Original article. Introduction

Transcription

1 Annals of Oncology 7: , O 1996 Kluwer Academic Publishers. Printed in the Netherlands. Original article A randomized trial of cisplatin, etoposide and bleomycin (PEB) versus carboplatin, etoposide and bleomycin (CEB) for patients with 'good-risk' metastatic non-seminomatous germ cell tumors C. Bokemeyer, 1 O. Kohrmann, 2 J. Tischler, 3 L. WeiBbach, 4 U. Rath, 5 A. Haupt, 1 P. Schoffski, 6 A. Harstrick 7 & H.-J. Schmoll 8 1 Department of Internal Medicine, University of Tubingen, Tubingen; 2 Department of Urology, Stadtisches Klinikum Mannheim, Mannheim; 3 Department of Hematology/Oncology, Klinikum Minden, Minden; 4 Department of Urology, Krankenhaus Am Urban, Berlin; ^Department of Hematology/Oncology, University of Heidelberg, Heidelberg; ''Department of Hematology/Oncology, Hannover University Medical School, Hannover; 1 Department of Oncology, West-German Cancer Center, Essen; B Department of Hematology/Oncology, University of Halle, Halle, Germany Summary Background: -based combination chemotherapy will cure 70% to 80% of patients with metastatic non-seminomatous germ cell tumors but is associated with the possibility of severe neuro-, oto- and nephro-toxicities., a cisplatin analogue, is an active drug in testicular cancer with a more favourable spectrum of side effects. In a randomized trial, the German Testicular Cancer Study Group compared a combination regimen of carboplatin, etoposide and bleomycin (CEB) to standard cisplatin, etoposide and bleomycin (PEB) chemotherapy for patients with 'minimal-' and 'moderate-disease' non-seminomatous germ cell tumors, according to the Indiana University classification. Patients and methods: PEB was given for three cycles at standard doses (given days 1-5), and the CEB regimen consisted of carboplatin (target AUC of 5 mg/ml x min) on day 1, etoposide 120 mg/m 2 on days 1 to 3 and bleomycin 30 mg on days 1, 8 and 15. Four cycles of CEB were given, with the omission of bleomycin in the fourth cycle. Thus, the cumulative doses of etoposide and bleomycin applied in the two treatment arms were comparable. Fifty-four patients were entered on the trial, 29 were treated with PEB and 25 with CEB chemotherapy. Patients were stratified according to disease extent (minimal versus moderate) and the degree of tumor marker elevation. Thirty-two patients (59%) belonged to the group with minimal disease and low markers. Results: No significant difference in response to chemotherapy was seen between the two arms, with CR rates of 81% for the PEB arm and 76% for CEB treatment. However, more patients treated with CEB (32% versus 13%) have relapsed after therapy, and 4 patients (16%) have died of disease progression after CEB in contrast to 1 (3%) after PEB therapy. The first interim analysis of negative events (relapse, vital tumor at secondary resection, death from the disease and therapy-associated death) showed a significantly higher rate after CEB than after PEB therapy, and the trial was terminated early. After a median follow-up of 33 months for all patients, the calculation of negative events is still significantly in favour of PEB-treated patients, particularly since three late relapses > 2 years have been observed in the CEB arm (P-0.03). Conclusion: This randomized trial demonstrates that even with the use of adequate doses of etoposide and full-dose bleomycin, carboplatin cannot altogether replace cisplatin in patients with testicular cancer. Treatment with the PEB regimen remains the standard approach in patients with 'goodrisk' non-seminomatous germ cell tumors. Key words: carboplatin, chemotherapy, cisplatin, Indiana University classification, prognostic factors, randomized study, testicular cancer, toxicity treatment strategies have focused on the reduction of treatment-associated toxicity [4]. A randomized trial in 1987 led to replacement of the more toxic cisplatin, vinblastine, bleomycin (PVB) regimen by cisplatin, etoposide and bleomycin (PEB) [5]. Subsequently, a randomized trial by the SECSG demonstrated that 3 cycles of PEB therapy are as effective as 4 cycles of PEB for patients with 'good-prognosis' non-seminoma [6]. However, the deletion of bleomycin from 3 cycles of PEB therapy during a follow-up trial resulted in inferior results and bleomycin has therefore remained part of the standard PEB