Abstract
Germ cell cancers (GCCs) represent the most common malignant tumours among men up to the age of 40 years. GCCs are characterised by high cure rates of >99% for localised stage I disease, but even at advanced metastatic stages, 67%–96% of patients can be cured with multimodal treatment approaches. The high curability owes to an exceptional responsiveness to cisplatin-based chemotherapy. However, about 30% of patients with metastatic disease eventually relapse after first-line multimodality treatment. At first relapse, salvage treatment still cures approximately 50% of patients. Platinum-based chemotherapy remains the standard of care in this setting, administered either as conventional-dose cisplatin-based regimens or as high-dose carboplatin-based combination chemotherapy. Across multiple large retrospective cohorts, high-dose chemotherapy has repeatedly been associated with superior progression-free and overall survival compared with conventional-dose approaches, although definitive prospective evidence is awaited. Patients suffering second or multiple relapses are considered platinum-resistant and face a very dismal prognosis with a life expectancy generally limited to less than 12 months. The mainstay of treatment in this setting is still conventional cytotoxic chemotherapy since no molecularly targeted treatments have shown clinically meaningful activity, to date. However, Claudin-6 (CLDN6)-directed approaches appear the most promising avenue of new treatment options in this situation. In this narrative review, we outline established and emerging treatment opportunities for this rare but challenging clinical scenario of platinum-resistant GCCs. This review provides a focused and up-to-date synthesis of therapeutic options for multiply relapsed, platinum-refractory GCCs, with a particular emphasis on late-line treatment sequencing and emerging biomarker-driven strategies, including CLDN6-directed therapies.
Plain language summary
Germ cell cancers (GCCs) are highly curable tumours, even when they have spread. Most patients respond very well to standard cisplatin-based chemotherapy. However, a small group of patients experience several relapses and eventually develop resistance to platinum-based treatment. In this situation, treatment options are limited and long-term survival is rare. This review summarises the available therapies for patients with platinum-refractory GCCs. High-dose chemotherapy can still help some patients, but its success decreases after multiple relapses. Standard combination chemotherapies may temporarily shrink the tumour, but responses usually do not last. Modern treatments such as targeted drugs or immunotherapy have shown very limited benefit to date. A promising new approach targets a protein called Claudin-6 (CLDN6), using innovative treatments such as antibody–drug conjugates or CAR-T cells. These early results offer hope for future therapeutic progress.
Keywords
Background
GCCs are relatively rare, representing about 1% of all cancers in men; however, they are the most common cancer among young men aged 15–40. 1 The incidence rate varies globally, with higher rates in Western countries and lower rates in Asia and Africa. While the overall incidence is increasing, survival rates are high even in metastatic stages with 5-year survival rates from 67% to 96% according to the International Germ Cell Cancer Collaborative Group-Update Consortium for patients with advanced disease due to excellent response to platinum-based chemotherapy. 2
Despite the generally outstanding chemotherapy sensitivity, 20%–30% of patients relapse after or progress during cisplatin-based first-line treatment and require salvage treatment with either conventional-dose chemotherapy (CDCT) or high-dose chemotherapy (HDCT) plus autologous stem cell transplant (ASCT). 3 Across the literature, the terms platinum-refractory and platinum-resistant are not used uniformly, which can limit cross-trial comparability – particularly when translational cohorts are included. In this review, we use platinum-refractory disease to denote progression during cisplatin-based chemotherapy or relapse within a short interval after completion of a cisplatin-containing regimen. By contrast, platinum-resistant disease refers to relapse after an initial response to cisplatin-based therapy, typically occurring beyond this early interval. We explicitly acknowledge that individual clinical trials and translational studies apply heterogeneous time cut-offs and response requirements, which needs to be considered when interpreting reported outcomes. 3
Salvage chemotherapy in combination with post-chemotherapy residual tumour resection can achieve longer lasting remissions and eventually cure in approximately 50% of relapsed patients. 4 Subsequent relapses are associated with substantially worse outcomes. Overall, 3%–5% of all GCCs patients develop platinum-refractory disease and potentially die from GCC. 5 Unfortunately, treatment options after failure of two or more lines of platinum-based combination treatments are limited and patient survival is generally limited to a few months only. 6 Despite the widespread use of targeted and molecular therapies in other cancers, their development in GCCs has shown little success to date. In this review, we aim to explore current and emerging therapeutic strategies for multiply relapsed, platinum-refractory GCC patients.
