Recent advances in management of patients with platinum-refractory testicular germ cell tumors


  • Christian Kollmannsberger M.D.,

    Corresponding author
    1. Division of Medical Oncology, British Columbia Cancer Agency, University of British Columbia–Vancouver Cancer Center, Vancouver, British Columbia, Canada
    • Division of Medical Oncology, British Columbia Cancer Agency–Vancouver Cancer Centre, 600 West 10th Avenue, Vancouver, BC V5Z 4E6, Canada===

    Search for more papers by this author
    • Fax: (604) 708-2144

  • Craig Nichols M.D.,

    1. Division of Hematology/Medical Oncology, Oregon Health Science University, Portland, Oregon
    Search for more papers by this author
  • Carsten Bokemeyer M.D.

    1. Division of Oncology, Hematology and Bone Marrow Transplantation, University Medical Center, Hamburg, Germany
    Search for more papers by this author


With the use of cisplatin-based combination chemotherapy, metastatic testicular germ cell tumors can be cured in 70–80% of patients, but patients refractory to cisplatin-based chemotherapy continue to have a very poor prognosis. Various chemotherapeutic agents have been evaluated in intensively pretreated or cisplatin-refractory patients, but as single agents, only orally administered etoposide, paclitaxel, gemcitabine, and, most recently, oxaliplatin have been shown to be active with selected patients achieving complete remissions. This has for the first time lead to clinical evaluation of combination chemotherapy regimens such as gemcitabine–paclitaxel or oxaliplatin–gemcitabine, demonstrating the feasibility of combination therapy in these heavily pretreated patients. High response rates of up to 45% were observed in particular with the latter combination. Salvage surgery remains a very important treatment option for patients with resectable disease. The molecular mechanisms of cisplatin resistance have been intensively studied, and several mechanisms have been discussed such as a decreased intracellular concentration of the drug, increased repair of the drug-induced damage, or an altered apoptotic response to this damage. This increasing knowledge may now allow design of new therapeutic options. Ongoing studies in refractory germ cell tumors are evaluating 3-drug regimens such as gemcitabine–paclitaxel–oxaliplatin but also biologic approaches such as inhibitors of the epidermal growth factor receptor or the vascular endothelial growth factor. This research may eventually allow the development of a noncross-resistant multidrug combination regimen that can be evaluated in an earlier line of therapy. Cancer 2006. © 2006 American Cancer Society.

Today, approximately 70–80% of all patients with metastatic germ cell cancer will achieve cure after standard-dose, cisplatin-based combination chemotherapy, such as PEB (cisplatin, etoposide, bleomycin) or VIP (etoposide, ifosfamide, cisplatin).1 Patients who relapse after cisplatin-based first-line chemotherapy have a less favorable prognosis, although 20–25% of patients are still cured with conventional second-line cisplatin-based chemotherapy.2, 3 High-dose chemotherapy plus autologous peripheral blood stem cell transplantation (ASCT) has become an accepted treatment option for patients who relapse after conventional chemotherapy. Durable responses are achieved in 20–60% of patients depending upon different prognostic factors of the patient population treated.4, 5 However, patients who relapse after conventional or high-dose salvage chemotherapy have a very poor prognosis, particularly those with cisplatin-refractory or absolutely refractory disease.6 Platinum-refractory disease is defined as response or disease stabilization during chemotherapy but disease progression within 4 weeks after cisplatin-based chemotherapy, whereas absolute cisplatin-refractory disease refers to patients with progression during cisplatin-based treatment, that is, they are not responsive to the treatment.5, 7 Identifying new treatment options with significant antitumor activity in these young patients with refractory disease remains a priority. Oral etoposide has, for a long time, been the only therapeutic option with a definite, but moderate, activity in patients with refractory germ cell cancer.8 In the past decade, several new drugs have been systematically investigated, but only recently, combination regimens have been studied for the first time in these heavily pretreated patients. Most refractory patients have been intensively pretreated with multiple conventional and/or high-dose chemotherapy cycles, so patient tolerance of new agents in this difficult therapeutic situation requires further evaluation. Underlying molecular mechanisms of cisplatin resistance are still incompletely understood, but several studies in recent years have provided more detailed insight into this complex problem.9, 10

