Original article. H. von der Maase, 1 L. Andersen, 1 L. Crino, 2 S. Weinknecht 3 & L. Dogliotti 4

Transcription

1 Annals of Oncology 10: Kluwer Academic Publishers. Printed in the Netherlands. Original article Weekly gemcitabine and cisplatin combination therapy in patients with transitional cell carcinoma of the urothelium: A phase II clinical trial H. von der Maase, 1 L. Andersen, 1 L. Crino, 2 S. Weinknecht 3 & L. Dogliotti 4 Department of Oncology, Aarhus University Hospital, Norrebrogade, Aarlnis C, Denmark; 2 Ospedale Pohclimco, Perugia, Italy; 3 Abteihmg filr Urologie, Berlin, Germany: * Medical Oncology. San Luigi University Hospital, Orbassano, Torino, Italy Summary Purpose: To determine the efficacy of gemcitabine and cisplatin combination therapy in patients with advanced and/or metastatic transitional cell urothelial carcinoma. Patients and methods: Forty-two chemonai've patients with Karnofsky performance status (KPS) ^ 70 were treated with cisplatin 35 mg/m 2 followed by gemcitabine 1000 mg/m 2 (30 min l.v. infusion) on days 1, 8, and 15 every twenty-eight days. Results: Thirty-eight patients were evaluable for efficacy. Half had visceral disease. There were seven complete (18%) and nine partial responses (24%), for a response rate of 42% (95% confidence interval (95% CI): 26%-59%). Responses were independently reviewed. Median response duration was 13.5 months (95% CI: months), median time to progressive disease 7.2 months (95% CI: months) and median survival 12.5 months (95% CI: months); oneyear survival was 52%. Laboratory toxicities included leucopenia (44% grade 3; 17% grade 4), neutropenia (25% grade 3; 33% grade 4) and thrombocytopenia (29% grade 3; 49% grade 4). Four patients had grade 4 symptomatic toxicity (three nausea and vomiting, one diarrhoea). There were no grade 4 infections and no toxic deaths. Conclusions: The combination of gemcitabine and cisplatin is active in patients with locally advanced and/or metastatic urothelial carcinoma. The weekly schedule of cisplatin is considered inappropriate. Key words: bladder cancer, gemcitabine-cisplatin combination, phase II study, transitional cell carcinoma Introduction There are 50,000 new cases of urothelial carcinoma in the US every year and over 66,000 in the European Union. Currently over 10,000 patients in the US and 30,000 in the European Union die annually from this disease [1, 2]. Prior to the development of effective chemotherapy, the median survival for patients with advanced or metastatic disease rarely exceeded three to six months. The combination of methotrexate, vinblastine, doxorubicin and cisplatin (MVAC) is considered the standard in the treatment of advanced bladder cancer, with response rates of 39%-72% and a median survival of months [3-6]. Most importantly, MVAC is the only regimen where the potential survival benefit has been confirmed in phase III studies, where MVAC has been compared with cisplatin alone [5] or with cisplatin in combination with cyclophosphamide and doxorubicin [4]. Long-term survival was rare, however, with only 3.7% of patients surviving for more than six years [6]. Furthermore, toxicity is considerable, with significant myelosuppression and febrile neutropenia in 25% of patients treated with MVAC. Other significant toxicities include mucositis, sepsis, renal toxicity and neurotoxicity, and approximately 3%-4% of patients treated with MVAC die as a result of treatment. The need for alternative strategies for the treatment of this disease, giving rise to improved long-term survival, has fuelled the search for new treatment regimens for patients with urothelial carcinoma. The nucleoside analogue gemcitabine (GEMZAR, Eli Lilly and Company, Indianapolis, IN) has been shown to have singleagent activity against urothelial carcinoma. In a phase I trial of gemcitabine in patients with urothelial carcinoma, responses were seen in patients who had previously received MVAC, which had been used to treat all but one patient [7]. Three subsequent phase II studies have investigated the single-agent activity of gemcitabine in patients with advanced or metastatic transitional cell carcinoma (TCC) [8-10]. In the first of these, an Italian study, four of 31 evaluable patients had a complete response and three a partial response, for a response rate of 23% [8]. All patients had received cisplatin-containing chemotherapy, in most cases MVAC. The second, a US study, reported four complete and seven partial responses in 39 chemonaive evaluable patients for an overall response rate of 28% [9]. In the third, a Canadian study, there were three complete and six partial responses in 37 evaluable chemonaive patients, for a response rate of 24% [10]. In all studies, treatment with gemcitabine was very well tolerated, with a low incidence of haematological and non-haematological toxicities. Given that cisplatin is considered to be the most