HCDT for multiply relapsed GCC patients
High-dose salvage chemotherapy with two or three cycles of a doublet regimen consisting of high-dose carboplatin and etoposide followed by ASCT is typically considered as first salvage treatment. Retrospective registry data suggest that HDCT may still achieve cure in multiply relapsed GCC patients if not previously administered. However, durable responses are less frequent when HDCT is used as a third or later line of treatment compared to its use as first or second salvage therapy following failure of a CDCT. According to a retrospective analysis from Indiana University, the 2-year progression-free survival (PFS) rate among 61 patients treated with HDCT as third or subsequent line of therapy was still 50%. 7 Seidel et al. conducted a retrospective observational cohort study including 283 patients treated at 11 international institutions. All patients had relapsed or refractory GCC of which 124 patients had undergone salvage HDCT after at least one salvage treatment line. The median follow-up for these patients was 17.0 months, and among these patients, the objective response rate (ORR) was 60% and the 2-year and 5-year overall survival (OS) rates were 53% and 37%, respectively. Importantly, HDCT as third or further line treatment was associated with a higher incidence of grade ⩾3 non-haematologic toxicities than when used as first salvage treatment. 8
In the retrospective analysis by Lorch et al., 9 HDCT as single or sequential therapy yielded responses in 55% of patients but resulted in a disappointing overall prognosis, with a projected 5-year OS of only 17%.
While HDCT remains an important curative option, it is associated with substantial acute and long-term toxicity. Acute complications include severe myelosuppression with febrile neutropenia and invasive infections, mucositis, gastrointestinal toxicity and the need for intensive supportive care, including transfusions and prolonged hospitalisation. Treatment-related mortality, although reduced in experienced centres, remains a relevant risk in heavily pretreated patients. In addition, cumulative non-haematologic toxicities may occur after multiple prior lines, including nephrotoxicity, ototoxicity, neurotoxicity, pulmonary toxicity and impaired marrow reserve. Long-term survivorship issues – particularly relevant in this young population – include infertility, endocrine dysfunction and the potential risk of secondary malignancies after extensive cytotoxic exposure. These risks underscore the importance of careful patient selection, treatment in expert centres and early integration of survivorship counselling and fertility preservation.7,10 In conclusion, HDCT in combination with post-chemotherapy resection of residual tumours can still be pursued with curative intent even when used as second or subsequent salvage treatment. However, reported long-term survival rates in this setting vary considerably across the literature and are generally markedly lower than those observed when HDCT is administered as first salvage therapy.
Durable disease control in these late-line settings is most consistently reported in patients who undergo complete post-chemotherapy resection of all residual disease whenever feasible. However, study-level reporting stratified by resection status and detailed relapse histology is inconsistent across cohorts, limiting more granular comparisons.
Palliative chemotherapy treatments
Due to their young age and general lack of relevant co-morbidities, refractory GCC patients are often capable of tolerating third and further line combination chemotherapy, even after HDCT with autologous stem cell transplantation. 6 Table 1 highlights available evidence for active single agent and combination chemotherapy regimens in this setting.
Chemotherapy regimens for palliative setting.
CI, confidence interval; DOR, duration of response; EMA CO, etoposide/methotrexate/actinomycin D alternating with cyclophosphamide/vincristin; GEMOX, gemcitabine/oxaliplatin; GOP, gemcitabine/oxaliplatin/paclitaxel; HDCT, high-dose chemotherapy; n.r./NR, not reported; ORR, objective response rate; OXIRI, oxaliplatin/irinotecan; PFS, progression-free survival; TPG, paclitaxel/cisplatin/gemcitabine; TTF, time to treatment failure.
