- Email: [email protected]
Combination chemotherapy with gemcitabine, oxaliplatin, and paclitaxel in patients with cisplatin-refractory or multiply relapsed germ-cell tumors: a study of the German Testicular Cancer Study Group
original article
Annals of Oncology 19: 448–453, 2008 doi:10.1093/annonc/mdm526 Published online 15 November 2007
Combination chemotherapy with gemcitabine, oxaliplatin, and paclitaxel in patients with cisplatinrefractory or multiply relapsed germ-cell tumors: a study of the German Testicular Cancer Study Group C. Bokemeyer1 , K. Oechsle1 , F. Honecker1, F. Mayer3, J. T. Hartmann3, C. F. Waller4, I. Bo¨hlke1 & C. Kollmannsberger2,3* 1
Received 20 August 2007; revised 10 October 2007; accepted 15 October 2007
Background: The aim of this study is to determine feasibility and efficacy of the combination regimen gemcitabine, oxaliplatin, and paclitaxel (GOP) in patients with cisplatin-refractory or multiply relapsed germ-cell tumors.
original article
Patients and methods: From April 2003 to October 2006, 41 patients refractory to cisplatin-based chemotherapy or with relapse after high-dose chemotherapy (HDCT) plus stem-cell support (peripheral blood stem-cell transplantation: PBSCT) received 800 mg/m2 gemcitabine, 80 mg/m2 paclitaxel (Taxol), both on days 1 + 8, and oxaliplatin 130 mg/m2 on day 1 of a 3-week cycle for a minimum of two cycles. Primary end point was response rate. Patients were pretreated with a median of two lines of platin-based chemotherapy (range, 1–3), and 78% had relapsed after HDCT/PBSCT. Results: Seventy-three percent of patients had relapsed within 3 months after the last cisplatin-based chemotherapy. Five percent of the patients achieved a complete response, and 34% and 12% a marker-negative and marker-positive partial response, respectively (overall response rate 51%). After a median follow-up of 5 months (range, 0–20), 15% of the patients remain in complete remission after GOP chemotherapy 6 residual tumor resection with a median response duration of 8 months (1 to 17+). Main toxicity was leucocytopenia grade 3/4 in 15%, anemia in 7%, and thrombocytopenia in 49% of the patients. Conclusion: Combination chemotherapy with GOP is feasible and effective with acceptable toxicity in patients with treatment-refractory germ-cell tumors. Key words: cisplatin refractory, gemcitabine, germ-cell tumor, oxaliplatin, paclitaxel, relapsed
introduction Today, 70%–80% of patients with metastatic germ-cell tumors achieve long-term survival with cisplatin-based chemotherapy [1, 2]. Patients suffering from relapse after first-line cisplatin-based chemotherapy may still be cured in 10%–50% of cases with salvage chemotherapy. Patients progressing during or relapsing after salvage chemotherapy, however, exhibit a very poor prognosis, and long-term survival is rare [3]. Therefore, the evaluation of new active drugs and treatment combinations remains a priority for these patients. Various chemotherapeutic agents have been *Correspondence to: Dr C. Kollmannsberger, Division of Medical Oncology, British Columbia Cancer Agency, Vancouver Cancer Centre, University of British Columbia, 600 West 10th Avenue, Vancouver, British Columbia V5Z 4E6, Canada. Tel: +1 604-877-6000; Fax: +1 604-708-2144; E-mail: [email protected]
These authors contributed equally to this work.
evaluated in intensely pretreated or cisplatin-refractory patients. As single agents, only orally administered etoposide or i.v. paclitaxel, gemcitabine, and, most recently, oxaliplatin have been shown to be active with selected patients achieving prolonged survival [4–9]. The German Testicular Cancer Study Group (GTCSG) systematically conducted a series of consecutive phase II trials in heavily pretreated germ-cell tumor patients which tested a number of agents and combinations each on the basis of a preclinical rationale. Gemcitabine, oxaliplatin, and paclitaxel (GOP) were developed as single agents in consecutive phase II trials [6, 7, 9]. Paclitaxel has demonstrated its activity in refractory or relapsed germ-cell tumor patients in several clinical trials and has become part of different salvage combination chemotherapy regimen, e.g. in combination with platinum and ifosfamide [10, 11]. For single-agent gemcitabine, two phase II studies from Germany and the United States have shown response rates of 15%–20% in
ª The Author 2007. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please e-mail: [email protected]
Downloaded from http://annonc.oxfordjournals.org/ at Univ of Iowa-Law Library on June 23, 2015
Department of Oncology/Hematology with sections Bone Marrow Transplantation and Pneumology, University Medical Centre Eppendorf, Hamburg, Germany; Divisions of Medical Oncology, British Columbia Cancer Agency, Vancouver Cancer Centre, University of British Columbia, Vancouver, Canada; 3Department of Oncology/Hematology/Immunology/Rheumatology, University of Tuebingen, Medical Centre, Tuebingen; 4Department of Hematology/Oncology, University of Freiburg, Freiburg, Germany 2
original article
Annals of Oncology
of granulocytes during any cycle of study treatment was <500/ll, G-CSF was strongly recommended and if the nadir of thrombocytes was <50.000/ll, oxaliplatin and gemcitabine had to be reduced by 25% in all further cycles. Scheduling and duration of G-CSF therapy were left to the treating physician. Patients were taken off study if no hematopoetic regeneration occurred within 2 weeks after the end of the previous cycle. Both oxaliplatin and paclitaxel were reduced to 75% dose if peripheral neuropathy Common Toxicity Criteria (CTC) grade II occurred. Both drugs were stopped in case of neuropathy grade III or IV. Gemcitabine was continued regardless of any grade of peripheral neuropathy, but had to be stopped if serum creatinine increased 1.5-fold above the upper normal limit. In all other cases of nonhematologic toxic grade II–IV, apart from alopecia and nausea, study treatment had to be held until recovery of symptoms to CTC grade 0 or I and all three drugs were reduced by 25% in the subsequent cycles. Pretreatment evaluation consisted of medical history, assessment of the Karnofsky performance status and preexisting toxic effects, physical examination and electrocardiogram, and routine laboratory including blood cell counts, liver function tests, creatinine clearance, electrolytes, and tumor markers (AFP and b-HCG).
patients and methods Eligibility criteria for this open-label, multicenter phase II trial included histologically confirmed seminomatous or nonseminomatous germ-cell tumor, relapse within 3 months after cisplatin-based chemotherapy, tumor progression during or relapse after salvage high-dose chemotherapy (HDCT), tumor progression during salvage cisplatin-based chemotherapy, or ineligibility for cisplatin-based or HDCT due to severe comorbidities. Patients had to have measurable disease and documented tumor progression on X-ray, ultrasound, computed tomography, or magnetic resonance imaging. An elevation of alpha-fetoprotein (AFP) or beta-human chorionic gonadotropin (b-HCG) was accepted as measurable disease if an increase of at least 25% within 3 weeks before study inclusion had been observed. Further inclusion criteria were a Karnofsky performance status of at least 60%, a life expectancy of >3 months, and age between 18 and 70 years. An adequate bone marrow function (leukocytes > 2500/ll, thrombocytes > 100 000/ll), liver function (bilirubin level < 1.5-fold the upper normal limit), and renal function (serum creatinine clearance < 1.5-fold the upper normal limit) were mandatory. All patients gave written informed consent. The study was approved by the ethics committees of the Universities of Tuebingen and Hamburg-Eppendorf, Germany, as well as by the local ethics committees of the participating centers. Gemcitabine was given at a dose of 800 mg/m2 as a 30-min infusion and paclitaxel at a dosage of 80 mg/m2 as a 1-h infusion, both on days 1 and 8 of a 3-week cycle. Oxaliplatin was administered only on day 1 with 130 mg/m2 i.v. >2 h. Antiemetic prophylaxis was left to the decision of the treating physician, but a combination of a 5-HT3 antagonist and dexamethasone was proposed. Prophylaxis for paclitaxel included anaphylaxis with 20 mg dexamethasone i.v. for 30 min and 50 mg diphenhydramine plus 300 mg cimetidine i.v. for 20 min before the infusion of paclitaxel was mandatory. Patients responding to study treatment were planned to receive two cycles beyond the best response up to a maximum of eight cycles. GOP chemotherapy was administered at full dose on days 1 and 8 if leukocytes were >3000/ll and thrombocytes >100 000/ll. If thrombocytes were 50 000–100 000/ll or leukocytes 1500–3000/ll on day 8 of the chemotherapy cycle, the doses of gemcitabine and paclitaxel were reduced to 75%. In case of thrombocytes <50 000/ll or leukocytes <1500/ll on day 8, study treatment was hold until complete regeneration and a prophylaxis with hematopoetic growth factors [granulocyte colonystimulating factor (G-CSF)] was given in subsequent cycles. If the nadir
Volume 19 | No. 3 | March 2008
definitions Disease was considered cisplatin refractory when at least tumor stabilization or a remission had been achieved, but tumor progression occurred again within 4 weeks of the last cisplatin-based chemotherapy. Disease was considered absolutely cisplatin refractory when tumor progression had occurred while patients were receiving cisplatin-based therapy [15, 16].