treatment regimen for testicular cancer [7]. was developed as a less toxic com- Introduction The use of cisplatin-based combination chemotherapy for the treatment of metastatic non-seminomatous germ cell tumors has led to a dramatic improvement in the prognosis of these patients [1]. Approximately 70% to 80% of patients with metastatic disease are now likely to be cured, and several prognostic factors, including tumor stage, tumor marker elevation and site of metastatic disease, have been identified as distinguishing 'good-prognosis' from 'poor-prognosis' patients [2, 3]. For those in the 'good-prognosis' group such as patients designated as having 'minimal' or 'moderate' disease according to Indiana University criteria, current

2 1016 pound of cisplatin; it has demonstrated significant activity in cisplatin-sensitive rumors, and is associated with less neuro-, oto- and nephro-toxicity than cisplatin [8, 9]. The combination of carboplatin, etoposide and bleomycin (CEB) for a total of 4 cycles was evaluated by the Royal Marsden Hospital as primary chemotherapy for patients with 'good-prognosis' metastatic nonseminomatous germ cell tumors. Only 5 of 76 treated patients have failed CEB therapy and the 2-year causespecific survival was 98.5%. The toxic effects of CEB chemotherapy were mild [10]. These favourable results comprised the rationale for a randomized trial conducted by the German Testicular Cancer Study Group comparing 3 cycles of the standard PEB regimen to 4 cycles of CEB treatment, using for CEB the schedule proposed by the Royal Marsden Hospital. Patient enrollment began in 1992 and an interim analysis after entry of the first 54 patients revealed an increased rate of negative events in patients treated with CEB chemotherapy. In accordance with an early cessation rule, the trial was closed in December All patients were followed until June 1996 and this report presents the initial analysis and the follow-up results after a median observation time of 33 months for all patients. Patients and methods From September 1992 to December 1993, patients with metastatic non-seminomatous testicular cancer and 'minimal' or 'moderate' disease according to Indiana University criteria [4 were randomized to receive either 3 cycles of cisplatin- or 4 cycles of carboplatinbased combination chemotherapy. Minimal disease patients had tumor marker elevation (AFP, p-hcg) only, cervical or supraclavicular lymph node metastases, abdominal lymph node metastases <10 cm in diameter or less than five pulmonary metastases, all smaller than 2 cm in diameter. Moderate disease was defined as abdominal lymph node metastases >10 cm without further metastatic disease, 5-10 pulmonary metastases all smaller than 3 cm in diameter, a singular pulmonary metastasis >3 cm or mediastinal lymphatic nodes <50% of the intrathoracic diameter. Patients with primary mediastinal germ-cell tumors and patients with advanced disease according to the Indiana University classification were excluded from the trial. Patients with histologically pure seminoma were also ineligible. Further inclusion criteria were normal bone marrow, renal and hepatic function, no previous chemotherapy, age between 18 and 65 years and informed consent. Ethical approval was given by the Hannover University Medical School ethics committee. Treatment Patients were randomized to receive either 3 cycles of PEB with cisplatin 20 mg/m 2 on days 1-5 and etoposide 100 mg/m 2 on days 1-5 as short infusions and bleomycin 30 mg/day bolus on days 1, 8 and 15. Treatment was repeated at 3-weekly intervals for a total of 3 cycles. combination therapy consisted of carboplatin dosed according to GFR based on the Calvert calculation to achieve an AUC of 5 mg/ml x min as a short infusion [11], etoposide 120 mg/m 2 on days 1-3 short infusion and bleomycin 30 mg bolus on days 1, 8 and 15. was given only in cycles 1-3 of the CEB regimen, while a total of 4 cycles of CEB were applied at intervals of 22 days. This CEB regimen had been proposed by the Royal Marsden Hospital in 1991 [10]. During subsequent courses the dose of carboplatin was adjusted according to the leukocyte and thrombocyte nadirs on day 16 of each previous treatment cycle. Dose reductions to 75% of the previous dose were performed in instances of leukocytes < 1000/ul and fever during the previous cycles or in patients