Monochemotherapy
As outlined above, cytotoxic chemotherapy remains the backbone of treatment for refractory GCCs. Several single-arm phase II studies have historically evaluated single-agent cytotoxic drugs in the platinum-resistant setting – including oral etoposide, paclitaxel, gemcitabine, oxaliplatin and ifosfamide – typically reporting ORRs of only 10%–20% and durable remissions in merely 3%–10% of patients. 5
More recently, cabazitaxel has been explored as a modern late-line option in small phase II and compassionate-use cohorts, typically after failure of multi-agent regimens such as GOP. 30 While cabazitaxel shows activity in individual patients, responses remain infrequent and rarely durable, underscoring the biological refractoriness of this disease stage.
Given the overall modest activity and lack of sustained disease control, single-agent chemotherapy is generally reserved for patients who are not fit for combination regimens owing to poor performance status, comorbidities or cumulative toxicities affecting marrow or renal reserve. 31
Polychemotherapy
Doublet regimens
Based on active single agents, a variety of combination chemotherapy regimens have been established and adopted in the management of platinum-refractory GCC patients.
Gemcitabine and oxaliplatin
The doublet combination of gemcitabine and oxaliplatin (GEMOX) is a well-established treatment regimen in the palliative setting, and it was assessed in two-phase II trials.22,23 In the first study, GEMOX was administered to 35 GCC patients, of which approximately half had previously undergone HDCT. Sixteen pts (46%) achieved a response, with 3 (9%) achieving complete remission, 2 following chemotherapies alone and 1 after post-chemotherapy surgical resection. 22
Oxaliplatin and irinotecan
Two trials evaluated the doublet combination of oxaliplatin and irinotecan (OXIRI), administered alone or in combination with paclitaxel, showing that the addition of paclitaxel resulted in a higher ORR (70% vs 40%). Nevertheless, these studies excluded patients who had previously undergone HDCT, which may overestimate activity of these combinations in patients who had prior salvage HDCT.25,32
Triplet regimens
The highest response rates so far have been achieved with two triplet combination chemotherapy regimens demonstrating ORRs of more than 50%, in patients who did not receive paclitaxel before: gemcitabine, oxaliplatin and paclitaxel (GOP) and paclitaxel, gemcitabine and cisplatin (TPG).26,28 Importantly, long-term disease control can generally only be achieved in subjects who undergo post-chemotherapy surgical removal of all residual masses. 33 Of note, the available evidence rarely provides uniform, granular outcomes stratified by completeness of resection (complete vs incomplete vs no surgery) or by detailed viable relapse histology (e.g. necrosis/teratoma/viable non-teratoma, somatic malignant transformation), which represents a key limitation of the current literature.
Gemcitabine, oxaliplatin and paclitaxel
Two phase II trials and a retrospective cohort study confirmed the efficacy of the triplet combination of GOP.26,27,34 Firstly, a phase II German Testicular Cancer Study Group (GTCSG) trial enrolled 41 patients with treatment-refractory disease, where approximately 80% of them had failed previous HDCT. These patients received a 3-weekly GOP regimen yielding an ORR of 51%. Two patients achieved a complete response (CR) with chemotherapy alone, while six others had no evidence of disease (NED) after post-chemotherapy surgical resection. 26 Another phase II trial reported similar treatment efficacy using a bi-weekly application schedule. Specifically, an ORR of 31% was observed with 17% of patients reaching CR after surgery and 7% a CR following GOP alone. 27 Finally, the GTCSG conducted a retrospective cohort study with 63 patients treated with the 3-weekly GOP regimen, demonstrating an ORR 44% and median OS and PFS of patients who achieved ORR were 22.6 months (95% CI: 0.00–48.47) and 10.2 months (95% CI: 4.31–16.01), respectively. 34
Gemcitabine, cisplatin and paclitaxel
Another triplet regimen, TPG (gemcitabine, cisplatin and paclitaxel), demonstrated a nearly 50% ORR in a cohort of 75 patients who had progressed after two or three cisplatin-based therapy schedules, comprising HDCT. Notably, more than half of these patients were cisplatin-refractory. A partial response (PR) with normalised tumour markers (PRm-) was observed in 29 subjects (38.7%), while eight patients (10.7%) obtained a CR, with median OS and PFS being 13 and 5 months. 28
Methotrexate/etoposide/actinomycin D administered in sequence with cyclophosphamide and vincristine
Concerning patients with hCG positive germ cell cancers (GCCs), in particular those with elements of chorion carcinoma, the methotrexate/etoposide/actinomycin D administered in sequence with cyclophosphamide and vincristine (EMA/CO) regimen displays another treatment option. Developed in the 1980s, it was designed for the treatment of high-risk gestational trophoblastic neoplasia, particularly in patients resistant to single-agent methotrexate or actinomycin D. The rationale was to combine agents with different mechanisms of action and non-overlapping toxicities in a dose-dense, alternating schedule to maximise efficacy while maintaining tolerability. The regimen emerged after earlier experiences with the EMA triplet (etoposide, methotrexate, actinomycin D) demonstrated high activity. Modifications such as EMA/EP (etoposide and cisplatin replacing CO) have been developed for resistant cases or as salvage options. In a retrospective registry with 41 treatment-resistant hCG-positive patients, who received this combination regimen, now established as an option for this setting, just under one-third of patients reached objective responses and 12% attaining a CR. OS and PFS were 8 and 3 months, respectively. 29 This limited activity may be due to heavy pre-treatment in this patient population, and further real-world data ares needed to better understand the role of EMA/CO and EMA/EP for this patient population.