response assessment Response assessment was carried out following every other cycle. Toxicity was assessed after each cycle. Response and toxicity were graded according to World Health Organization and National Cancer Institute—CTC (version 2.0) criteria, respectively. In addition, patients with reduction of the size of a tumor lesion and normalization of previously elevated tumor markers were considered partial remission (PR) with tumor marker normalization (PR negative). A reduction ‡50% in the sum of the perpendicular diameters of measurable disease plus a tumor marker decrease for at least 1 month, but without complete normalization, was considered as a marker-positive PR (PR positive). If elevated markers were the only evidence of progressive disease, a decrease of at least 90% was required for a PR. Levels of serum tumor markers were measured every 3 weeks. All responses as well as the diagnosis of stable disease had to be confirmed after a 4-week interval. All patients were scheduled to receive at least two cycles of treatment. If the clinical situation after one cycle indicated tumor progression and if in these cases a significant marker (50%) and/or radiological progression (25%), however, had occurred already after one cycle, the treatment was stopped and the patient was classified as having progressive disease.
statistical considerations The primary end point of this study was response rate. Tolerability, progression free, and overall survival time represented secondary end points. An optimal Simon two-stage design was used to determine the number of patients required [17]. Assuming a response rate of clinical interest of ‡40%, a minimal response rate of 20%, a probability of 5% for rejecting an active drug combination (type II error), and a probability of 20% to further evaluate an ineffective drug combination (type I error), 18 patients had to be enrolled into the first cohort. If <4 responses to study therapy were observed among the first 18 patients, the study was to be terminated. If an objective remission occurred in at least
doi:10.1093/annonc/mdm526 | 449
Downloaded from http://annonc.oxfordjournals.org/ at Univ of Iowa-Law Library on June 23, 2015
refractory patients [7, 8]. On the basis of preclinical investigations indicating an incomplete cross-resistance between oxaliplatin and cisplatin [12], a GTCSG phase II trial evaluated oxaliplatin given at a dose of 130 mg/m2 every 2 weeks in cisplatin-refractory or relapsed patients, yielding a response rate of 19% [9]. With single-agent activity for both gemcitabine and oxaliplatin and in vitro analyses demonstrating supraadditive synergistic effects of oxaliplatin in combination with gemcitabine, both agents were combined within a subsequent GTCSG phase II trial [13, 14]. These two drug regimens were proven to be feasible with acceptable toxicity in this heavily pretreated patient group and a high response rate of 46% was observed including several long-term survivors [14]. These results formed the rationale for the present trial which evaluated the combination of all three active agents, GOP, in patients with relapsed or refractory germ-cell tumors.
original article four patients, the study was to be continued with a second cohort of 20 patients.
results
450 | Bokemeyer et al.
Table 1. Patient characteristics, n = 41 patients No. of patients Age (median) 38 years UICC stage at primary diagnosis I 2 II 4 IIIB 7 IIIC 21 Not evaluable 8 Location of primary tumor Gonadal 32 Retroperitoneal 6 Mediastinal 3 Histology Pure seminoma 2 Embryonal carcinoma 10 Yolk sac tumor 3 Choriocarcinoma 3 Immature teratoma 7 Mixed nonseminomatous 16 IGCCCG stage at initial diagnosis Good 6 Intermediate 11 Poor 22 NA 2 Previous lines of platin-based 2 chemotherapy (median) Previous paclitaxel 10 Previous high-dose chemotherapy 32 (median) A primary diagnosis 16 Salvage high-dose chemotherapy 16 Refractoriness after previous platin-based chemotherapy PD during chemotherapy 5 PD within 4 weeks 9 PD within 5–8 weeks 1 PD within 9–12 weeks 15 PD >3 months 11 Late relapse 3 Metastases at study entry Lungs 29 Liver 11 Bones 4 Central nervous system 4 Retroperitoneal lymph nodes 18 Distant lymph nodes 14 Other 9 Elevation of tumor markers at study entry AFP 11 Median AFP value 3257 kU/l b-HCG 21 Median b-HCG value 816 U/l LDH 26 Median LHD value 586 U/l
% Range, 25–62 5 10 17 51 19 78 14 8 5 24 7 7 17 39 6 27 54 4 Range, 1–4 24 78 39 39
12 23 2 36 27 7 71 27 10 10 44 34 22 27 Range, 36–71 560 51 Range, 10–108 721 63 Range, 266–3200
UICC, International Union Against Cancer; NA, not available; PD, progressive disease; AFP, alpha-fetoprotein; b-HCG, beta-human chorionic gonadotropin; LDH, lactate dehydrogenase; IGCCCG, International Germ Cell Cancer Cooperative Group.
Volume 19 | No. 3 | March 2008
Downloaded from http://annonc.oxfordjournals.org/ at Univ of Iowa-Law Library on June 23, 2015
Forty-one patients with heavily pretreated or cisplatinrefractory nonseminomatous germ-cell cancer were entered into the study from April 2003 to October 2006. Patient characteristics are listed in Table 1. All patients had previously received a median number of two (range, 1–3) platinum-based regimens. Seventy-eight percent of patients had relapsed after HDCT with peripheral blood stemcell transplantation either within the primary treatment (39%) or as salvage regimen (39%). Five patients (12%) had progressed during cisplatin-based chemotherapy (absolute cisplatin refractory) and nine (23%) had progressed within 4 weeks after completion of cisplatin-based chemotherapy (cisplatin refractory). Another 38% of patients had tumor progression between 5 and 12 weeks after cisplatin-based chemotherapy. At the time of study inclusion, 10% of patients each had bone and central nervous system metastases. A total of 150 cycles of GOP were applied with a median number of four cycles per patient (range, 1–7). Study treatment was stopped due to tumor progression in 54% and for withdrawal of consent or unacceptable toxicity in 5% of the patients. A complete remission (CR) was observed in 5% of the patients, 34% achieved a marker-negative, and 12% a tumor marker-positive PR as best response, resulting in an overall response rate of 51% (95% confidence interval 35% to 67%). The first of the two CR patients had initially presented with a stage IIC nonseminoma and had received three cycles of cisplatin/etoposide/bleomycin (PEB). He attained a markernegative remission and underwent residual tumor resection showing necrosis only. He relapsed 6 months later and while receiving salvage etoposide/ifosfamide/cisplatin (VIP) chemotherapy, the patient progressed further. Classified as absolutely cisplatin-refractory disease, he was started on GOP and achieved a continuous CR. The second patient initially presented with good prognosis metastatic disease. He progressed shortly after three cycles of PEB and went on to salvage HDCT. He relapsed a second time 7 months after HDCT and subsequently achieved a continuous CR to GOP. In five other responding patients, a complete tumor resection was feasible after completion of treatment, resulting in an overall rate of complete responses of 17%. Another three responding patients had an incomplete resection of their residual lesions. They completed treatment and are alive with disease but progression free at the time of last follow-up. Table 2 summarizes the characteristics of all six long-term responding patients. After a median follow-up of 5 months (range, 0–20), 17% of the patients are still disease free with a median response duration of 8 months (range, 3 to 17+), 24% are alive with disease, 57% have died due to disease, and one patient (2%) committed suicide. Median progression-free survival time was 3 months (range, 1 to 17+) and the median overall survival was 6 months (range, 1 to 19+). Combination chemotherapy with GOP was given in 51% of cycles without dose reduction and in 79% without treatment
Annals of Oncology
original article
PR2 CR PD 5 7 0 4· PEB; 1· VIP + 3· HD-CE 3· PEB; 1· VIP + 3· HD-CE 4· PEB; 4· VIP 4 5 6
CR, complete remission; GOP, gemcitabine, oxaliplatin, and paclitaxel; PEB, cisplatin/etoposide/bleomycin; VIP, etoposide/ifosfamide/cisplatin; HD-CE, high-dose carboplatin/etoposide; PR2, marker-negative partial remission; b-HCG, beta-human chorionic gonadotropin; HD-CEC, high-dose carboplatin/etoposide/cyclophosphamide; LDH, lactate dehydrogenase; PD, progressive disease; TI, Taxol/Ifosfamide.