with thrombocytes < 50000/ul and clinical symptoms of thrombocytopenia. If no leukocytopenia (>2000/\L\) or thrombocytopenia (> /u.l) were observed, the dose of carboplatin was increased by 10%. The use of 3 cycles of the established PEB regimen and 4 cycles of the proposed CEB regimen resulted in comparative cumulative doses of etoposide (1500 mg/m 2 and 1440 mg/ m 2, respectively) and bleomycin (270 mg), since no bleomycin was given in the fourth cycle of CEB. Thus, the major difference between the two regimens was the substitution of carboplatin for cisplatin. Prior to randomisation all patients were stratified according to 4 criteria; minimal versus moderate disease and low versus high tumor markers (high markers: AFP > 1000 ng/ml and/or fj-hcg > IU/1). During treatment, patients were restaged after two cycles and at the end of therapy using tumor markers and a radiological assessment for response. In patients with marker normalisation and persisting residua] disease (PR marker negative: PRm-) surgical resection was attempted, while those with marker normalisation and complete remission were only followed. Screening for ototoxicity, renal toxicity, peripheral neuropathy and pulmonary function was performed to assess the side effects of treatment. Apart from the tumor markers, standard WHO criteria were used for the classification of tumor response and for treatment toxicity. The statistical analysis was based on the assumption of a 1-year relapse-free survival of 90% in both treatment arms. The anticipated number of evaluable patients was 154 for each arm. Recruitment was planned for a total of 4 years. Response rate, rate of primary progression, number of relapses and relapse-free survival were used as end points. In addition, toxicity parameters were considered. An interim analysis was prospectively planned after entry into the trial of every 50 patients. As an early assessment of efficacy, the rate of 'negative events', defined as primary disease progression, therapyinduced death, persistence of vital carcinoma in post-chemotherapy resected residual disease, occurrence of relapse and death from tumor, was used as the primary end point for this analysis. Early termination was considered at a significant difference of 10% between both arms. In order to detect the inferiority of one treatment arm early and to avoid endangering patients, both the total number of negative events and the number of patients experiencing a negative event were calculated. The maximum number of negative events per patient was three: (1) progression or residual tumor at surgery, (2) relapse, (3) death from therapy or death from progressive tumor. Adequate corrections for multiple testing during a consecutive trial were planned. Results By the time of the first interim analysis 54 evaluable patients, with a median age of 28 years (18-56), had been accrued. Most of them (59%) had only minimal disease and low tumor markers. Seven patients (13%) had elevated tumor markers only, 34 (63%) had one metastatic localisation and 13 had (24%) two or three localisations of metastases. The median follow-up for all patients was 33 months (2-44+). Two patients in CR (one per arm) were lost to follow-up during the first year after therapy. Of the patients still alive 96% were followed for longer than 1 year after therapy and 69% for longer than 2 years. Twenty-five patients were randomized to receive CEB therapy and 29 to PEB therapy. There were no imbalances between the two groups (Table 1). There

3 1017 were no significant differences in response rates between the arms, with 81% CR after PEB and 76% after CEB therapy. One patient in each arm (4%) showed primary tumor progression. 10 patients after PEB (32%) and 12 patients after CEB (48%) underwent a secondary residual tumor resection after chemotherapy. Vital undifferentiated tumor at secondary resection was persistent in 16% after PEB and 20% after CEB therapy. One-year progression-free survival rates were 86% after PEB and 84% after CEB therapy. During the follow-up period 4 patients (13%) relapsed after PEB compared to 8 patients after CEB (32%). One patient died of progressive tumor after PEB therapy compared to 4 patients after CEB treatment. Calculating the total number of negative events in the two arms CEB therapy was associated with a significantly higher rate of 17 