Small molecule inhibitors
Tyrosine kinase inhibitors
Receptor tyrosine kinases (RTKs) have been shown to be activated in pre-treated GCC patients. The RTKs most commonly involved include platelet-derived growth factor, vascular endothelial growth factor receptor, hepatocyte growth factor receptor, stem cell growth factor receptor and epidermal growth factor receptor. However, although extensively evaluated in multiple clinical trials, RTK-targeted therapies have yet failed to produce clinically meaningful responses in unselected platinum-resistant GCC patient cohorts. 31
Importantly, most of these studies were conducted without prior comprehensive molecular profiling and molecular patient selection, and therefore largely tested tyrosine kinase inhibitors (TKIs) in unselected GCC patients, which may in fact have had detrimental effect on TKI activity in these trial populations.
Sunitinib is a multikinase inhibitor administered in 32 refractory germ cell tumour (GCT) patients enrolled in a phase II trial, showing a 13% of total response rate (only PR) with a median OS and median PFS of 3.8 months (95% CI, 3–6.6) and 2 months (95% CI, 1.4–2.6), respectively. 35 Another phase II trial included 43 patients treated with pazopanib in monotherapy, another multi-kinase inhibitor, reported a tumour marker response in approximately 70% of cases. Transient radiographic disease stabilisation was achieved in 44%, but only 4.7% achieved a PR. After a median follow-up of 29.6 months (range = 10.6–35.8 months), the 24-month OS rate was 14.2% (95% CI, 6.0%–33.7%) and 3-month PFS rate was 12.8% (95% CI, 5.7%–28.9%). 36 Sorafenib was evaluated in a phase II trial that enrolled 18 patients with refractory GCTs, reaching disease stabilisation lasting more than 1 year in 16.7% of cases (3 patients) and a reduction serum tumour marker in 44.4% of subjects (8 patients). 37
Cabozantinib is a multikinase inhibitor targeting mainly mesenchymal-epithelial transition (MET) factor. It is currently being administered as monotherapy in an ongoing phase II trial involving refractory GCT patients, who relapsed after one salvage treatment. Notably, this trial is also investigating potential biomarkers, including next-generation sequencing and c-MET expression assessed by immunohistochemistry (IHC), with the aim of identifying predictors of response and resistance to therapy (NCT04876456). 38 Between 10% and 40% of GCTs harbour c-KIT mutations, another RTK often overactivated in seminomas, but less frequently in non-seminomas. These mutations predominantly involve exon 17, which is commonly associated with resistance to c-KIT-targeted therapy. This might explain the limited clinical efficacy observed in a phase II trial evaluating imatinib, a predominant c-KIT inhibitor, in 18 refractory GCC patients, where no patient achieved an objective radiographic response and only one patient had a 3-months disease stabilisation. 39 Across drug classes, the lack of systematic molecular profiling and biomarker enrichment is a recurring limitation in the late-line GCC trial landscape and should be addressed prospectively.