CR CR CR 9+ 6+ 6+ LDH: 590 b-HCG: 170; LDH: 1104 LDH: 1385
PR2 CR CR
Yes No No
9+ 6+ 6+
CR CR CR 11+ 15+ 17+ 11+ 15+ 13 Yes Yes Yes 5 3 19 4· PEB; 1· VIP + 3· HD-CE 4· VIP; 2· TI + HD-CE 3· PEB; 3· VIP + 1· HD-CEC 1 2 3
PR2 PR2 CR
Lungs Lungs Mediastinal + retroperitoneal lymph nodes Mediastinal lymph nodes and lungs Mediastinal lymph nodes, lungs, pleura Retroperitoneal lymph nodes, lungs, and liver
b-HCG: 332 b-HCG: 5076 LDH: 581
PR2 PR2 PR2
Time to progression (months) Elevated marker at study entry (in U/l) Site of metastases at study entry Response to last cisplatinbased chemotherapy Time to relapse (months) Prior therapy Patient no.
Table 2. Characteristics of patients achieving a continuous CR after chemotherapy 6 salvage surgery
Volume 19 | No. 3 | March 2008
delay in this heavily pretreated patient population. Overall, dose reductions were necessary due to grades 2–4 toxicity in 66% of patients with a large number of these dose reductions carried out for grade 2 toxicity (Table 3). Treatment was delayed in 41% of patients. CTC grade 3 or 4 thrombocytopenia occurred in 49% of the patients (34% grade 3 and 15% grade 4), but no severe hemorrhagic complication was observed. Anemia grade 3/4 and leucocytopenia were documented in only 7% and 15% of the patients, respectively. No severe infections were observed. Nonhematologic toxicity was favorable with nausea/emesis, peripheral neurotoxicity, and diarrhea grade 3 occurring in one patient each (2%). No toxic death occurred in this heavily pretreated population.
discussion Despite the overall success of cisplatin-based chemotherapy in patients with metastatic germ-cell cancer, the prognosis for cisplatin-refractory patients remains poor. Currently, no standard salvage chemotherapy exists for these patients and treatment options are very limited [3, 18]. Various doublet combinations of the active agents GOP have been tested in clinical trials in recent years [14, 19–24]. The present study is the first successfully completed triple combination study in heavily pretreated refractory patients. The majority of our patients were either truly cisplatin refractory or had progressed within 12 weeks after cisplatinbased therapy (73%) or even HDCT (78%), indicating a high prevalence of chemotherapy resistance and particularly poor prognosis of these patients. Despite these unfavorable criteria, an encouraging response rate of 51% was observed including 39% complete or marker-negative partial responses. These results compare favorably to the response rates observed in the trials utilizing two drug regimens, in particular, if complete and marker-negative remission are considered (Table 4). Six of our patients (15%) remain in prolonged CR including two patients who achieved a CR without subsequent resection and one patient who relapsed 4 months after chemotherapy plus resection and was again rendered tumor free by a second salvage surgery (Table 2). Of note, five of these six patients had previously failed salvage HDCT with autologous stem-cell transplantation. One of the two patients who achieved a complete response without subsequent resection was deemed absolutely cisplatin refractory due to progression during cisplatin-based secondline salvage chemotherapy. Promising activity was also observed with the combination of gemcitabine and oxaliplatin in two phase II trials with response rates of 32% and 46% and complete responses resulting in long-term survival in 10%–12% of patients [14, 21]. Pectasides reported a high response rate with the combination of oxaliplatin and irinotecan; however, these patients had not been pretreated with HDCT and a different definition of resistant disease was used. Einhorn et al. [25] recently described 13% prolonged CRs after gemcitabine and paclitaxel in patients who had failed HDCT in a retrospective analysis. It remains currently unclear whether the addition of a third drug confers a benefit compared with two drug regimens. This question
doi:10.1093/annonc/mdm526 | 451
Downloaded from http://annonc.oxfordjournals.org/ at Univ of Iowa-Law Library on June 23, 2015
Best response to GOP
Secondary resection post-GOP
Overall survival (months)
Status
Annals of Oncology
original article
452 | Bokemeyer et al.
All patients ‘resistant’ defined as failure to achieve a durable complete remission to a cisplatin-based regimen and one or more of the following unfavorable prognostic features: relapse after CR or progression within 4 weeks after first-line therapy; poor or no response to prior conventional dose cisplatin/ifosfamide therapy; and extragonadal primary site. b Refractory being defined as a disease progression during or within 2 months after cisplatin-based chemotherapy. CI, confidence interval; CR, complete remission; PR, partial remission; NS, not significant; HD-CT, high-dose chemotherapy.
a
to 67%)
49%) 73%) 64%) 53%) 55%) 64%) to to to to to to
12% 27% 30% 16% 16% 17% NS) 35% CI CI CI CI CI CI CI CI (95% (95% (95% (95% (95% (95% (95% (95% 21% 50% 46% 32% 30% 40% 17% 51% 36% NSa 63% 100% 61%b All patients resistanta 100% 35%
could only be answered within a randomized trial. Unfortunately, given the rare patient population, randomized trials will almost be impossible to conduct. Achieving response rates as high as possible in refractory patients, however, is clinically also important since the induction of a remission may subsequently allow for the complete resection of residual masses and thus offer the potential for long-term survival. The complete resection of residual masses after a chemotherapy-induced remission seems to be an important part in the treatment of these patients, as demonstrated by a retrospective analysis from Indiana University [18]. In this specific analysis of 101 patients with relapsed and/or cisplatin-refractory disease, only 5% of patients achieved long-term survival. All these had surgery of metastatic disease as a vital part of their salvage treatment. A high response rate as achieved with the GOP regimen may allow for a higher number of patients to be resected after chemotherapy. This triple regimen with its high response rate can therefore serve as a therapeutic option for patients with refractory germ-cell cancer and in particular for patients with a high need for a tumor response or patients who may be amenable to surgery after tumor shrinkage. The GOP regimen was feasible with acceptable toxicity, although a number of patients underwent dose reductions and to a lesser extent dose delays. The relatively high number of dose reductions and delays can be explained by the cautious dose reduction/delay schedule which mandated changes already for grade 2 toxic effects. In particular, no excess neurotoxicity in these heavily cisplatin-pretreated patients was observed despite the combination of two potentially neurotoxic drugs. As expected in such an intensively pretreated patient population, thrombocytopenia was the most frequent grade 3/4 toxicity and occurred in 49% of patients. No hemorrhagic complications, however, occurred. In contrast, a small study of nine patients from Italy utilizing GOP given weekly in germ-cell tumor patients found this combination not to be feasible due to extensive
36% 22% 89% 14% NS 0% 22% 78%
(1%)
0 4 (3%) 8 (5%) 0 0 0 0
NS NS 29% 10% NS 6% 11% NS
(7%) (4%) (21%) (1%) (1%)
28 18 35 28 26 18 18 41
11 6 31 1 1 0 2
Paclitaxel/gemcitabine Irinotecan/cisplatin Gemcitabine/oxaliplatin Gemcitabine/oxaliplatin Paclitaxel/oxaliplatin Oxaliplatin/irinotecan Oxaliplatin/gemcitabine Oxaliplatin/gemcitabine/ paclitaxel
(22%) (21%) (22%) (9%) (23%) (1%) (1%)
Hinton et al. [19] Miki et al. [23] Kollmannsberger et al. [14] Pectasides et al. [21] Theodore et al. [22] Pectasides et al. [20] Di Georgi et al. [24] Present study
33 32 33 14 34 2 2
(2%)
Overall response rate (CI)
0 1 (2%) 6 (15%) 0 0 0 0
Cisplatin-refractory disease
(15%) (5%) (34%) (2%) (2%)
Patients pretreated with HD-CT
6 2 14 1 1 0 1
Late relapse (>2 years)
(39%) (37%) (22%) (12%) (27%) (2%) (5%)
No.of patients
16 15 9 5 11 1 2
Treatment
Grade IV
Author
Grade III
Volume 19 | No. 3 | March 2008
Downloaded from http://annonc.oxfordjournals.org/ at Univ of Iowa-Law Library on June 23, 2015
Toxic effects per patient (n = 41) Leukocytopenia 7 (17%) Anemia 17 (41%) Thrombocytopenia 6 (15%) Neurotoxicity 8 (20%) Nausea/emesis 8 (20%) Fever 3 (7%) Diarrhea 1 (2%) Toxic effects per cycle (n = 150) Leukocytopenia 39 (26%) Anemia 79 (53%) Thrombocytopenia 45 (30%) Neurotoxicity 27 (18%) Nausea/emesis 21 (14%) Fever 3 (2%) Diarrhea 1 (1%)
Grade II
Table 4. Comparison of patient characteristics and outcomes in trials of combination regimens for the treatment of refractory germ-cell cancer
Grade I
CR/PR marker negative
Table 3. Toxic effects per patient (n = 41) and per cycle (n = 150)
10%, PR2: NS CR: 9%, PR2: 36% CR: 9%, PR2: 9% CR: 14%, PR2: NS CR: 0, PR2: 4% CR: 22%, PR2: NS CR: 5%, PR2: 5% CR: 5%, PR2: 34%
Annals of Oncology
Annals of Oncology
funding Sanofi-Aventis.