Mechanisms of Platinum Resistance

Despite various studies, no uniform hypothesis has been developed to explain the exquisite chemosensitivity of most germ cell tumors (GCTs) or the chemoresistance of the minority of malignant GCTs. Multiple factors on different cellular levels seem to play roles in induction of cell death after cisplatin-based chemotherapy, but knowledge of the molecular basis of chemoresistance development is still poorly understood.

Cisplatin is believed to kill cells through interaction with DNA, mainly by formation of various DNA adducts, which lead to initiation of apoptosis.11 The tumor cell can escape initiation of apoptosis on several levels. First, cisplatin can be inactivated by changes in the level of thiol-containing cell compounds, such as glutathione or metallothionein, or it can be exported from the cell by several export pumps, even before it reaches DNA. Second, cisplatin-induced DNA damage can be repaired, predominantly by the so-called nuclear excision repair pathway (NER) before activation of the apoptotic cascade; Third, recognition of critical DNA damage, by mechanisms that initiate apoptosis, can fail. Fourth, the execution of apoptotic cell death may be prevented by antiapoptotic signals or by defects of apoptosis effectors.

Several studies, predominantly performed in cell lines and xenograft models, have suggested a correlation between glutathione and metallothionein levels and cisplatin resistance as well as between various export pumps, such as ABC transporters, lung-resistant protein, and cisplatin resistance.12–15 However, validation on clinical tumor samples is generally lacking. Investigations of the role of nuclear excision repair (NER), which is thought to be the most important DNA repair mechanism for cisplatin-induced damage, indicate a low capacity of GCT cells for NER. The low intrinsic capacity of NER demonstrated in GCT cell lines has been attributed to low levels of xeroderma pigmentosum complementation Group A protein (XPA) and the NER protein ERCC1. Alternatively, it has been proposed that DNA adducts could be concealed by testis-specific high-mobility group (HMG) box proteins, which would, then, prevent NER factors from performing damage detection and repair.9, 16, 17 The finding of a low NER capacity itself and its potential clinical relevance have not been confirmed in samples from patients with GCTs, yet it is conceivable that a low NER activity contributes to overall chemosensitivity of GCTs. A high level of wild-type p53 in GCTs has commonly been regarded as the biologic explanation for GCT chemosensitivity.18 Two recent studies, which investigated the role of p53 in refractory GCT, have demonstrated that p53 mutations are rare even after treatment, that these mutations are unlikely to cause chemotherapy resistance, and that the inactivation of p53 does not lead to a sensitivity change to chemotherapy in vitro.19, 20 These findings suggest that induction of apoptosis in testis cancer cells can be executed independently, without p53 involvement; therefore, the previously suggested important role of p53 in cisplatin resistance is questionable. A high incidence of microsatellite instability (MSI) was found in samples of patients with refractory GCT.21 In most refractory cases, MSI was found in several loci, whereas in unselected GCT, only very few patients had an MSI and most of them just in 1 locus. MSI in refractory GCT indicates that defects in the DNA mismatch repair pathway may represent a clinically relevant resistance mechanism. Other proteins involved in regulation of apoptosis, such as BAX, BCL-2, BCL-XL, and others, have also been investigated, but no single factor has correlated with treatment response.22 More research is necessary to conclusively define the role of antiapoptotic regulators downstream of the initiation of apoptosis in chemotherapy-resistant GCTs.

Unfortunately, most models described above are based predominantly on in vitro analyses of cell lines and largely lack confirmation of their relevance in clinical material.