2 1462 active agent in urothelial carcinoma, and that gemcitabine has clearly demonstrable single-agent activity in patients with advanced urothelial carcinoma, it seemed reasonable to investigate the use of these two agents in combination. The combination of gemcitabine and cisplatin has already been shown to be more active than single-agent gemcitabine in patients with non-small-cell lung cancer (NSCLC) [11-14]. The present study was initiated to investigate the activity and toxicity profile of a combination of gemcitabine and cisplatin administered weekly for three weeks, followed by one week of rest, to chemonaive patients with advanced or metastatic urothelial carcinoma. Patients and methods Patients Patients were eligible for entry into the study if they met the following inclusion criteria: histological diagnosis of advanced and/or metastatic transitional cell carcinoma of the urothehum (stage III and IV not amenable to curative surgery or radiotherapy); no prior systemic immunotherapy or chemotherapy; local intravesicular immunotherapy or chemotherapy was allowed if received > 6 months prior to study, radiation therapy was allowed if received > 3 weeks prior to study and if the irradiated area was not the only source of measurable disease. Other eligibility requirements included: clinically measurable disease; a Karnofsky performance status of 70%-100%; adequate bone marrow reserves defined by a white blood cell (WBC) count 3=3 x 10 9 /l, platelets > 100 x 10 g/dl or >7.0 mmol/1; adequate renal function, defined as normal to moderately decreased chromium-ethylenediaminetelra-acetic acid (Cr-EDTA) clearance or calculated creatinine clearance >60 ml/min; a life expectancy of at least 12 weeks and signed informed consent. Exclusion criteria included' inadequate liver function, defined as bilirubin levels greater than 1.5 times normal, prothrombin and partial thromboplastin time greater than 1.5 times normal and transaminases greater than 3 times normal; abnormal calcium levels; active infection; central nervous system (CNS) metastases or any second primary carcinoma. Measurable disease was defined as lesions of at least 1 cm x I cm as determined by computerised tomography (CT) scan, magnetic resonance imaging (MRI). X-ray or physical examination. Table I. Baseline patient characteristics Total number of patients Qualified for efficacy Sex Male Female Age (in years) Median Range Disease stage III IV Karnofsky performance status $80 Number of sites of metastatic disease 1 2 S=3 Sites of metastatic disease (patients) Lymph nodes Liver Bone Lung Other II received four additional cycles. Patients with SD received a maximum of six cycles. A physical examination was conducted before each treatment and the number of units required for transfusion was recorded at every cycle. Further safety and toxicity evaluations were as follows, haematology was performed weekly and blood chemistry and urinalysis performed every two weeks, renal function was monitored every eighth week. Toxicity was assessed using WHO scales. Endpoints and statistics The primary study endpoint was response rate. The study was conducted in a two-stage design. An initial 20 qualified patients were entered into the first stage of the study with at least two responses required to proceed to the next accrual target. The study was designed to accrue up to 42 patients if the initial two responses were observed. Time to progressive disease and duration of response were also studied using Kaplan-Meier estimations. Survival curves were generated by the Kaplan-Meier method Therapy and patient monitoring Patients were treated with cisplatin 35 mg/m 2 followed by gemcitabine 1000 mg/m 2. on days 1. 8 and 15 of each 28-day cycle. Cisplatin was administered after prehydration with 1000 ml saline and 20 meq potassium chloride over 60 min. A further 1000 ml of normal saline or glucose was given over the two-hour following cisplatin administration and before gemcitabine. Cycles were repeated every 28 days, as long as platelet counts were > 50 x 10 9 /l, white blood cell counts were > 1 x 10 9 /l and creatinine clearance was >40 ml/min. Prophylactic therapy for nausea and vomiting was recommended. Colony stimulating factors were not used. Disease evaluation was performed every two cycles using chest X-ray and the pretreatment method in evaluable tumour sites, while radiological imaging studies demonstrated sites of disease other than those chosen for evaluability. Standard WHO criteria for response were used [15]. All responses were reconfirmed after a minimal interval of four weeks. All responses were reviewed by an independent radiologist. Treatment was discontinued in patients with