references 1. Bosl GJ, Motzer RJ. Medical progress: testicular germ-cell cancer. N Engl J Med 1997; 337: 242–253. 2. Einhorn LH. Curing metastatic testicular cancer. Proc Natl Acad Sci USA 2002; 99: 4592–4595. 3. Kollmannsberger C, Nichols C, Bokemeyer C. Recent advances in management of patients with platinum-refractory testicular germ cell tumors. Cancer 2006; 106: 1217–1226. 4. Miller JC, Einhorn LH. Phase II study of daily oral etoposide in refractory germ cell tumors. Semin Oncol 1990; 17: 36–39. 5. Motzer RJ, Bajorin D, Schwartz LH et al. Phase II trial of paclitaxel shows antitumor activity in patients with previously treated germ cell tumors. J Clin Oncol 1994; 12: 2277–2283. 6. Bokemeyer C, Beyer J, Metzner B et al. Phase II study of paclitaxel in patients with relapsed or cisplatin-refractory testicular cancer. Ann Oncol 1996; 7: 31–34. 7. Bokemeyer C, Gerl A, Scho¨ffski P et al. Gemcitabine in patients with relapsed or cisplatin-refractory testicular cancer. J Clin Oncol 1999; 17: 512–516. 8. Einhorn LH, Stender MJ, Williams SD. Phase II trial of gemcitabine in refractory germ cell tumors. J Clin Oncol 1999; 17: 509–511. 9. Kollmannsberger C, Rick O, Derigs H-G et al. Activity of oxaliplatin in patients with relapsed or cisplatin-refractory germ cell cancer: a study of the German Testicular Cancer Study Group. J Clin Oncol 2002; 20: 2031–2037.
Volume 19 | No. 3 | March 2008
10. Motzer RJ, Mazumdar M, Sheinfeld J et al. Sequential dose-intensive paclitaxel, ifosfamide, carboplatin, and etoposide salvage therapy for germ cell tumor patients. J Clin Oncol 2000; 18: 1173–1180. 11. Rick O, Bokemeyer C, Beyer J et al. Salvage treatment with paclitaxel, ifosfamide, and cisplatin plus high-dose carboplatin, etoposide, and thiotepa followed by autologous stell cell rescue in patients with relapsed or refractory germ cell cancer. J Clin Oncol 2001; 19: 81–88. 12. Dunn TA, Schmoll HJ, Gru¨nwald V et al. Comparative cytotoxicity of oxaliplatin and cisplatin in non-seminomatous germ cell lines. Invest New Drugs 1997; 15: 109–114. 13. Faivre S, Raymond E, Woynarowski JM, Cvitkovic E. Supraadditive effect of 2#,2#-difluorodeoxycytidine (gemcitabine) in combination with oxaliplatin in human cancer cell lines. Cancer Chemother Pharmacol 1999; 44: 117–123. 14. Kollmannsberger C, Beyer J, Liersch R et al. Combination chemotherapy with gemcitabine plus oxaliplatin in patients with intensively pretreated or refractory germ cell cancer: a study of the German Testicular Cancer Study Group. J Clin Oncol 2004; 22: 108–114. 15. Nichols CR, Tricot G, Williams SD et al. Dose-intensive chemotherapy in refractory germ cell cancer—a phase I/II trial of high-dose carboplatin and etoposide with autologous bone marrow transplantation. J Clin Oncol 1989 7: 932–939. 16. Beyer J, Kramar A, Mandanas R et al. High-dose chemotherapy as salvage treatment in germ cell tumors: a multivariate analysis of prognostic variables. J Clin Oncol 1996; 14: 2638–2645. 17. Simon R. Optimal two-stage designs for phase II clinical trials. Controlled Clinical Trtials 1989; 10: 1–10. 18. Porcu P, Bhatia S, Einhorn LH. Results of treatment after relapse from high-dose chemotherapy in germ cell tumors. J Clin Oncol 2000; 18: 1181–1186. 19. Hinton S, Catalano PJ, Einhorn L et al. Phase II trial of paclitaxel and gemcitabine in refractory germ cell tumors. J Clin Oncol 2001; 20: 1859–1863. 20. Pectasides D, Pectasides M, Farmakis D et al. Oxaliplatin and irinotecan plus granulocyte-colony stimulating factor as third-line treatment in relapsed or cisplatin-refractory germ-cell tumor patients: a phase II study. Eur Urol 2004; 46: 216–221. 21. Pectasides D, Pectasides M, Farmakis D et al. Gemcitabine and oxaliplatin (GEMOX) in patients with cisplatin-refractory germ cell tumors: a phase II study. Ann Oncol 2004; 15: 493–497. 22. Theodore C, Flechon A, Fizazi K et al. A phase II multicentre study of oxaliplatin (Ox) in combination with paclitaxel (Px) in patients (pts) who failed cisplatin (CDDP) based chemotherapy (CT) for germ cell tumors (GCT). Proc Am Soc Clin Oncol 2004; 23: 389 (Abstr 4534). 23. Miki T, Mizutani Y, Nonomura N et al. Irinotecan plus cisplatin has substantial antitumor effect as salvage chemotherapy against germ cell tumors. Cancer 2002; 95: 1879–1885. 24. De Giorgi U, Rosti G, Aieta M et al. Phase II study of oxaliplatin and gemcitabine salvage chemotherapy in patients with cisplatin-refractory nonseminomatous germ cell tumor. Eur urol 2006; 50: 1032–1038, discussion 1038–1039. 25. Einhorn LH, Brames MJ, Juliar B, Williams SD. Phase II study of paclitaxel plus gemcitabine salvage chemotherapy for germ cell tumors after progression following high-dose chemotherapy with tandem transplant. J Clin Oncol 2007; 25: 513–516. 26. De Giorgi U, Rosti G, Papiani G et al. Weekly gemcitabine, paclitaxel, oxaliplatin combination chemotherapy in patients with cisplatin-refractory germ cell tumor: preliminary experience. Am J Clin Oncol 2004; 27: 457–460. 27. de Wit R, Louwerens M, de Mulder PHM et al. Management of intermediate prognosis germ cell cancer: results of a phase I/II study of taxol-BEP. Int J Cancer 1999; 83: 831–833.
doi:10.1093/annonc/mdm526 | 453
Downloaded from http://annonc.oxfordjournals.org/ at Univ of Iowa-Law Library on June 23, 2015
hematotoxicity, in particular severe thrombocytopenia [26]. Since these patients were comparably heavily pretreated to our patients, the differences in schedule, with gemcitabine given at a dose of 800 mg/m2, paclitaxel at 70 mg/m2, and oxaliplatin at 50 mg/m2 once weekly for three consecutive weeks followed by 1 week rest, may have been responsible for the differences in hematotoxicity. In summary, the triple combination consisting of GOP given as described in this report demonstrates a high activity in platinum-refractory germ-cell tumor patients and may offer a chance for prolonged or even long-term survival in selected patients even after failure of multiple lines of treatment including HDCT. Further development of this regimen in earlier lines of therapy may, however, be challenging. Paclitaxel is already widely used in the second-line setting and is currently being explored as an addition to the BEP regimen in intermediate prognosis patients [27]. All investigations of novel regimens will face the challenge of defining the appropriate clinical setting for second- or even first-line therapy in germcell cancer patients. Clinical situation and inclusion criteria will have to be very carefully chosen. New combinations, however, hold the promise to take us another step forward in treating germ-cell cancer.