Drugs with No or Minor Activity in Refractory GCTs

A large number of different agents were evaluated in patients with refractory testicular cancer, mostly based on promising preclinical activity in cell lines or xenograft models. However, the vast majority of these agents could not demonstrate any meaningful clinical activity despite well received preclinical rationale and results (Table 1).23–34

Table 1. Agents Without or Minor Clinical Activity in Patients with Refractory Germ Cell Cancer
AuthorAgentNo. patientsResponses
Williams et al. 198380Amsacrine60/6
Williams et al. 198523Mitoxantrone140/14
Drasga et al. 198781Iproplatin141/14
Harstrick et al. 199024Epirubicine161/16
Hoskins et al. 199026Mitomycin C72/7
Murphy et al. 199282Iproplatin150/14
Stoter et al. 199225Epirubicine180/18
Bokemeyer et al. 199383Vinorelbine70/7
Motzer et al. 199365Suramin140/14
Moasser et al. 199566All-trans Retinoic acid160/14
Puc et al. 199530Topotecan150/14
Kollmannsberger et al. 200028Bendamustine191/19
Kollmannsberger et al 200222Irinotecan150/15
Kondagunta et al 200434Temozolomide140/14

Paclitaxel in cisplatin-refractory disease

With different mechanisms of action and resistance to DNA-damaging agents, such as cisplatin and ifosfamide, paclitaxel demonstrated activity when given to the teratocarcinoma cell line 833k and its cisplatin-refractory variant 833k-63CP10.35 Motzer et al. were first in demonstrating a 26% response rate in patients who had failed cisplatin-based therapy.36 These results were subsequently confirmed by a study of the German Testicular Cancer Study Group (GTCSG), which revealed 6 responders among 24 (25%) patients.37 Despite intensive pretreatment with a median of 7 previous cisplatin-based cycles,3–12 peripheral neuropathy was tolerable with 29% World Health Organization grade III. To date, 5 studies that used single-agent paclitaxel in refractory germ cell cancer have been published and describe a response rate of 21% (range, 11–30%) in a total of 98 patients (Table 2). Interestingly, some of these patients who were pretreated with high-dose chemotherapy plus ASCT, and even individual patients with absolutely platinum-refractory disease, responded to paclitaxel.38 These results have led to investigations of paclitaxel in combination with cisplatin and/or ifosfamide as first-line salvage in patients at first relapse and as first-line therapy for patients with an intermediate prognosis.39

Table 2. Activity of Single-Agent Paclitaxel in Relapsed or Refractory Germ Cell Tumor
AuthorNo. of ptsPaclitaxel dose (mg/m2)Responses no. pts (total %)
Bokemeyer et al. 19948310135–225/3 hrs3 PR (30)
Motzer et al. 19943631250 + G-CSF/24 hrs3 CR, 5 PR (26)
Nasario et al. 19958415250 + G-CSF/24 hrs0 CR, 2 PR (13)
Bokemeyer et al. 19963724225/3 hrs2 CR, 4 PR (25)
Sandler et al. 19988518170–200/24 hrs2 PR (11)
Total98 5 CR, 16 PR (21)

Gemcitabine in relapsed disease

Gemcitabine is a nucleoside analog with an overall low toxicity profile and no cross-resistance to cisplatin. Two studies have evaluated the role of gemcitabine in refractory GCTs.40, 41 Investigators at Indiana University treated 21 patients with a schedule of 1250 mg/m2 gemcitabine once weekly for 3 weeks every 28 days. They reported 1 complete response (CR) and 2 partial responses (PR) in their patients for an overall response rate of 15%.41 The GTCSG included 31 patients in a protocol that examined gemcitabine at a dose of 1000 mg/m2 once weekly for 3 weeks every 28 days. Six (19%) patients achieved a response to gemcitabine including 3 of 22 (17%) patients after previous high-dose chemotherapy, 3 of 19 (16%) patients previously treated with paclitaxel, and 1 of 4 patients with mediastinal GCTs.40 The median progression-free interval was 4 months; however, individual patients responded for almost 1 year to gemcitabine. None of the patients with absolutely refractory disease responded, but 3 patients with refractory disease had a tumor-marker decline of more than 90%. Combining toxicity data of 52 published patients to date, thrombocytopenia was the main side effect (25%); nonhematologic toxicities were infrequent and mild.

