Other Grade 3 events included eight patients with fatigue, one patient with dyspnea, five patients with anorexia, three patients with dehydration, and two patients with stomatitis. Fourteen patients had Grade 1 or 2 nausea.


Twenty-seven patients were assessable for response to therapy. Of the two patients who were unevaluable for response, one withdrew consent after the first cycle of therapy and declined additional imaging and one patient developed excessive toxicity prior to completing the first cycle of therapy.

Five patients obtained an objective response for an overall response rate of 17.2% (90% confidence interval, [90% CI], 7–33%). One patient obtained a clinical complete response and four patients achieved a clinical partial response. Responses were observed in patients with lymph node and lung metastases and unresectable bladder masses. Using the Mehta exact test for ordered categoric data, there was no apparent association between the Bajorin risk category and response (P = 0.67).14 The median survival of the study cohort of 29 patients was 7.7 months.


Although there has been only modest progress made in the management of advanced urothelial carcinoma since the introduction of cisplatin-based multiagent therapy in the 1980s, several developments have had a significant impact on the utilization of chemotherapy in this patient population. A clinically relevant stage migration has occurred over the last 20 years, with patients for the most part having improved performance status at the time of diagnosis. In addition, increased utilization of neoadjuvant and adjuvant chemotherapy has led to an increasing number of patients with metastatic disease having already received first-line chemotherapy. As a result, there is an expanding, albeit still modest, number of patients with advanced urothelial carcinoma who are fit enough to receive second-line chemotherapy. The veritable explosion of new chemotherapy agents with documented activity in advanced urothelial carcinoma and the recognition that these patients have a very poor prognosis with survival rates of < 1 year have fueled interest in exploring the potential effect of second-line chemotherapy on quality of life and, ultimately, survival.

A number of Phase II trials in the salvage setting have been reported.4, 9, 11, 15, 16 Witte et al. performed a large Phase II trial of ifosfamide that demonstrated a response rate of 20%, with 5 patients obtaining a complete response. However, the toxicity associated with this therapy was not insignificant, with myelosuppression being problematic.9 More recently, Lorusso et al. administered gemcitabine at a dose of 1200 mg/m2 weekly for 3 of 4 weeks in patients who previously were treated with cisplatin-based therapy and reported an overall response rate of 22.5%, with 4 patients (13%) achieving a complete response.17 Albers et al. treated cisplatin-refractory patients with gemcitabine, 1250 mg/m2, on Days 1 and 8 every 21 days and observed an 11% overall response rate with 2 of the responding patients (7%) obtaining a complete response.18 McCaffrey et al. treated 30 patients with residual or progressive disease after cisplatin-based combination therapy with docetaxel at a dose of 100 mg/m2 administered on a 3-week schedule. Four patients (13%) achieved a partial response.4 Sternberg et al.16 treated 41 patients who had previously been treated with cisplatin-based regimens (primarily M-VAC) with gemcitabine and paclitaxel and observed an impressive 60% response rate with 11 patients achieving a complete response. The median survival in their experience was 14. 4 months. It is interesting to note that 63% of the patients enrolled had received their initial chemotherapy in either the neoadjuvant or adjuvant settings.16

Response rates in Phase II trials of patients with advanced bladder carcinoma can be influenced significantly by patient selection, as described by Bajorin et al.14 Their study concentrated on prognostic factors associated with response and identified performance status and the sites of metastatic disease as major response parameters. In the salvage setting, other factors in addition to those described the Bajorin et al.14 will likely influence response rates, including the time to disease progression, the drugs initially received by the patients, and the setting of initial chemotherapy (i.e., adjuvant/neoadjuvant versus therapy for metastatic disease). As is the case for many solid tumors, the therapeutic goals of salvage therapy need further characterization (i.e., palliation versus survival prolongation), given the limited impact of chemotherapy on survival in the initial treatment of patients with advanced urothelial carcinoma reported to date.

The combination of gemcitabine and docetaxel in the salvage setting of advanced urothelial carcinoma appears to have modest activity at the expense of moderate toxicity, primarily myelosuppression. Given the activity of various single agents in the salvage setting, the roles of doublet or triplet combinations remain unproven and are reported to be associated with more toxicity. The ultimate utility of this and all salvage chemotherapy regimens in this clinical setting remains undefined.