original article
Annals of Oncology 19: 448–453, 2008 doi:10.1093/annonc/mdm526 Published online 15 November 2007
Combination chemotherapy with gemcitabine, oxaliplatin, and paclitaxel in patients with cisplatinrefractory or multiply relapsed germ-cell tumors: a study of the German Testicular Cancer Study Group C. Bokemeyer1 , K. Oechsle1 , F. Honecker1, F. Mayer3, J. T. Hartmann3, C. F. Waller4, I. Bo¨hlke1 & C. Kollmannsberger2,3* 1
Received 20 August 2007; revised 10 October 2007; accepted 15 October 2007
Background: The aim of this study is to determine feasibility and efficacy of the combination regimen gemcitabine, oxaliplatin, and paclitaxel (GOP) in patients with cisplatin-refractory or multiply relapsed germ-cell tumors.
original article
Patients and methods: From April 2003 to October 2006, 41 patients refractory to cisplatin-based chemotherapy or with relapse after high-dose chemotherapy (HDCT) plus stem-cell support (peripheral blood stem-cell transplantation: PBSCT) received 800 mg/m2 gemcitabine, 80 mg/m2 paclitaxel (Taxol), both on days 1 + 8, and oxaliplatin 130 mg/m2 on day 1 of a 3-week cycle for a minimum of two cycles. Primary end point was response rate. Patients were pretreated with a median of two lines of platin-based chemotherapy (range, 1–3), and 78% had relapsed after HDCT/PBSCT. Results: Seventy-three percent of patients had relapsed within 3 months after the last cisplatin-based chemotherapy. Five percent of the patients achieved a complete response, and 34% and 12% a marker-negative and marker-positive partial response, respectively (overall response rate 51%). After a median follow-up of 5 months (range, 0–20), 15% of the patients remain in complete remission after GOP chemotherapy 6 residual tumor resection with a median response duration of 8 months (1 to 17+). Main toxicity was leucocytopenia grade 3/4 in 15%, anemia in 7%, and thrombocytopenia in 49% of the patients. Conclusion: Combination chemotherapy with GOP is feasible and effective with acceptable toxicity in patients with treatment-refractory germ-cell tumors. Key words: cisplatin refractory, gemcitabine, germ-cell tumor, oxaliplatin, paclitaxel, relapsed
introduction Today, 70%–80% of patients with metastatic germ-cell tumors achieve long-term survival with cisplatin-based chemotherapy [1, 2]. Patients suffering from relapse after first-line cisplatin-based chemotherapy may still be cured in 10%–50% of cases with salvage chemotherapy. Patients progressing during or relapsing after salvage chemotherapy, however, exhibit a very poor prognosis, and long-term survival is rare [3]. Therefore, the evaluation of new active drugs and treatment combinations remains a priority for these patients. Various chemotherapeutic agents have been *Correspondence to: Dr C. Kollmannsberger, Division of Medical Oncology, British Columbia Cancer Agency, Vancouver Cancer Centre, University of British Columbia, 600 West 10th Avenue, Vancouver, British Columbia V5Z 4E6, Canada. Tel: +1 604-877-6000; Fax: +1 604-708-2144; E-mail: [email protected]
These authors contributed equally to this work.
evaluated in intensely pretreated or cisplatin-refractory patients. As single agents, only orally administered etoposide or i.v. paclitaxel, gemcitabine, and, most recently, oxaliplatin have been shown to be active with selected patients achieving prolonged survival [4–9]. The German Testicular Cancer Study Group (GTCSG) systematically conducted a series of consecutive phase II trials in heavily pretreated germ-cell tumor patients which tested a number of agents and combinations each on the basis of a preclinical rationale. Gemcitabine, oxaliplatin, and paclitaxel (GOP) were developed as single agents in consecutive phase II trials [6, 7, 9]. Paclitaxel has demonstrated its activity in refractory or relapsed germ-cell tumor patients in several clinical trials and has become part of different salvage combination chemotherapy regimen, e.g. in combination with platinum and ifosfamide [10, 11]. For single-agent gemcitabine, two phase II studies from Germany and the United States have shown response rates of 15%–20% in
ª The Author 2007. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please e-mail: [email protected]
Downloaded from http://annonc.oxfordjournals.org/ at Univ of Iowa-Law Library on June 23, 2015
Department of Oncology/Hematology with sections Bone Marrow Transplantation and Pneumology, University Medical Centre Eppendorf, Hamburg, Germany; Divisions of Medical Oncology, British Columbia Cancer Agency, Vancouver Cancer Centre, University of British Columbia, Vancouver, Canada; 3Department of Oncology/Hematology/Immunology/Rheumatology, University of Tuebingen, Medical Centre, Tuebingen; 4Department of Hematology/Oncology, University of Freiburg, Freiburg, Germany 2
original article
Annals of Oncology
of granulocytes during any cycle of study treatment was <500/ll, G-CSF was strongly recommended and if the nadir of thrombocytes was <50.000/ll, oxaliplatin and gemcitabine had to be reduced by 25% in all further cycles. Scheduling and duration of G-CSF therapy were left to the treating physician. Patients were taken off study if no hematopoetic regeneration occurred within 2 weeks after the end of the previous cycle. Both oxaliplatin and paclitaxel were reduced to 75% dose if peripheral neuropathy Common Toxicity Criteria (CTC) grade II occurred. Both drugs were stopped in case of neuropathy grade III or IV. Gemcitabine was continued regardless of any grade of peripheral neuropathy, but had to be stopped if serum creatinine increased 1.5-fold above the upper normal limit. In all other cases of nonhematologic toxic grade II–IV, apart from alopecia and nausea, study treatment had to be held until recovery of symptoms to CTC grade 0 or I and all three drugs were reduced by 25% in the subsequent cycles. Pretreatment evaluation consisted of medical history, assessment of the Karnofsky performance status and preexisting toxic effects, physical examination and electrocardiogram, and routine laboratory including blood cell counts, liver function tests, creatinine clearance, electrolytes, and tumor markers (AFP and b-HCG).
patients and methods Eligibility criteria for this open-label, multicenter phase II trial included histologically confirmed seminomatous or nonseminomatous germ-cell tumor, relapse within 3 months after cisplatin-based chemotherapy, tumor progression during or relapse after salvage high-dose chemotherapy (HDCT), tumor progression during salvage cisplatin-based chemotherapy, or ineligibility for cisplatin-based or HDCT due to severe comorbidities. Patients had to have measurable disease and documented tumor progression on X-ray, ultrasound, computed tomography, or magnetic resonance imaging. An elevation of alpha-fetoprotein (AFP) or beta-human chorionic gonadotropin (b-HCG) was accepted as measurable disease if an increase of at least 25% within 3 weeks before study inclusion had been observed. Further inclusion criteria were a Karnofsky performance status of at least 60%, a life expectancy of >3 months, and age between 18 and 70 years. An adequate bone marrow function (leukocytes > 2500/ll, thrombocytes > 100 000/ll), liver function (bilirubin level < 1.5-fold the upper normal limit), and renal function (serum creatinine clearance < 1.5-fold the upper normal limit) were mandatory. All patients gave written informed consent. The study was approved by the ethics committees of the Universities of Tuebingen and Hamburg-Eppendorf, Germany, as well as by the local ethics committees of the participating centers. Gemcitabine was given at a dose of 800 mg/m2 as a 30-min infusion and paclitaxel at a dosage of 80 mg/m2 as a 1-h infusion, both on days 1 and 8 of a 3-week cycle. Oxaliplatin was administered only on day 1 with 130 mg/m2 i.v. >2 h. Antiemetic prophylaxis was left to the decision of the treating physician, but a combination of a 5-HT3 antagonist and dexamethasone was proposed. Prophylaxis for paclitaxel included anaphylaxis with 20 mg dexamethasone i.v. for 30 min and 50 mg diphenhydramine plus 300 mg cimetidine i.v. for 20 min before the infusion of paclitaxel was mandatory. Patients responding to study treatment were planned to receive two cycles beyond the best response up to a maximum of eight cycles. GOP chemotherapy was administered at full dose on days 1 and 8 if leukocytes were >3000/ll and thrombocytes >100 000/ll. If thrombocytes were 50 000–100 000/ll or leukocytes 1500–3000/ll on day 8 of the chemotherapy cycle, the doses of gemcitabine and paclitaxel were reduced to 75%. In case of thrombocytes <50 000/ll or leukocytes <1500/ll on day 8, study treatment was hold until complete regeneration and a prophylaxis with hematopoetic growth factors [granulocyte colonystimulating factor (G-CSF)] was given in subsequent cycles. If the nadir
Volume 19 | No. 3 | March 2008
definitions Disease was considered cisplatin refractory when at least tumor stabilization or a remission had been achieved, but tumor progression occurred again within 4 weeks of the last cisplatin-based chemotherapy. Disease was considered absolutely cisplatin refractory when tumor progression had occurred while patients were receiving cisplatin-based therapy [15, 16].