negative events compared to 9 negative events after PEB therapy. This result was statistically significant regardless of whether the possible number of adverse events (P 0.03; Pearson) or the number of patients per treatment arm (P 0.02; Pearson) were used as the basis for comparison. Ten of 41 (25%) patients with 'minimal disease' experienced negative events in comparison to 9 of 13 patients (69%) with 'moderate-disease' non-seminomatous germ-cell tumors. After PEB therapy the median time to relapse was 4 months (3-12). The median time to relapse after CEB therapy was also 4 months (2-38). No relapses more than 1 year after therapy have been observed in the PEB arm in contrast to three late relapses >2 years after therapy in the CEB arm. The approach to patients with relapsed disease was based on the decision of the individual treating physician without regard to whether they had received CEB or PEB treatment. Most patients received standard platinbased chemotherapy followed by high-dose chemo- Table 1. Characteristics of patients with 'good-risk' testicular cancer entered in a randomized trial of cisplatin- versus carboplatin-based first line chemotherapy. Number of patients 29 Median follow-up (months) 31 Range 9-44 Median age (years) 29 Range Stratification criteria Minimal disease + low markers' 17 (59%) Minimal disease + high markers' 4 (14%) Moderate disease + low markers 6 (21%) Moderate disease + high markers 2 (7%) N mclastatic sites 0 (marker elevation only) 3(10%) 1 19(66%) 2 5 (17%) 3 2 (7%) PEB CEB Total (60%) 32 (59%) 5(20%) 9(17%) 4(16%) 10(19%) 1 (4%) 3 (6%) 4(16%) 7(13%) 15 (60%) 34 (63%) 6 (24%) 11 (20%) 0 2 (4%) High markers - p"-hcg > IU/1 and/or AFP > 1000 ng/ ml. therapy and stem cell rescue. Results of treatment and current status are summarized in Table 2. With respect to toxic effects of treatment, 68% of patients after PEB in contrast to 47% after CEB had severe nausea and vomiting. Nephrotoxicity of mild degree (WHO grade 1) occurred in 16% of patients after PEB and in none after CEB treatment. Ototoxicity (WHO grade 1-2) occurred in 10% of patients in both treatment arms. Polyneuropathy occurred in 10% of PEB-treated patients, and no cases of Raynaud's syndrome and no severe pulmonary toxicity were reported. Hematotoxicity showed leukopenia WHO grade 3-4 in 50% of PEB- versus 26% of CEB-treated patients and severe thrombocytopenia WHO grade 3-4 in 22% of PEB- and 10% of CEB-treated patients. Dose modifications were performed with the deletion of bleomycin in 1 patient in each arm and a reduction of the dose of etoposide in 1 patient per treatment arm. In 3 patients receiving CEB the dose of carboplatin was escalated by 10% and in 2 patients carboplatin was decreased by 10% for non-symptomatic hematological toxicity (Table 3). Discussion -based chemotherapy regimens are active in 'good-risk' non-seminomatous germ-cell tumors and Table 2. Treatment results and current status in 'good-risk' testicular cancer patients treated with cisplatin- or carboplatin-based chemotherapy. Response to therapy p NED C PD Relapse Number of patients progression free at 1 year Death of disease No of patients with negative events 1 b c PEB 18 (65%) 5 (16%) 5(16%) 1 (3%) 4 (13%) 25 (86%) 1 (3%) CEB ("-25) 12 (48%) 7 (28%) 5 (20%) 1 (4%) 8 (32%) Total (n-54) 30 (55%) 12 (23%) 10(19%) 2 (3%) 12 (22%) 21 (84%) 46 (85%) 4 (16%) 5 (10%) 7(24%) 11(44%) 18(33%) CR after chemotherapy alone. Necrosis at secondary tumor resection. Resection of vital carcinoma (no evidence of disease). Table 3. Summary of non-hematological toxicity of PEB or CEBtreatment. PEB CEB Number of patients Nausea/vomiting (>WHO grade 2) 68% 47% Nephrotoxicity (WHO grade I) 16% 0% Ototoxicity (WHO grade 1-2) 10% 10% Polyneuropathy (WHO grade 1-2) 10% 0% Raynaud's syndrome - - Pulmonary toxicity - -