mTOR inhibitors
An immunohistochemical study revealed a strong activation of mTOR pathway in seminomas, while loss of PTEN gene was also frequently observed in GCTs, prompting the evaluation of Everolimus, an mTOR inhibitor, in this setting. Two phase II trials were conducted, enrolling 15 and 25 patients. However, neither trial demonstrated any objective response, and the median OS and median PFS were less than 4 months and less than 2 months, respectively, in both studies.40,41
Poly(ADP-ribose)polymerase inhibitors
One of the most important types of DNA damage caused by cisplatin is DNA single-strand breaks (SSB) and resulting replication-associated DNA double-strand breaks (DSB). Poly(ADP-ribose)polymerase (PARP) enzymes are key players of cellular SSB repair. PARP inhibitors (PARPi) disrupt SSB repair which causes replication-associated DSBs, which need to be repaired by homologous recombination (HR). In case of HR-deficiency, PARPi-induced DSB remain unrepaired which can lead to cell death. 42 An in vitro study highlighted that the concomitant administration of PARPi and low-dose cisplatin in patients with cisplatin-refractory disease could be effective building on a PARPi-induced platinum-sensitisation in a synergistic manner. 43 According to what was mentioned above, two separate trials investigated the efficacy of PARPi in heavily pretreated GCTs patients.
Olaparib was assessed in a phase II trial involving patients who had received at least two previous treatments. Among the 18 subjects enrolled, 5 patients achieved stable disease (SD; 27.8%), while no PRs were observed in the remaining participants. The 12-month OS was 27.8% (95% CI, 10.1%–48.9%), while 12-week PFS was also 27.8% (95% CI, 10.1%–48.9%). Germline DNA was analysed using target sequencing, and only one patient was identified as BRCA1 mutation carrier; this individual maintained SD over a period of 4 months. 44 Another phase II study evaluated the PARPi veliparib in combination with carboplatin and gemcitabine. Fifteen patients were enrolled; SD or PR were documented in 5 (33.3%) and 4 (26.7%) cases, respectively. The primary aim of this study was 12-months PFS, but it was achieved only by one patient (6.7%). The median OS was 10.5 months (95% CI, 8.9–11.1), while the median PFS was 3.1 months (95% CI, 2.2–3.9). 45 In conclusion, so far, PARPi fail to demonstrate meaningful therapeutic efficacy in a GCC setting. Possible reasons include cross-resistance between PARPi and platinum-based agents, inter- and intra-patient heterogeneity observed in refractory GCCs and the absence of predictive genetic alterations and biomarkers associated with PARPi responsiveness. 46 The prevalence of somatic and germline homologous recombination repair (HRR) defects in GCTs is low. 44 Notably, in cohort of 187 GCT patients, approximately 25 individuals exhibited at least one germline mutation in a DNA damage repair gene despite no reported familial predisposition to cancer, and around 5 of these cases involved BRCA1/2 mutations. 47
Epigenetic targeting
Aberrant DNA methylation of CpG islands within gene promotors silences gene expression and is mediated by DNA methyltransferases (DNMTs), which are frequently overexpressed in non-seminomatous GCCs. 48 Consequently, non-seminomas display a globally hypermethylated genome, whereas seminomas are generally hypomethylated. 49 Promoter hypermethylation of RASSF1, HIC1, MGMT and CALCA has been linked to non-seminomatous histology, and methylation of RASSF1, HIC1 and CALCA correlated with reduced cisplatin sensitivity.50,51 In addition, a subset of non-seminomas shows hypermethylation of HRR genes (BRCA1, RAD51C), which was associated with PARPi sensitivity in preclinical models. 52
Demethylating agents such as 5-azacytidine or 5-aza-deoxycytidine can reverse aberrant methylation and restore TP53-mediated apoptosis, particularly in DNMT3B-overexpressing embryonal carcinoma models. 53 These agents also sensitise tumour cells to cisplatin.48,54,55 In a phase I trial, guadecitabine combined with cisplatin achieved a 23% response rate in heavily pretreated, platinum-refractory GCCs, with manageable haematologic toxicity. 56 Ongoing trials are expected to define the therapeutic relevance of guadecitabine-based epigenetic modulation and to establish whether hypomethylating agents can be leveraged to achieve durable re-sensitisation to cisplatin in patients with refractory GCCs.