PD or in cases of unacceptable toxicity. at the imestigators' discretion. Patients with a CR received up to four additional cvcles. while patients with a PR Results Patients Between March 1995 and July 1996, 42 patients were entered into this multicentre study: 2 from 1 center in Belgium; 6 in 2 centres in Germany; 16 from 5 centers in Italy and 18 from 1 center in Denmark. All patients were included in safety analyses. Their demographics are shown in Table 1. The median age of patients at entry was 64 years (range 48-74). Twenty patients had lymph node involvement, 12 had liver metastases, 9 had bone and 4 had lung metastases. Nine patients had lymph node metastases only and two had advanced bladder tumours (T4b) only. All 42 patients had undergone prior surgery (mainly biopsies), three had received radiotherapy and one had prior intravesical chemotherapy

3 1463 Table 2. Response to treatment with gemcitabine and cisplatin. Evaluable patients 38 (100) Number of patients (%) Complete response 7(18) Partial response 9 (24) Stable disease 10(26) Progressive disease 8 (21) Not evaluable at follow-up 2 (5) Follow-up measurements not performed 2 (5) Overall response rate (95% Cl) 16 (42) (26-59) Table 3. Maximum laboratory toxicities (WHO grade 3 and 4). Toxicity Total number WHO grade 3. WHO grade 4. of patients n (%) n (%) Leucopenia Neutropenia Thrombocytopenia Anaemia (44) 9(25) 12(29) 12(29) 7(17) 12(33) 20 (49) 0(0) (mitomycin) before entry into the trial. A further four patients had intravesical bacillus Calmette-Guerin (BCG) immunotherapy prior to study entry. In total, 38 patients qualified for efficacy analysis. Two patients did not receive at least one cycle of treatment: one had a performance status of 40 due to severe bone pain resulting from osseous metastases, the other had a creatinine clearance of 57 ml/min which was lower than the protocol-defined minimum of 60 ml/min. A further two patients did not have bidimensionally measurable lesions and could not be included in efficacy analysis. Response All responses were confirmed at four weeks and independently reviewed. There were seven CRs (18%) and nine PRs (24%), for an overall response rate of 42% (Table 2). A further 10 patients had SD and 8 patients had PD. One patient with a CR had one residual lymph node at the end of treatment. This lesion was surgically removed and shown to be benign. One PR patient was rendered disease free by surgical resection of a single residual lymph node metastasis. Two patients were not evaluable at follow-up and follow-up measurements were not done for two others. Responses were seen in lesions in the liver, lung, bone and penis, as well as lymph nodes. Seven out of nine patients with disease confined to the lymph nodes responded to treatment. All seven complete responses were seen in patients with stage IV disease. The median duration of response for all responding patients was 13.5 months (95% CI: months) and the median time to progressive disease was 7.2 months (95% CI: months). The median survival for all patients was 12.5 months (95% CI: months) and the one-year survival rate was 52% (Figure 1). At the time of analysis, four responses were ongoing (32+ Survival Time Figure 1. Overall survival curve for all patients («= 42) with advanced urothelial carcinoma who received gemcitabine and cisplatin months to 48+ months). One of these patients had bone metastases as well as lymph node and bladder involvement. One patient had bladder and lymph node involvement and the two remaining patients had lymph node disease only. The two patients with bladder involvement had no further treatment following chemotherapy. Toxicity WHO laboratory toxicities are summarised in Table 3. Grade 3 and 4 leucopenia was reported in 44% and 17% of patients, respectively. Grade 3 and 4 neutropenia was seen in 25% and 33% of patients, respectively. One patient with grade 4 neutropenia also had grade 3 infection, while three patients with grade 3 or 4 neutropenia had grade 2 infection. No grade 4 infections were seen. Eight patients were hospitalised for fever; two exhibited grade 3 or 4 neutropenia. Grade 3 and 4 thrombocytopenia was reported in 29% and 49% of patients, respectively, resulting in grade 3 haemorrhage in one patient. This patient had vaginal bleeding, and on physical examination a large, bleeding, necrotic mass was observed. Of the 20 patients with grade 4 thrombocytopenia, 9 received platelet transfusions. Six patients had grade 3 haematuria, all of whom had grade 2-4 thrombocytopenia. Grade 3 anaemia was seen in 29% of patients. In total, 30 patients had transfusions during the study, 14 had platelet transfusions and 28 received red blood cells. Other laboratory toxicities included one patient with grade 4 alkaline phosphatase, who also had liver metastases. No grade 3 or 4 liver toxicities were reported. It is notable that no grade 3 or 4 renal toxicity was reported. Only five patients had grade 1 blood urea nitrogen, with no creatinine toxicity. Three patients were discontinued as a result of decreasing renal function, one had decreasing creatinine clearance after five cycles, one had baseline creatinine clearance below the required level and was discontinued after the first injection, and one had abnormal kidney function and reduced Cr-EDTA