Oxaliplatin in relapsed or cisplatin-refractory disease

The rationale of investigating oxaliplatin was based on the activity of oxaliplatin in cisplatin-resistant cell lines in vitro and functional studies that suggested mechanisms of DNA damage other than those of cisplatin or carboplatin.42 Subsequent in vitro studies indicated incomplete cross-resistance between cisplatin and oxaliplatin in nonseminomatous cisplatin-resistant germ-cell cancer cell lines.43 A French study group very early investigated oxaliplatin in combination with other agents in relapsed or refractory germ cell cancer. Oxaliplatin was either combined with cisplatin alone or with cisplatin and 1 or more of several other potentially active agents, such as ifosfamide, etoposide, bleomycin, and dactinomycin. None of the patients on oxaliplatin–cisplatin alone but 7 patients treated with oxaliplatin–cisplatin-based multidrug regimens responded. Because all responses were achieved with a combination of several potentially active drugs, the role of oxaliplatin alone could not be conclusively defined from this study. Further investigation of oxaliplatin in refractory germ cell cancer appeared warranted.44

The GTCSG subsequently performed a Phase II study in which oxaliplatin was administered as a single agent to 32 patients at 2 different dose levels of 60 mg/m2 once weekly for 3 weeks every 4 weeks and 130 mg/m2 every 2 weeks.45 Rationale for a higher single dose of oxaliplatin was based on in vitro data that suggested a relation between cell apoptosis and peak drug concentrations of oxaliplatin, both in cisplatin-resistant and cisplatin-sensitive colony-forming units of various tumor types.46 Compared with characteristics of patients treated within other trials that evaluated the role of paclitaxel or gemcitabine in relapsed germ cell cancer, this study was conducted in a prognostically more unfavorable group of patients. Thirty-four percent of these patients presented with a late relapse, 78% were pretreated with high-dose chemotherapy plus ASCT, and 85% of patients in this study had to be classified as cisplatin-refractory (Table 3). Overall, 4 (13%) patients achieved remission, with 3 of them occurring in the group of 19 patients treated with biweekly 130 mg/m2 of oxaliplatin (3 of 19, 19%). Considering poor prognostic characteristics, an overall response rate of 13% appears comparable to results reported for paclitaxel and gemcitabine, both of which are considered active in this therapeutic setting. Because all patients who responded to oxaliplatin were considered cisplatin-refractory, these data also seem to confirm preclinical data of incomplete cross-resistance between oxaliplatin and cisplatin in GCT.

Table 3. Comparison of Patient Characteristics in Trials of New Agents or Regimens for the Treatment of Refractory Germ Cell Cancer
Author, yrTreatmentNo pts.Late relapse (> 2 yrs), %Pts pretreated with HD-CT, %Cisplatin-refractory disease, %Response rate, % (95% CI)
  • a

    All patients “resistant” is defined as failure to achieve a durable complete remission to a cisplatin-based regimen and one or more of the following unfavourable prognostic features: relapse after complete remission or progression within 4 weeks after first-line therapy; poor or no response to prior conventional dose cisplatin/ifosfamide therapy; or extragonadal primary site.

  • b

    Refractory is defined as disease progression during or within 2 months after cisplatin-based chemotherapy.

Motzer 199436Paclitaxel31n.s.167626 (3–29)
Bokemeyer 199637Paclitaxel24n.s.507525 (10–47)
Bokemeyer 199940Gemcitabine3113715519 (13–45)
Einhorn 199941Gemcitabine2010556515 (3–38)
Kollmannsberger 200245Oxaliplatin3234788513 (1–24)
Hinton 200247Paclitaxel/Gemcitabine28n.s.363621 (12–49)
Miki 200252Irinotecan/Cisplatin18n.s.22n.s.a50 (27–73)
Kollmannsberger 200449Gemcitabine/Oxaliplatin3529896346 (30–64)
Pectasides 200450Gemcitabine/Oxaliplatin28101410032 (16–53)
Theodore 200451Paclitaxel/Oxaliplatin26n.s.n.s.61b30 (16–55)
Pectasides 200456Oxaliplatin/Irinotecan1860All pts resistanta40 (17–64)