4 1018 therapy for 'good-risk' non-seminomatous germ-cell tumors. A trial by the Memorial Sloan-Kettering Cancer Center compared 4 cycles of carboplatin and etoposide (EC) versus cisplatin and etoposide (EP) in 270 patients with 'good-risk' germ-cell tumors [12]. Although no differences in response rates were seen between EC (88%) and EP (90%) therapy, 16 patients (12%) treated with EC have relapsed from CR versus 4 patients (3%) treated with EP. Event-free and relapsefree survival were inferior for patients treated with 4 cycles of etoposide and carboplatin. The difference in overall survival did not quite reach statistical significance. Recently these unfavourable results for the EC combination were confirmed by an International Joint Russian-American trial [13]. After a median follow-up of 52 months 6 of 23 (30%) 'good-risk' patients receiving etoposide and carboplatin have relapsed from CR. A significantly higher relapse rate and a lower proportion of patients with NED were observed for patients receiving EC than in those on EP therapy. Both studies clearly suggest that 4 cycles of carboplatin and etoposide are inferior to 4 cycles of the same dose of etoposide combined with cisplatin. However, some pitfalls remain in the interpretation of these trials: (1) a substantial number of patients with pure seminoma, a very chemosensitive histological entity of germ-cell cancers, was also included in this trial; (2) bleomycin, now considered an essential component of the treatment of non-seminomatous germ-cell tumors, was not included in any of the treatment arms [7, 14]; (3) the intervals between EC cycles were 28 days instead of the usual 21 days, resulting in a lower dose intensity for both carboplatin and etoposide compared to a standard 3-weekly regimen. However, a higher dose-intensity of carboplatin, which was the objective of the second part of the trial by the MSKCC investigators, with most patients reaching an area-under-the-curve (AUC) > 5 mg/m] x min, did not improve the results of the carboplatin arm, but was associated with significantly more myelotoxicity. Poor results using the combination of carboplatin and etoposide (EC) at 4-weekly intervals were also reported in a phase n trial by French investigators [15]. 0.8They treated 24 patients with 'good-risk' non-seminomatous germ-cell tumors with the EC combination and 0.6observed a complete response rate of 83%. Adverse c o> events occurred in 9 patients (37.5%) and after a meo.> 0.4dian follow-up of 24 months 4 patients had died, 3 k «patients were alive with progressive disease and only 17 O O) Q. 0) 0.2had no evidence of disease. However, the low dose-peb(n=29) o c intensity of carboplatin could not explain the inferior CEB(n=25) 0.0results in this trial, since the dosing of the carboplatin was retrospectively calculated according to renal function and all patients had received a median area-undermonths the-curve (AUC) of 6.1 mg/ml x min. No significant Figure 1. Proportion of patients without a negative event among 54 differences in AUC values could be observed between patients with 'good-risk' non-seminomatous testicular cancer treated either with 3 cycles of PEB or 4 cycles of CEB combination patients successfully treated and those failing treatment [15-17]. chemotherapy (log-rank test n.s.). jl L the problem of drug toxicity is currently the most important issue in the selection of chemotherapy regimens [1]., a clinically active cisplatin analogue, exhibits clinical advantages over those of its parent compound: it is not only less emetogenic but also has no significant nephro-, neuro- or ototoxicity when given at standard doses [9]. However, myelosuppression, the major toxicity of carboplatin, may compromise the dosing of carboplatin when it is combined with other myelotoxic drugs. The initial introduction of carboplatin in the treatment of 'good-risk' non-seminomatous germ-cell tumors is based on a phase II trial of the Royal Marsden Hospital. Of 67 patients with 'good-risk' non-seminomas treated with carboplatin, etoposide and bleomycin (CEB) only 5 (6.5%) experienced adverse events. was given at a dose of 120 mg/m2 on days 1-3, bleomycin 30 mg on days 2, 9 and 16 and carboplatin at doses between mg/ m2 calculated according to the renal function by the Calvert formula. The treatment failures observed in 5 patients were considered to be the consequence of an insufficient dosage of carboplatin (carboplatin <360 mg/m2) and all 5 of the patients were rendered diseasefree after salvage chemotherapy [10]. The favourable results reported for CEB therapy comprised the rationale for the current reported trial of CEB - vs. PEB - combination chemotherapy. However, in our direct randomized comparison of the Royal Marsden CEB regimen against a standard PEB regimen significantly more adverse events were seen in CEB-treated patients. The trial was terminated after recruitment of 54 patients. All patients entered on trial continued to be followed. After a median observation time of 33 months the negative event-free survival is still significantly worse for patients treated with CEB combination chemotherapy (Figure 1). In particular, after CEB therapy 4 of 25 patients have died of the disease in contrast to 1 of 29 patients after PEB therapy. There are 4 currently available randomized trials comparing carboplatin versus cisplatin-based chemo-