response assessment Response assessment was carried out following every other cycle. Toxicity was assessed after each cycle. Response and toxicity were graded according to World Health Organization and National Cancer Institute—CTC (version 2.0) criteria, respectively. In addition, patients with reduction of the size of a tumor lesion and normalization of previously elevated tumor markers were considered partial remission (PR) with tumor marker normalization (PR negative). A reduction ‡50% in the sum of the perpendicular diameters of measurable disease plus a tumor marker decrease for at least 1 month, but without complete normalization, was considered as a marker-positive PR (PR positive). If elevated markers were the only evidence of progressive disease, a decrease of at least 90% was required for a PR. Levels of serum tumor markers were measured every 3 weeks. All responses as well as the diagnosis of stable disease had to be confirmed after a 4-week interval. All patients were scheduled to receive at least two cycles of treatment. If the clinical situation after one cycle indicated tumor progression and if in these cases a significant marker (50%) and/or radiological progression (25%), however, had occurred already after one cycle, the treatment was stopped and the patient was classified as having progressive disease.
statistical considerations The primary end point of this study was response rate. Tolerability, progression free, and overall survival time represented secondary end points. An optimal Simon two-stage design was used to determine the number of patients required [17]. Assuming a response rate of clinical interest of ‡40%, a minimal response rate of 20%, a probability of 5% for rejecting an active drug combination (type II error), and a probability of 20% to further evaluate an ineffective drug combination (type I error), 18 patients had to be enrolled into the first cohort. If <4 responses to study therapy were observed among the first 18 patients, the study was to be terminated. If an objective remission occurred in at least
doi:10.1093/annonc/mdm526 | 449
Downloaded from http://annonc.oxfordjournals.org/ at Univ of Iowa-Law Library on June 23, 2015
refractory patients [7, 8]. On the basis of preclinical investigations indicating an incomplete cross-resistance between oxaliplatin and cisplatin [12], a GTCSG phase II trial evaluated oxaliplatin given at a dose of 130 mg/m2 every 2 weeks in cisplatin-refractory or relapsed patients, yielding a response rate of 19% [9]. With single-agent activity for both gemcitabine and oxaliplatin and in vitro analyses demonstrating supraadditive synergistic effects of oxaliplatin in combination with gemcitabine, both agents were combined within a subsequent GTCSG phase II trial [13, 14]. These two drug regimens were proven to be feasible with acceptable toxicity in this heavily pretreated patient group and a high response rate of 46% was observed including several long-term survivors [14]. These results formed the rationale for the present trial which evaluated the combination of all three active agents, GOP, in patients with relapsed or refractory germ-cell tumors.
original article four patients, the study was to be continued with a second cohort of 20 patients.
results
450 | Bokemeyer et al.
Table 1. Patient characteristics, n = 41 patients No. of patients Age (median) 38 years UICC stage at primary diagnosis I 2 II 4 IIIB 7 IIIC 21 Not evaluable 8 Location of primary tumor Gonadal 32 Retroperitoneal 6 Mediastinal 3 Histology Pure seminoma 2 Embryonal carcinoma 10 Yolk sac tumor 3 Choriocarcinoma 3 Immature teratoma 7 Mixed nonseminomatous 16 IGCCCG stage at initial diagnosis Good 6 Intermediate 11 Poor 22 NA 2 Previous lines of platin-based 2 chemotherapy (median) Previous paclitaxel 10 Previous high-dose chemotherapy 32 (median) A primary diagnosis 16 Salvage high-dose chemotherapy 16 Refractoriness after previous platin-based chemotherapy PD during chemotherapy 5 PD within 4 weeks 9 PD within 5–8 weeks 1 PD within 9–12 weeks 15 PD >3 months 11 Late relapse 3 Metastases at study entry Lungs 29 Liver 11 Bones 4 Central nervous system 4 Retroperitoneal lymph nodes 18 Distant lymph nodes 14 Other 9 Elevation of tumor markers at study entry AFP 11 Median AFP value 3257 kU/l b-HCG 21 Median b-HCG value 816 U/l LDH 26 Median LHD value 586 U/l
% Range, 25–62 5 10 17 51 19 78 14 8 5 24 7 7 17 39 6 27 54 4 Range, 1–4 24 78 39 39
12 23 2 36 27 7 71 27 10 10 44 34 22 27 Range, 36–71 560 51 Range, 10–108 721 63 Range, 266–3200
UICC, International Union Against Cancer; NA, not available; PD, progressive disease; AFP, alpha-fetoprotein; b-HCG, beta-human chorionic gonadotropin; LDH, lactate dehydrogenase; IGCCCG, International Germ Cell Cancer Cooperative Group.
Volume 19 | No. 3 | March 2008
Downloaded from http://annonc.oxfordjournals.org/ at Univ of Iowa-Law Library on June 23, 2015
Forty-one patients with heavily pretreated or cisplatinrefractory nonseminomatous germ-cell cancer were entered into the study from April 2003 to October 2006. Patient characteristics are listed in Table 1. All patients had previously received a median number of two (range, 1–3) platinum-based regimens. Seventy-eight percent of patients had relapsed after HDCT with peripheral blood stemcell transplantation either within the primary treatment (39%) or as salvage regimen (39%). Five patients (12%) had progressed during cisplatin-based chemotherapy (absolute cisplatin refractory) and nine (23%) had progressed within 4 weeks after completion of cisplatin-based chemotherapy (cisplatin refractory). Another 38% of patients had tumor progression between 5 and 12 weeks after cisplatin-based chemotherapy. At the time of study inclusion, 10% of patients each had bone and central nervous system metastases. A total of 150 cycles of GOP were applied with a median number of four cycles per patient (range, 1–7). Study treatment was stopped due to tumor progression in 54% and for withdrawal of consent or unacceptable toxicity in 5% of the patients. A complete remission (CR) was observed in 5% of the patients, 34% achieved a marker-negative, and 12% a tumor marker-positive PR as best response, resulting in an overall response rate of 51% (95% confidence interval 35% to 67%). The first of the two CR patients had initially presented with a stage IIC nonseminoma and had received three cycles of cisplatin/etoposide/bleomycin (PEB). He attained a markernegative remission and underwent residual tumor resection showing necrosis only. He relapsed 6 months later and while receiving salvage etoposide/ifosfamide/cisplatin (VIP) chemotherapy, the patient progressed further. Classified as absolutely cisplatin-refractory disease, he was started on GOP and achieved a continuous CR. The second patient initially presented with good prognosis metastatic disease. He progressed shortly after three cycles of PEB and went on to salvage HDCT. He relapsed a second time 7 months after HDCT and subsequently achieved a continuous CR to GOP. In five other responding patients, a complete tumor resection was feasible after completion of treatment, resulting in an overall rate of complete responses of 17%. Another three responding patients had an incomplete resection of their residual lesions. They completed treatment and are alive with disease but progression free at the time of last follow-up. Table 2 summarizes the characteristics of all six long-term responding patients. After a median follow-up of 5 months (range, 0–20), 17% of the patients are still disease free with a median response duration of 8 months (range, 3 to 17+), 24% are alive with disease, 57% have died due to disease, and one patient (2%) committed suicide. Median progression-free survival time was 3 months (range, 1 to 17+) and the median overall survival was 6 months (range, 1 to 19+). Combination chemotherapy with GOP was given in 51% of cycles without dose reduction and in 79% without treatment
Annals of Oncology
original article
PR2 CR PD 5 7 0 4· PEB; 1· VIP + 3· HD-CE 3· PEB; 1· VIP + 3· HD-CE 4· PEB; 4· VIP 4 5 6
CR, complete remission; GOP, gemcitabine, oxaliplatin, and paclitaxel; PEB, cisplatin/etoposide/bleomycin; VIP, etoposide/ifosfamide/cisplatin; HD-CE, high-dose carboplatin/etoposide; PR2, marker-negative partial remission; b-HCG, beta-human chorionic gonadotropin; HD-CEC, high-dose carboplatin/etoposide/cyclophosphamide; LDH, lactate dehydrogenase; PD, progressive disease; TI, Taxol/Ifosfamide.