Combination Chemotherapy for Patients with Multiply Relapsed or Cisplatin-Refractory GCT

Only with the availability of different new active agents in recent years, 2 or 3 drug combination regimens have become possible. Several studies are now available that have tested different combinations of paclitaxel, gemcitabine, oxaliplatin, and irinotecan. The main objectives of these studies have been to prove feasibility of combination chemotherapy in these intensively pretreated patients and to document response rate and survival. Achieving a high response rate in refractory patients is important because the induction of a remission may subsequently allow resection of residual masses and may, therefore, present a chance to achieve long-term survival in selected patients.6


Hinton et al. were among the first who demonstrated the feasibility of a combination regimen in relapsed or cisplatin-refractory germ cell cancer.47 Twenty-eight patients, with 75% of them having previously received 2 or more platinum-containing regimens and 10 patients being overtly platinum-refractory, were treated with a combination of paclitaxel and gemcitabine. A 21% response rate was observed, which appears not substantially different from the single-agent response rates reported for each of these drugs. However, remissions were observed in cisplatin-refractory patients and patients pretreated with high-dose chemotherapy plus ASCT. Although durable remissions have thus far been extremely rare with single agent therapy in these patients, among the 3 complete remissions observed, 2 were still ongoing at > 15 and > 25 months. Hematologic toxicity was the most prevalent toxicity, but, overall, this treatment was well tolerated. These results proved for the first time that combination chemotherapy is indeed not only feasible in heavily pretreated patients, but that it may also offer a chance of cure for some of these young patients.


In vitro studies suggest a supraadditive effect for the combination of oxaliplatin and gemcitabine, particularly for a sequence of gemcitabine followed by oxaliplatin.48 In view of these preclinical data and the clinical single-agent activity of both drugs, 2 Phase II studies have examined activity and toxicity of a combination comprising oxaliplatin and gemcitabine in refractory germ cell cancer patients.49, 50 Within a Phase II study by the GTCSG, 35 patients were treated with a combination of gemcitabine 1000 mg/m2 given on Days 1 and 8 and oxaliplatin 130 mg/m2 given on Day 1 of a 3-week cycle. Patients were heavily pretreated with a median number of 6 previous cisplatin-based standard chemotherapy cycles, and 89% of them had also received previous high-dose chemotherapy with ASCT. Half of the patients had also already received paclitaxel as part of their salvage regimens, and 63% of all patients were classified as cisplatin-refractory or absolutely refractory, which reflects the poor prognostic features of this patient population. Despite these unfavorable characteristics, a response rate of 46% and a median overall survival time of 13 months for responding patients were achieved. Two patients had a CR to chemotherapy, and another responding patient was rendered disease free with additional surgery, with all 3 of them achieving a continuously disease-free status at the time of study publication. This response rate is considerably higher than the response rates of 15–25% reported for single-agent therapy with 1 of the 4 active agents. In this study, approximately 50% of remissions—including 1 complete remission—occurred in cisplatin-refractory patients. These results were subsequently confirmed by a Greek study that investigated the same regimen in 29 cisplatin-refractory patients, who had comparably unfavorable prognostic features. Most of these 29 patients had been pretreated with 2 or 3 cisplatin-based regimens, and all patients were considered absolutely cisplatin-refractory or cisplatin-refractory. A response rate of 32%, including 4 complete remissions, was reported with all 4 patients continuously disease free at > 14 to > 28 months. All patients in these 2 studies who are continuously disease free in complete remission may have a reasonable chance of cure, which is rarely achieved with single-agent therapy alone. Grade 3/4 toxicities were comparable in both studies and consisted mainly of neutropenia and thrombocytopenia in approximately 50% of patients, with 10% of patients developing neutropenic fever.