4 1464 clearance (from 65.1 ml/min at baseline to 26.5 ml/min after four cycles). This patient also had grade 1 creatinine toxicity. Non-laboratory toxicities included grade 4 diarrhoea in one patient who was removed from the study after one cycle for progressive disease and grade 3 diarrhoea in one other patient. Grade 3 and 4 nausea and vomiting occurred in 29% of patients. Grade 3 pulmonary toxicity was observed in three patients (7%), one of whom had a previous history of severe obstructive pulmonary disease. This patient was withdrawn from the study as the symptoms worsened during treatment and the patient had not responded to therapy. Ten patients (24%) had grade 2 and one patient had grade 3 alopecia. Nine patients discontinued treatment during the study. Reasons for discontinuation were (one patient each) anaphylactic reaction to cisplatin, asthenia, increased creatinine, pre-existing hydronephrosis, abnormal kidney function, leucopenia, thrombocytopenia, pain, and pneumonia. There were no toxic deaths. The median number of cycles received was four (range 0-8). The median gemcitabine dose administered was 768 mg/m 2 per week (range mg/m 2 ) and the median cisplatin dose was 28 mg/m 2 per week (range 9-37). Patients received, on average, 80% of the planned cisplatin dose and 74% of the planned gemcitabine dose. Of the total number of gemcitabine doses planned, 34% were reduced and 20% omitted, while 21% of cisplatin doses were reduced and 16% omitted. Haematological toxicity accounted for the majority of these alterations. Leucopenia was responsible for 14% and 27% of cisplatin and gemcitabine reductions, respectively, while thrombocytopenia was responsible for 36% and 41% of cisplatin and gemcitabine reductions, respectively. Abnormal renal function, as assessed by measured creatinine clearance and Cr-EDTA, led to reductions in 30% and 17% of cisplatin and gemcitabine doses, respectively. These were not generally associated with increased creatinine levels. Discussion The combination of gemcitabine and cisplatin, administered on days 1, 8 and 15 of a 28-day cycle, has been shown to be active in patients with advanced urothelial carcinoma. In this study, 42% of qualified patients had an objective response to treatment and 18% had a complete response. Responses, which were independently reviewed, were seen in a variety of metastatic sites, including liver, lung, and bones. The median response duration of 13.5 months and the median survival of 12.5 months for all patients, including five long-lasting responses, was encouraging, given the fact that the majority of patients had stage IV disease (88%), 50% had visceral disease and almost half (47%) had a KPS of 80 or less. Many of the conventional treatments for advanced urothelial carcinoma, including the MVAC regimen, have been associated with considerable toxicities. Gemcitabine, on the other hand, is generally well tolerated as a single agent and the combination of gemcitabine and cisplatin is already known to be well tolerated in patients with NSCLC [11-14]. In the present study, myelosuppression was the most frequently reported side effect. Grade 4 neutropenia and thrombocytopenia were seen in 33% and 49% of patients, respectively, but there was only one case each of grade 3 infection and haemorrhage and there were no toxic deaths. Thus haematological toxicity was mainly a laboratory concern, which did not result in major clinical problems. Renal toxicity, a potential complication arising from cisplatin therapy, was not common with this combination, although three patients were withdrawn from the study as a result of decreased renal function. There were no reports of grade 3 or 4 renal toxicity (creatinine clearance or blood urea nitrogen). Indeed, one patient had creatinine clearance levels that did not meet the entry criteria. The main nonlaboratory toxicity was nausea and vomiting: 12 patients (29%) had grade 3 and 4 nausea and vomiting. The high incidence of this toxicity and the frequency of dose reductions and omissions was considered to be at least partly due to the weekly administration of cisplatin, which is felt to be inappropriate in this patient group. There was only one other incidence of grade 4 nonlaboratory toxicity, and only occasional reports of grade 