5 1019 The largest trial currently available comparing cisplatin- versus carboplatin-based chemotherapy was conducted by the UK Medical Research Council in cooperation with the EORTC [18]. This trial randomized 598 patients between 4 cycles of cisplatin, etoposide and bleomycin and 4 cycles of carboplatin, etoposide and bleomycin. and etoposide dosing were performed as in the previous Royal Marsden phase II trial, except that bleomycin was given only once per cycle (total dose 30 mg) in both treatment arms. The results demonstrate a significantly higher failure rate in patients on CEB (n - 299) than in those on PEB (n - 299) therapy (19 versus 10%; P = 0.02) and an inferior 1-year event-free survival (80% versus 90%). In the CEB arm 14 deaths (5%) have occurred in contrast to 4 deaths (1%) in PEB-treated patients. However, the fact that bleomycin was not used in the standard dose of 90 mg per cycle but was reduced to only 30 mg still hampers interpretation of this trial. Compared to the randomized MRC trial full-dose bleomycin was used in both treatment arms in our randomized study. In the CEB arm bleomycin was applied for the first 3 cycles until a total dose of 270 mg was reached, similar to the dose for patients treated with 3 cycles of PEB. dosage in our trial was based on the Calvert formula using a target area under the curve (AUC) of 5 mg/ml x min [11]. The total cumulative doses of etoposide applied in the two treatment arms were 1500 mg/m 2 (PEB) and 1440 mg/ m 2 (CEB) and the cumulative doses of bleomycin were 270 mg in both treatment arms, and both regimens were repeated at 3-weekly intervals. With die use of comparable doses of bleomycin and etoposide, the substitution of carboplatin for cisplatin achieved inferior results (Table 2). Most of the negative events observed in the current trial occurred in patients with moderate disease (69% of whom experienced a negative event) while a smaller percentage of patients with minimal disease (25%) experienced negative events (P< 0.005). The high proportion of patients in this trial with moderate disease (13 pts = 25%) as opposed to other low-risk trials may explain the overall unfavourable results in both treatment arms. Table 4 summarizes the inclusion criteria, treatment regimens and results in randomized trials of carboplatin in low-risk testicular cancer. Of particular concern in our patients treated with CEB were 3 late relapses observed at 38, 37 and 35 months. The elevation of tumor markers did not predict for the occurrence of negative events in our patients. Salvage treatment for patients relapsing after CEB appeared to be as difficult as after PEB therapy and only 3 of 7 patients after CEB could be rendered Table 4. Comparison of randomized trials of cisplatin- versus carboplatin-based chemotherapy for 'low-risk' metastatic testicular cancer. Institution MSKCC [12] Russian Cancer Center 13] UK Medical Research Council - EORTC [18] German Testicular Cancer Study Group Inclusion criteria - Seminoma stage II C, m - Extragonadal seminoma - MSKCC criteria 'good risk' with a chance of CR> Seminoma stage II C; HI - Extragonadal seminoma - MSKCC criteria 'good risk' with chance of CR > Recalculation according to Indiana University criteria - 'good-risk' non-seminoma according to Royal Marsden Classification - Indiana University criteria for 'minimal' and 'moderate' disease Regimens mg/m 2 dl 20 mg/m 2 dl mg/m 2 dl 20 mg/m 2 dl-5 AUC 5 dl 120 mg/m 2 30IUdl dl-3 50 mg/m 2 dl mg/m 2 dl-3 30 IU dl AUC 5 dl 120 mg/m 2 dl-3 30IUdl, 8, mg/m 2 dl-5 30IUdl, 8,15 v A- n x 4; q 21 d V 4* n x 4;q 21 d x 4; q 21 d X4;q 78 A ^o U 98,-1 U ^o 21 d x 4;q 21 d x 3;q 21 d No. of patients 111 i j i 134 TK j Failure/ negative events?4%" 13% 9% 20% c 10% 44%" 24% * Total number of negative events. b Number of relapses. c Failures at one year. d Patients with negative events at 3 years.