These results represent the best results achieved thus far in heavily pretreated or cisplatin-refractory germ cell cancer patients.


The French Study Group combined paclitaxel at a dose of 175 mg/m2 with oxaliplatin at a dose of 130 mg/m2 administered every 3 weeks in 26 intensively pretreated patients.51 Sixteen patients were classified as refractory and 10 as relapsed, with refractory defined as disease progression during or within 2 months after cisplatin-based chemotherapy, and relapsed disease was defined as progression between 2 and 6 months after the last cisplatin-based chemotherapy. One patient achieved a PR with tumor-marker normalization, and 3 further patients had a > 90% marker decrease. With 2 patients continuously disease free at > 13 and > 23 months after additional surgery and a median survival of 8.8 months, these results demonstrate the activity of oxaliplatin–paclitaxel in intensively pretreated patients. Severe neuropathy, a major concern when combining 2 potentially neurotoxic drugs, occurred in only 1 patient.

Irinotecan–cisplatin and irinotecan–oxaliplatin

Despite negative results reported for single-agent irinotecan in a Phase II study by the GTCSG, Miki et al. conducted a Phase II study with a combination of irinotecan with cisplatin or nedaplatin, a platin analog.52 This was based on a synergistic interaction that had been reported for the combination of irinotecan and cisplatin or nedaplatin in vitro.53 A very high response rate of 50% was reported for this combination in 20 cisplatin-refractory germ cell cancer patients, which suggested that the synergism of irinotecan with cisplatin is clinically meaningful.52 However, no exact details of patient characteristics were given in the study report. Most patients appeared to have had 2 prior lines of standard cisplatin-based therapy, but only 4 patients had received high-dose chemotherapy before irinotecan–cisplatin, and 2 patients had been pretreated with BEP alone. The extent of the previous therapy, such as the number of previously received cisplatin-containing cycles and the number of patients with prior high-dose chemotherapy who presented as late relapse or truly cisplatin-refractory, certainly has an impact on results of studies on relapsed or refractory GCT (Table 3). These patients had different prognostic features and seemed to have had considerably less prior therapy than patients included in other trials. Most importantly, refractory disease was defined as GCT that was only “resistant” to first-line or second-line cisplatin-based chemotherapy, and no exact definition for resistant disease was specified. This makes interpretation of these results difficult, and the favorable results may be partly because of patient selection and some degree of residual cisplatin sensitivity. Irinotecan–cisplatin was also toxic with all patients developing Grade 3/4 neutropenia and 94% Grade 3/4 thrombocytopenia, which limits its use in intensively pretreated patients.

A synergistic activity was also demonstrated for the combination of oxaliplatin and SN-38, the active metabolite of irinotecan, in colon cancer cell lines.54 This was attributed to a modification of tumor resistance to platinum.55 A Phase II study subsequently investigated the combination of oxaliplatin 85 mg/m2 given on Days 1 and 15 and irinotecan 80 mg/m2 on Days 1, 8, and 15 of a 4 week cycle.56 All 18 enrolled patients had failed to achieve a durable response to a cisplatin-based regimen. Seventy-eight percent of patients had received paclitaxel as part of their previous salvage therapies. Because of their extensive pretreatment, all patients received G-CSF prophylactically. Seven patients responded, which resulted in a response rate of 40% (95% confidence interval [95% CI], 17.3–64.2%). Four patients achieved a CR with 3 of them continuously disease free at > 11, > 14, and > 19 months, which offered them a chance of cure. Overall, the combination of oxaliplatin–irinotecan was well tolerated; in particular, the rate of neurotoxicity was surprisingly low, although two-thirds of patients had been pretreated with cisplatin and paclitaxel. These results appear comparable to those obtained with gemcitabine–oxaliplatin, although patients included in the gemcitabine–oxaliplatin trials were even more intensively pretreated than patients included in this trial.49, 50 Despite poor results reported for single-agent topoisomerase I inhibitor therapy in refractory GCT patients, the results of this study also support the hypothesis of a clinically meaningful synergistic interaction between irinotecan and platinum analogs.