3 non-laboratory toxicities. As an alternative to the dosing schedule used in this study, Stadler et al. [16] have used gemcitabine (1000 mg/m 2 ) on days 1, 8 and 15, and cisplatin (75 mg/m 2 ) on day 1 of a 28-day cycle in patients with metastatic urothelial cancer. This combination appears to be well tolerated and highly active, with a response rate of 66% (95% CI: 51%-79%). Day 2 administration of cisplatin is also under investigation and preliminary results suggest good activity, with four complete responses and eight partial responses in 17 evaluable patients, for a response rate of 71%, and reasonable toxicity [17]. Grade 3 and 4 neutropenia and thrombocytopenia were less frequent in the two studies using a single dose of cisplatin once every 28 days than in the present study. Consequently, a phase III trial, comparing the combination of day 2 cisplatin (70 mg/m 2 ) and days 1, 8 and 15 gemcitabine (1000 mg/m 2 ) with MVAC, has been initiated. We conclude that the combination of gemcitabine and cisplatin is active in patients with transitional cell carcinoma of the urothelium. We consider the weekly schedule of cisplatin administration to be inappropriate and the schedule has been optimised by use of monthly cisplatin. A direct comparison with MVAC is underway. Acknowledgements The authors would like to thank Eli Lilly and Company for supporting this study and to acknowledge the contribution of the following clinicians who entered patients

5 1465 into this study: Dr P. F. Conte, Ospedale Santa Chiara, Pisa, Italy; Prof. Y. Humblet, UCL St. Luc, Brussels, Belgium; Prof. G. Jakse, Urologische Klinik der RWTH, Aachen, Germany; Prof. L. Weissbach, Urban Hospital, Berlin, Germany; Dr D. Perroni, Ospidale Santa Croce, Cuneo, Italy; Dr G. F. Porcile, Ospidale San Lazzaro, Alba, Italy; Prof. M. Tonato, Policlinico Monteluce, Perugia, Italy. References 1. Parker SL. Tong T. Bolden S, Wingo PA Cancer statistics, Ca Cancer J Clin 1996; 46: Black RJ. Bray F. Ferlay J. Parkin DM. Cancer incidence and mortality in the European Union. Cancer registry data and estimates of national incidence for Eur J Cancer 1997, 33: 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 1989: 64: Logothetis CJ, Dexus FH, Finn L et al. A prospective, randomized trial comparing MVAC and CISCA chemotherapy for patients with metastatic urothelial tumors. J Clin Oncol 1990; 8: 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 1992: 10: Saxman SB, Propert KJ, Einhorn LH et al. Long-term follow-up of a phase III intergroup study of methotrexate, vinblastine and doxorubicin in patients with metastatic urothelial carcinoma: A Cooperative Group study. J Clin Oncol 1997; Pollera CF. Ceribelli A, Crecco M. Calabresi F. Weekly gemcitabine in advanced bladder cancer: A preliminary report from a phase I study. Ann Oncol 1994, Lorusso V, Pollera CF, Antimi M et al. A phase II study of gemcitabine in patients with transitional cell carcinoma of the urinary tract previously treated with platinum. Eur J Cancer 1998; 34: Stadler WM, Kuzel T, Roth B et al Phase II study of single-agent gemcitabine in previously untreated patients with metastatic urothelial cancer. J Clin Oncol 1997: 15: Moore MJ. Tannock IF. Scott Ernst D et al. Gemcitabine: A promising new agent in the treatment of advanced urothelial cancer. J Clin Oncol 1997; 15: Crino L, Scagliotli G. Marangolo M et al. Cisplatin-gemcitabine combination in advanced non-small-cell lung cancer: A phase II study J Clin Oncol 1997; 15: Abratt RP, Bezwoda WR, Goedhals L. Hacking DJ. Weekly gemcitabine with monthly cisplatin: Effective chemotherapy for advanced non-small-cell lung cancer. J Clin Oncol 1997; Anton A, Carrato A, Gonzalez-Larriba JL et al. Phase II activity of gemcitabine in combination with cisplatin in advanced nonsmall-cell lung cancer. Proc Am Soc Clin Oncol 1996: 15: 380 (Abstrll34). 14. Sandier A, Nemunaitis J, Dehnam C et al. Phase III study of cisplatin (C) with or without gemcitabine (G) in patients with advanced non-small-cell lung cancer (NSCLC). Proc Am Soc Clin Oncol 1998; 17: 454a (Abstr 1747). 15. World Health Organization. Handbook for Reporting Results of Cancer Treatment WHO Offset Publication No. 48. Geneva. WHO Stadler W, Carducci M, Raghavan D et al. Gemcitabine (GEM) plus cisplatin (CDDP) in metastatic transitional cell carcinoma (TCC): Final results of a phase II study. Proc Am Soc Clin Oncol 1998; 17: 320a (Abstr 1235). 17. Moore MJ, Tannock I, Winquist E et al. Gemcitabine (G) + cisplatin (C): an active regimen in advanced transitional cell carcinoma (TCC). Proc Am Soc Clin Oncol 1998; 17' 320a (Abstr 1234). Received 28 July 1999; accepted 25 October Correspondence to. Prof. H. von der Maase, MD Department of Oncology Aarhus University Hospital Norrebrogade DK-8000 Aarhus C Denmark

6