6 1020 disease-free by standard-dose cisplatin salvage chemotherapy and high-dose chemotherapy followed by autologous stem cell rescue. There was no significant difference in toxic effects between the two treatment arms and in fact, patients treated with 3 cycles of PEB experienced only moderate nephro-, oto- or neurotoxicity. Thus, this trial demonstrates that even with the use of adequate etoposide and bleomycin doses, substitution of carboplatin for cisplatin in the treatment of 'good-risk' non-seminomatous germ cell tumors remains problematical. Any attempt to explain the unfavourable results achieved with the CEB combination must include consideration of the dosage of carboplatin needed to achieve antitumor activity equal to that of a standard dose of cisplatin. On the basis of in vitro results and the antitumor activity of carboplatin against heterotransplanted human testicular cancer cell lines in a nude mouse model, it appears that a dose of 400 mg/m 2 of carboplatin is less active than 100 mg/m 2 of cisplatin [19, 20]. Even using the Calvert formula for carboplatin dosing based on glomerular filtration rate may not result in sufficient antitumor activity, at least if an AUC of 5 mg/ml x min is planned [16,17]. as a single agent is also investigated in patients with metastatic seminoma, a more chemosensitive tumor than non-seminomatous germ cell cancer [21, 22]. Randomized trials are evaluating the use of carboplatin versus cisplatin-based combination chemotherapy in patients with advanced seminoma. Toxicity aspects in the treatment of seminoma are even more important since these patients are usually 10 years older than patients treated for non-seminomatous tumors. Currently, carboplatin is also employed for high-dose chemotherapy in patients with relapsed testicular cancer. The absence of severe organ toxicity is the basis for the use of high-dose carboplatin in this treatment setting. However, high-dose carboplatin in combination with other cytotoxic agents is myelosuppressive and in order to apply this type of high-dose chemotherapy a hematopoietic support with the use of autologous bone marrow or peripheral blood stem cells is needed. The current role of carboplatin in the treatment of non-seminomatous germ cell cancer is therefore limited to the use of high-dose carboplatin (approximately mg/m 2, AUC of mg/ml x min) in combination with high-dose etoposide and possibly a third drug (ifosfamide or cyclophosphamide) for the treatment of relapsed germ cell tumors [23, 24]. There appears to be no role for standard-dose carboplatin in any combination regimen used in patients with 'good-prognosis' non-seminomatous germ cell tumors. References 1. Einhorn LH. Treatment of testicular cancer A new and improved model. J Clin Oncol 1990; 8: Bosl GJ, Geller NL, Cirricone C et al. Multivariate analysis of prognostic variables in patients with metastatic testicular cancer. Cancer Res 1983; 43: Vogelzang NJ. Prognostic factors in metastatic testicular cancer. Int J Androl 1987; 10: Birch R, Williams S, Cone A et aj. Prognostic factors for favourable outcome in disseminated germ cell tumors. J Clin Oncol 1986; 4: Williams S, Birch R, Einhorn LH et al. Treatment of dissiminated germ cell tumors with cisplatin, bleomycin and either vinblastin or etoposide. N Engl J Med 1987; 316(23): Einhorn LH, Wiliams SD, Loehrer PJ et al. A comparison of four versus three courses of cisplatin, VP-16 and bleomycin in