CR CR CR 9+ 6+ 6+ LDH: 590 b-HCG: 170; LDH: 1104 LDH: 1385
PR2 CR CR
Yes No No
9+ 6+ 6+
CR CR CR 11+ 15+ 17+ 11+ 15+ 13 Yes Yes Yes 5 3 19 4· PEB; 1· VIP + 3· HD-CE 4· VIP; 2· TI + HD-CE 3· PEB; 3· VIP + 1· HD-CEC 1 2 3
PR2 PR2 CR
Lungs Lungs Mediastinal + retroperitoneal lymph nodes Mediastinal lymph nodes and lungs Mediastinal lymph nodes, lungs, pleura Retroperitoneal lymph nodes, lungs, and liver
b-HCG: 332 b-HCG: 5076 LDH: 581
PR2 PR2 PR2
Time to progression (months) Elevated marker at study entry (in U/l) Site of metastases at study entry Response to last cisplatinbased chemotherapy Time to relapse (months) Prior therapy Patient no.
Table 2. Characteristics of patients achieving a continuous CR after chemotherapy 6 salvage surgery
Volume 19 | No. 3 | March 2008
delay in this heavily pretreated patient population. Overall, dose reductions were necessary due to grades 2–4 toxicity in 66% of patients with a large number of these dose reductions carried out for grade 2 toxicity (Table 3). Treatment was delayed in 41% of patients. CTC grade 3 or 4 thrombocytopenia occurred in 49% of the patients (34% grade 3 and 15% grade 4), but no severe hemorrhagic complication was observed. Anemia grade 3/4 and leucocytopenia were documented in only 7% and 15% of the patients, respectively. No severe infections were observed. Nonhematologic toxicity was favorable with nausea/emesis, peripheral neurotoxicity, and diarrhea grade 3 occurring in one patient each (2%). No toxic death occurred in this heavily pretreated population.
discussion Despite the overall success of cisplatin-based chemotherapy in patients with metastatic germ-cell cancer, the prognosis for cisplatin-refractory patients remains poor. Currently, no standard salvage chemotherapy exists for these patients and treatment options are very limited [3, 18]. Various doublet combinations of the active agents GOP have been tested in clinical trials in recent years [14, 19–24]. The present study is the first successfully completed triple combination study in heavily pretreated refractory patients. The majority of our patients were either truly cisplatin refractory or had progressed within 12 weeks after cisplatinbased therapy (73%) or even HDCT (78%), indicating a high prevalence of chemotherapy resistance and particularly poor prognosis of these patients. Despite these unfavorable criteria, an encouraging response rate of 51% was observed including 39% complete or marker-negative partial responses. These results compare favorably to the response rates observed in the trials utilizing two drug regimens, in particular, if complete and marker-negative remission are considered (Table 4). Six of our patients (15%) remain in prolonged CR including two patients who achieved a CR without subsequent resection and one patient who relapsed 4 months after chemotherapy plus resection and was again rendered tumor free by a second salvage surgery (Table 2). Of note, five of these six patients had previously failed salvage HDCT with autologous stem-cell transplantation. One of the two patients who achieved a complete response without subsequent resection was deemed absolutely cisplatin refractory due to progression during cisplatin-based secondline salvage chemotherapy. Promising activity was also observed with the combination of gemcitabine and oxaliplatin in two phase II trials with response rates of 32% and 46% and complete responses resulting in long-term survival in 10%–12% of patients [14, 21]. Pectasides reported a high response rate with the combination of oxaliplatin and irinotecan; however, these patients had not been pretreated with HDCT and a different definition of resistant disease was used. Einhorn et al. [25] recently described 13% prolonged CRs after gemcitabine and paclitaxel in patients who had failed HDCT in a retrospective analysis. It remains currently unclear whether the addition of a third drug confers a benefit compared with two drug regimens. This question
doi:10.1093/annonc/mdm526 | 451
Downloaded from http://annonc.oxfordjournals.org/ at Univ of Iowa-Law Library on June 23, 2015
Best response to GOP
Secondary resection post-GOP
Overall survival (months)
Status
Annals of Oncology
original article
452 | Bokemeyer et al.
All patients ‘resistant’ defined as failure to achieve a durable complete remission to a cisplatin-based regimen and one or more of the following unfavorable prognostic features: relapse after CR or progression within 4 weeks after first-line therapy; poor or no response to prior conventional dose cisplatin/ifosfamide therapy; and extragonadal primary site. b Refractory being defined as a disease progression during or within 2 months after cisplatin-based chemotherapy. CI, confidence interval; CR, complete remission; PR, partial remission; NS, not significant; HD-CT, high-dose chemotherapy.
a
to 67%)
49%) 73%) 64%) 53%) 55%) 64%) to to to to to to
12% 27% 30% 16% 16% 17% NS) 35% CI CI CI CI CI CI CI CI (95% (95% (95% (95% (95% (95% (95% (95% 21% 50% 46% 32% 30% 40% 17% 51% 36% NSa 63% 100% 61%b All patients resistanta 100% 35%
could only be answered within a randomized trial. Unfortunately, given the rare patient population, randomized trials will almost be impossible to conduct. Achieving response rates as high as possible in refractory patients, however, is clinically also important since the induction of a remission may subsequently allow for the complete resection of residual masses and thus offer the potential for long-term survival. The complete resection of residual masses after a chemotherapy-induced remission seems to be an important part in the treatment of these patients, as demonstrated by a retrospective analysis from Indiana University [18]. In this specific analysis of 101 patients with relapsed and/or cisplatin-refractory disease, only 5% of patients achieved long-term survival. All these had surgery of metastatic disease as a vital part of their salvage treatment. A high response rate as achieved with the GOP regimen may allow for a higher number of patients to be resected after chemotherapy. This triple regimen with its high response rate can therefore serve as a therapeutic option for patients with refractory germ-cell cancer and in particular for patients with a high need for a tumor response or patients who may be amenable to surgery after tumor shrinkage. The GOP regimen was feasible with acceptable toxicity, although a number of patients underwent dose reductions and to a lesser extent dose delays. The relatively high number of dose reductions and delays can be explained by the cautious dose reduction/delay schedule which mandated changes already for grade 2 toxic effects. In particular, no excess neurotoxicity in these heavily cisplatin-pretreated patients was observed despite the combination of two potentially neurotoxic drugs. As expected in such an intensively pretreated patient population, thrombocytopenia was the most frequent grade 3/4 toxicity and occurred in 49% of patients. No hemorrhagic complications, however, occurred. In contrast, a small study of nine patients from Italy utilizing GOP given weekly in germ-cell tumor patients found this combination not to be feasible due to extensive
36% 22% 89% 14% NS 0% 22% 78%
(1%)
0 4 (3%) 8 (5%) 0 0 0 0
NS NS 29% 10% NS 6% 11% NS
(7%) (4%) (21%) (1%) (1%)
28 18 35 28 26 18 18 41
11 6 31 1 1 0 2
Paclitaxel/gemcitabine Irinotecan/cisplatin Gemcitabine/oxaliplatin Gemcitabine/oxaliplatin Paclitaxel/oxaliplatin Oxaliplatin/irinotecan Oxaliplatin/gemcitabine Oxaliplatin/gemcitabine/ paclitaxel
(22%) (21%) (22%) (9%) (23%) (1%) (1%)
Hinton et al. [19] Miki et al. [23] Kollmannsberger et al. [14] Pectasides et al. [21] Theodore et al. [22] Pectasides et al. [20] Di Georgi et al. [24] Present study
33 32 33 14 34 2 2
(2%)
Overall response rate (CI)
0 1 (2%) 6 (15%) 0 0 0 0
Cisplatin-refractory disease
(15%) (5%) (34%) (2%) (2%)
Patients pretreated with HD-CT
6 2 14 1 1 0 1
Late relapse (>2 years)
(39%) (37%) (22%) (12%) (27%) (2%) (5%)
No.of patients
16 15 9 5 11 1 2
Treatment
Grade IV
Author
Grade III
Volume 19 | No. 3 | March 2008
Downloaded from http://annonc.oxfordjournals.org/ at Univ of Iowa-Law Library on June 23, 2015
Toxic effects per patient (n = 41) Leukocytopenia 7 (17%) Anemia 17 (41%) Thrombocytopenia 6 (15%) Neurotoxicity 8 (20%) Nausea/emesis 8 (20%) Fever 3 (7%) Diarrhea 1 (2%) Toxic effects per cycle (n = 150) Leukocytopenia 39 (26%) Anemia 79 (53%) Thrombocytopenia 45 (30%) Neurotoxicity 27 (18%) Nausea/emesis 21 (14%) Fever 3 (2%) Diarrhea 1 (1%)
Grade II
Table 4. Comparison of patient characteristics and outcomes in trials of combination regimens for the treatment of refractory germ-cell cancer
Grade I
CR/PR marker negative
Table 3. Toxic effects per patient (n = 41) and per cycle (n = 150)
10%, PR2: NS CR: 9%, PR2: 36% CR: 9%, PR2: 9% CR: 14%, PR2: NS CR: 0, PR2: 4% CR: 22%, PR2: NS CR: 5%, PR2: 5% CR: 5%, PR2: 34%
Annals of Oncology
Annals of Oncology
funding Sanofi-Aventis.