Salvage Surgery

Several studies have investigated the role of surgery for patients with persistently elevated tumor markers after cisplatin-based chemotherapy or for patients who relapse after salvage chemotherapy (sometimes referred to as “desperation surgery”).57–60 Murphy et al. reported 48 chemotherapy-refractory patients with elevated tumor markers or other signs of tumor progression. Seventy-nine percent of patients were rendered grossly disease free and 60% obtained a serologic remission. Of these patients, 21% remained continuously disease free without any further treatment.57 These results were confirmed by Albers et al., who reported a high long-term disease-free rate after salvage surgery for patients with persistently elevated tumor markers.60 Smaller studies have also described long-term disease-free survival (DFS) rates of up to 25% with salvage surgery.58, 59 Complete resection of residual masses is the most crucial factor for success in this setting.

Complete resection of residual masses also appears to be an important step toward long term DFS in cisplatin-refractory patients who achieve a remission on salvage chemotherapy as demonstrated by a retrospective analysis from Indiana University.6 In this analysis, only 5% of 101 patients with relapsed and/or cisplatin-refractory metastatic disease achieved long-term survival, and all of these had had surgery for metastatic disease as an important component of their salvage treatment. Similarly, most long-term survivors, reported in recent trials that have investigated novel combination regimens, had surgery as part of their salvage treatment for refractory disease, which emphasizes the importance of surgical resection of residual masses in these patients. Surgery alone or surgery after salvage chemotherapy should, therefore, always be considered for selected patients with cisplatin-refractory disease, particularly in patients who present with localized and potentially completely resectable recurrences or with completely resectable masses after salvage chemotherapy.

Patients with late relapse represent a particular treatment problem due to tumor transformation and the possible presence of nongerm cell cancer components or mature teratoma. George et al. recently reported 83 patients among whom 23 (28%) relapsed between 2–5 years, and 60 (72%) relapsed 5 years after cisplatin-based chemotherapy.61 Overall, 47% of patients remained disease free after salvage surgery alone or chemotherapy most often followed by surgery. Only 5 patients achieved CR after chemotherapy alone, which emphasizes the importance of surgery in this setting. All larger series published to date have consistently shown that even if the disease is still chemotherapy sensitive, surgery is the most important treatment for these patients.61–63 After complete surgical resection, long-term DFS rates of 25–40% can be achieved. Surgical resection should, therefore, be attempted in all patients with late relapse whenever possible.

Biologic and Molecular Targeted Therapy for Refractory Germ Cell Cancer

Molecular mechanisms involved in testicular cancer development and in cisplatin resistance are still incompletely understood. Currently under investigation is whether there are molecular differences that separate sensitive disease from resistant disease. New techniques, such as comparative genomic hybridization, may give more information about genetic changes in patients who relapse after cisplatin-based therapy or in patients who have an incomplete response to therapy. Some trials have investigated biologic agents for use in treatment of relapsed GCTs. On the basis of in vitro results in 2 human teratocarcinoma cell lines, suramin was evaluated in a Phase II study in 14 intensively pretreated patients. Serum alpha-fetoprotein levels declined in 1 patient, but no CR or PR was observed.64 ATRA (all-trans–retinoic acid) induced differentiation as a response in 6 human germ cell tumor cell lines. Subsequently, 16 patients were treated with retinoic acid for an average duration of 7 weeks, but no responses were noted.65