favorable prognosis disseminated germ cell tumors. J Clin Oncol 1989; 7: Loehrer PJ Sr, Johnson D, Elson P et al. Importance of bleomycin in favorable-prognosis disseminated germ cell tumors: An Eastern Cooperative Oncology Group trial. J Clin Oncol 1995; 13: Vermorken JB, Bokkel Huinink ten WW, Eisenhauer EA et al. versus cisplatin. Ann Oncol 1993; 4 (Suppl 4): O'Reilly SM, Rustin GJS, Smith DB, Newlands ES. Single agent activity of carboplatin in patient with previously untreated non-seminomatous germ cell tumours. Ann Oncol 1992; 3: Horwich A, Dearnaley DP, Nicholls J et al. Effectiveness of carboplatin, etoposide and bleomycin combination chemotherapy in good-prognosis metastatic testicular nonseminomatous germ cell tumors. J Clin Oncol 1991; 9: Calvert AH, Newell DR, Gumbrell LA et al. dosage: Prospective evaluation of a simple formula based on renal function. J Clin Oncol 1989; 7: Bajorin DF, Sarosdy MF, Pfister DG et al. Randomized trial of etoposide and carboplatin in patients with good-risk germ cell tumors: A multiinstirutional study. J Clin Oncol 1993; 11: Tjulandin SA, Garin AM, Mescheryakov AA et al. etoposide and carboplatin-etoposide induction chemotherapy for good-risk patients with germ cell tumors. Ann Oncol 1993; 4: Culine S, Mahjoubi M, Philoppot I, Droz JP. Treatment of good-risk disseminated non-seminomatous germ cell tumours; the less bleomycin, the more cisplatin? Eur J Cancer 1991; 12: 1715 (Letter). 15. Kattan J, Mahjoubi M, Droz JP et al. High failure rate of carboplatin-etoposide combination in good risk non-seminomatous germ cell tumours. Eur J Cancer 1993;29A: Fish RG, Shelley MD, Griffiths H et al. A dosing scheme for carboplatin in adult cancer patients bases upon pre-infusion renal function and platelet count. Anticancer-Drugs 1994; 5: Childs WJ, Nicholls EJ, Horwich A. The optimisation of carboplatin dose in carboplatin, etoposide and bleomycin combination chemotherapy for good prognosis metastatic nonseminomatous germ cell tumours of the testis. Ann Oncol 1992; 3: Horwich A, Sleijfer D, Fossa S et al. on behalf of the UK Medical Research Council Testicular Tumour Working Party & EORTC Genito-Urinary Group. A trial of carboplatin-based combination chemotherapy in good prognosis metastatic testicular cancer. Proc Am Soc Clin Oncol 1994; 13: 709 (Abstr). 19. Bokemeyer C, Harstrick A, Schmoll HJ. The use of carboplatin in malignant germ cell tumours. Eur J Cancer 1994, 30A: Harstrick A, Caspar J, Guba R et al. Comparison of the antitumor activity of cisplatin, carboplatin and iproplatin against established human testicular cancer cell lines in vivo and in vitro. Cancer 1989; 63: Schmoll HJ, Harstrick A, Bokemeyer C et al. Single-agent carboplatin for advanced seminoma. A phase II study. Cancer 1993; 72:

7 Horwich A, Dearnaley DP, Duchesne GM et al. Simple nontoxic treatment of advanced metastatic seminoma with carboplatin. J Clin Oncol 1989; 7: Lampe H, Dearnaley DP, Price A et al. High-dose carboplatin and etoposide for salvage chemotherapy of germ cell tumours. Eur J Cancer 1995, 31A: Nichols CR, Tricot G, Williams S et al. Dose-intensive chemotherapy in refractrory germ cell cancer A phase I/O trial of high-dose carboplatin and etoposide with autologous bone marrow transplantation. J Clin Oncol 1989; 7: Received 2 July 1996; accepted 23 October Correspondence to: C. Bokemeyer, MD DepL of Internal Medicine University of Tubingen Otfried-Miiller-Str Tubingen Germany