references 1. Bosl GJ, Motzer RJ. Medical progress: testicular germ-cell cancer. N Engl J Med 1997; 337: 242–253. 2. Einhorn LH. Curing metastatic testicular cancer. Proc Natl Acad Sci USA 2002; 99: 4592–4595. 3. Kollmannsberger C, Nichols C, Bokemeyer C. Recent advances in management of patients with platinum-refractory testicular germ cell tumors. Cancer 2006; 106: 1217–1226. 4. Miller JC, Einhorn LH. Phase II study of daily oral etoposide in refractory germ cell tumors. Semin Oncol 1990; 17: 36–39. 5. Motzer RJ, Bajorin D, Schwartz LH et al. Phase II trial of paclitaxel shows antitumor activity in patients with previously treated germ cell tumors. J Clin Oncol 1994; 12: 2277–2283. 6. Bokemeyer C, Beyer J, Metzner B et al. Phase II study of paclitaxel in patients with relapsed or cisplatin-refractory testicular cancer. Ann Oncol 1996; 7: 31–34. 7. Bokemeyer C, Gerl A, Scho¨ffski P et al. Gemcitabine in patients with relapsed or cisplatin-refractory testicular cancer. J Clin Oncol 1999; 17: 512–516. 8. Einhorn LH, Stender MJ, Williams SD. Phase II trial of gemcitabine in refractory germ cell tumors. J Clin Oncol 1999; 17: 509–511. 9. Kollmannsberger C, Rick O, Derigs H-G et al. Activity of oxaliplatin in patients with relapsed or cisplatin-refractory germ cell cancer: a study of the German Testicular Cancer Study Group. J Clin Oncol 2002; 20: 2031–2037.
Volume 19 | No. 3 | March 2008
10. Motzer RJ, Mazumdar M, Sheinfeld J et al. Sequential dose-intensive paclitaxel, ifosfamide, carboplatin, and etoposide salvage therapy for germ cell tumor patients. J Clin Oncol 2000; 18: 1173–1180. 11. Rick O, Bokemeyer C, Beyer J et al. Salvage treatment with paclitaxel, ifosfamide, and cisplatin plus high-dose carboplatin, etoposide, and thiotepa followed by autologous stell cell rescue in patients with relapsed or refractory germ cell cancer. J Clin Oncol 2001; 19: 81–88. 12. Dunn TA, Schmoll HJ, Gru¨nwald V et al. Comparative cytotoxicity of oxaliplatin and cisplatin in non-seminomatous germ cell lines. Invest New Drugs 1997; 15: 109–114. 13. Faivre S, Raymond E, Woynarowski JM, Cvitkovic E. Supraadditive effect of 2#,2#-difluorodeoxycytidine (gemcitabine) in combination with oxaliplatin in human cancer cell lines. Cancer Chemother Pharmacol 1999; 44: 117–123. 14. Kollmannsberger C, Beyer J, Liersch R et al. Combination chemotherapy with gemcitabine plus oxaliplatin in patients with intensively pretreated or refractory germ cell cancer: a study of the German Testicular Cancer Study Group. J Clin Oncol 2004; 22: 108–114. 15. Nichols CR, Tricot G, Williams SD et al. Dose-intensive chemotherapy in refractory germ cell cancer—a phase I/II trial of high-dose carboplatin and etoposide with autologous bone marrow transplantation. J Clin Oncol 1989 7: 932–939. 16. Beyer J, Kramar A, Mandanas R et al. High-dose chemotherapy as salvage treatment in germ cell tumors: a multivariate analysis of prognostic variables. J Clin Oncol 1996; 14: 2638–2645. 17. Simon R. Optimal two-stage designs for phase II clinical trials. Controlled Clinical Trtials 1989; 10: 1–10. 18. Porcu P, Bhatia S, Einhorn LH. Results of treatment after relapse from high-dose chemotherapy in germ cell tumors. J Clin Oncol 2000; 18: 1181–1186. 19. Hinton S, Catalano PJ, Einhorn L et al. Phase II trial of paclitaxel and gemcitabine in refractory germ cell tumors. J Clin Oncol 2001; 20: 1859–1863. 20. Pectasides D, Pectasides M, Farmakis D et al. Oxaliplatin and irinotecan plus granulocyte-colony stimulating factor as third-line treatment in relapsed or cisplatin-refractory germ-cell tumor patients: a phase II study. Eur Urol 2004; 46: 216–221. 21. Pectasides D, Pectasides M, Farmakis D et al. Gemcitabine and oxaliplatin (GEMOX) in patients with cisplatin-refractory germ cell tumors: a phase II study. Ann Oncol 2004; 15: 493–497. 22. Theodore C, Flechon A, Fizazi K et al. A phase II multicentre study of oxaliplatin (Ox) in combination with paclitaxel (Px) in patients (pts) who failed cisplatin (CDDP) based chemotherapy (CT) for germ cell tumors (GCT). Proc Am Soc Clin Oncol 2004; 23: 389 (Abstr 4534). 23. Miki T, Mizutani Y, Nonomura N et al. Irinotecan plus cisplatin has substantial antitumor effect as salvage chemotherapy against germ cell tumors. Cancer 2002; 95: 1879–1885. 24. De Giorgi U, Rosti G, Aieta M et al. Phase II study of oxaliplatin and gemcitabine salvage chemotherapy in patients with cisplatin-refractory nonseminomatous germ cell tumor. Eur urol 2006; 50: 1032–1038, discussion 1038–1039. 25. Einhorn LH, Brames MJ, Juliar B, Williams SD. Phase II study of paclitaxel plus gemcitabine salvage chemotherapy for germ cell tumors after progression following high-dose chemotherapy with tandem transplant. J Clin Oncol 2007; 25: 513–516. 26. De Giorgi U, Rosti G, Papiani G et al. Weekly gemcitabine, paclitaxel, oxaliplatin combination chemotherapy in patients with cisplatin-refractory germ cell tumor: preliminary experience. Am J Clin Oncol 2004; 27: 457–460. 27. de Wit R, Louwerens M, de Mulder PHM et al. Management of intermediate prognosis germ cell cancer: results of a phase I/II study of taxol-BEP. Int J Cancer 1999; 83: 831–833.
doi:10.1093/annonc/mdm526 | 453
Downloaded from http://annonc.oxfordjournals.org/ at Univ of Iowa-Law Library on June 23, 2015
hematotoxicity, in particular severe thrombocytopenia [26]. Since these patients were comparably heavily pretreated to our patients, the differences in schedule, with gemcitabine given at a dose of 800 mg/m2, paclitaxel at 70 mg/m2, and oxaliplatin at 50 mg/m2 once weekly for three consecutive weeks followed by 1 week rest, may have been responsible for the differences in hematotoxicity. In summary, the triple combination consisting of GOP given as described in this report demonstrates a high activity in platinum-refractory germ-cell tumor patients and may offer a chance for prolonged or even long-term survival in selected patients even after failure of multiple lines of treatment including HDCT. Further development of this regimen in earlier lines of therapy may, however, be challenging. Paclitaxel is already widely used in the second-line setting and is currently being explored as an addition to the BEP regimen in intermediate prognosis patients [27]. All investigations of novel regimens will face the challenge of defining the appropriate clinical setting for second- or even first-line therapy in germcell cancer patients. Clinical situation and inclusion criteria will have to be very carefully chosen. New combinations, however, hold the promise to take us another step forward in treating germ-cell cancer.
original article