Novel molecular targets are now also being explored in germ cell cancer. Initial studies suggest that vascular endothelial growth factor (VEGF) may play an important role in development and metastasis of testicular cancer, which makes the evaluation of VEGF inhibitors noteworthy.66, 67 c-KIT is expressed in seminomas and tyrosinekinase inhibitors, so targeting c-KIT may be a valuable therapeutic option for treating refractory seminomas.68, 69 However, detectable activating KIT mutations appear to be rare in nonseminomas, so it is unlikely that they would respond to targeted therapy with imatinib.70, 71 Epidermal growth factor receptor (EGFR) expression in relapsed or refractory germ cell cancer was investigated by the GTCSG as well as by an Italian study group. Both studies report expression of EGFR restricted to trophoblastic giant cells and syncytiotrophoblastic components of nonseminomatous germ cell tumors with no EGFR expression in other cell types.68, 72 These results by Moroni et al. encouraged the Italian group to initiate a Phase II study testing erlotinib, a small-molecule EGFR inhibitor, in patients with GCTs. Mandani et al. reported a significant proportion of patients expressing EGFR in a study population comprising late-relapse patients who had an exclusively yolk sac histology or myxoid differentiation or transformed teratomas with a sarcomatous differentiation. HER-2/NEU also can be detected by immunohistochemistry in some germ cell tumors, but only very few patients exhibit a HER-2/NEU gene amplification by fluorescence in situ hybridization (FISH), indicating both that a resistant phenotype in GCTs may not be related to alterations in the ERBB2 signaling pathway and HER-2/NEU.73 Only recently, the protein product of the T-cell leukemia–lymphoma 1 (TLC1) oncogene was identified as a novel AKT kinase activator. TLC1 is expressed in a high proportion of testicular seminomas of germ cell origin indicating its potential to serve as a novel anticancer drug target.76 A variety of other potential novel targets are currently under investigation, and it is hoped that, in the future, insight into their molecular mechanisms may allow designing new treatment approaches for these patients.

Summary and Future Prospects

The 10-year survival rate of patients with metastatic testicular germ cell tumor who were treated with cisplatin-based chemotherapy continued to significantly increase during the last 20 years. This improvement has resulted mainly from an increase in survival of poor-prognosis patients with metastatic disease.75 Progress has also been made in patients with refractory germ cell cancer, but cisplatin-refractory germ cell cancer still carries a very unfavorable prognosis. The exploration of new drugs and regimens in testis cancer has mostly been performed in these intensively pretreated patients without an established curative treatment option. Several new active agents have been identified thus allowing development of combination chemotherapy with 2-drug combinations. Response rates of > 30% have been achieved for the first time in cisplatin-refractory patients.49, 50, 76 The most favorable results have thus far been obtained with the combination gemcitabine–oxaliplatin in a prognostically very unfavorable patient population (Table 3).49, 50 All studies investigating combination chemotherapy regimens have reported patients with CR and potentially long-term DFS, which may, again for the first time, offer a chance of cure for these patients. Combination chemotherapy and, if possible, resection of residual masses should, therefore, be considered now as useful therapeutic options for patients in good performance status with a sufficient bone marrow reserve.

Three-drug regimens, such as gemcitabine–oxaliplatin–paclitaxel or paclitaxel–gemcitabine–cisplatin are now being developed in clinical trials. Despite noteworthy preliminary response rates, these 3-drug combinations are still associated with frequent and substantial toxicity in these heavily pretreated patients resulting in a high number of dosage and/or schedule modifications and early treatment termination.77, 78 These studies may, however, lead to development of a 3-drug combination regimen that is entirely noncross-resistant to standard cisplatin-based regimens currently used for germ cell cancer. The identification of new active regimens may not only allow improved treatment of refractory patients but also may offer new options for combination chemotherapy regimens for patients at an earlier phase of their disease.

The identification of new therapeutic options for patients with cisplatin-refractory disease is, however, hindered by the small number of patients and extent and intensity of chemotherapy these patients have previously received. All investigations of novel regimens will face the challenge of defining the appropriate clinical setting for second- or even first-line therapy in GCT patients. Clinical situation and inclusion criteria will have to be very carefully chosen, but they may take us another step forward in treating germ cell cancer. Cooperative efforts that use mutual, experimental, and palliative treatment protocols for new agents and regimens are necessary, and ongoing research in this area may not only improve treatment of patients with metastatic germ cell cancer but also may yield important discoveries of mechanisms of drug resistance and possibilities of combining novel agents applicable to patients with other types of cancer.