Immunotherapy
Programmed cell death protein 1 (PD-1) is an inhibitory receptor expressed on multiple immune cells, including T cells, NK cells (Natural Killer cells) and B cells. Its interaction with PD-L1 on tumour cells suppresses cytotoxic T-cell responses and promotes immune evasion. Immune checkpoint inhibitors (ICIs) targeting PD-1 or PD-L1 restore T-cell activity and antitumour immunity. 57
The rationale for ICIs in GCCs stems from their immune-rich tumour microenvironment, characterised by macrophages, lymphocytes, NK cells and dendritic cells.58,59 PD-L1 expressions are common, particularly in embryonal carcinoma and seminoma, but varies between studies. High PD-L1 expression in tumour cells has been associated with poor outcomes, whereas PD-L1 expression in tumour-infiltrating lymphocytes may correlate with improved survival.60,61
Despite frequent immune-cell infiltration and variable PD-L1 expression, most GCC exhibit a comparatively low tumour mutational burden and consequently a low neoantigen load. This biological feature likely constrains the generation of strong, mutation-driven tumour-specific T-cell responses and may contribute to the consistently low response rates observed with PD-1/PD-L1 blockade. Taken together, an immune-infiltrated microenvironment alone appears insufficient to predict benefit in refractory GCCs in the absence of higher neoantigenicity.
Early-phase clinical trials of ICIs in refractory GCCs have shown limited or no efficacy. In the phase II GU14-206 trial, pembrolizumab induced no objective responses in 12 non-seminoma patients. 62 Small case series reported isolated durable remissions with PD-1 blockade, mostly in PD-L1-positive tumours. 63 In a Japanese phase II study, nivolumab produced 1 PR and 3 SDs among 17 patients, with benefit unrelated to PD-L1 expression but linked to high tumour mutational burden. 64 Another pembrolizumab trial achieved only SD in 3 of 12 cases. 65
Anti-PD-L1 therapy with avelumab failed to produce responses in eight patients. 66 In the APACHE study, durvalumab ± tremelimumab led to 1 PR and 1 SD among 22 patients (both patients were in combination arm), but hyperprogression occurred in up to 72% of cases. 67
Overall, checkpoint inhibitors have not demonstrated clinical benefit in refractory GCCs. Responses, when observed, are rare and short-lived, and no validated predictive biomarkers currently support routine ICI use in this setting (Table 2).
Immunotherapy trials with checkpoint-inhibitors in platinum-refractory GCTs.
Avel, Avelumab; CR, complete response; Durv, Durvalumab; mDOR, median duration of response; mGCT, metastatic germ cell tumour; mo, months; mOS, median overall survival; mPFS, median progression free survival; Nivo, Nivolumab; ORs, objective responses; PD, programmed cell death protein; Pemb, Pembrolizumab; PR, partial response; pts, patients; SD, stable disease; TGCT, testicular germ cell tumour; Treme, tremelimumab.
Emerging targeted therapies for platinum-resistant testicular GCTs: Antibody–drug conjugates and chimeric antigen receptor-T
Targeting CD30
CD30 is a transmembrane glycoprotein of the TNF receptor family involved in tumour cell growth regulation. While uniformly expressed in Hodgkin lymphoma and anaplastic large cell lymphoma, CD30 is also variably detected in GCCs, particularly in embryonal carcinoma (≈93%–99%) and in about 20% of seminomas. 68 CD30 expression in post-chemotherapy residual viable GCC tissue has been linked to inferior survival. 69
Brentuximab vedotin, an antibody–drug conjugate (ADC) targeting CD30, has shown strong efficacy in lymphomas. In a phase II basket trial including five GCC patients, two achieved objective responses (one CR, one PR). 70 However, two subsequent phase II trials with 24 and 18 chemo-refractory GCC patients, failed to demonstrate meaningful clinical benefit with only short-lived serum tumour markers declines and disease stabilisation in a few patients.71,72
Importantly, several targeted-therapy studies in refractory GCCs – including ADCs – were conducted in largely unselected populations without biomarker enrichment, which may have diluted potential efficacy signals. Biomarker-driven selection (e.g. confirmed target expression in viable tumour tissue) should be prioritised in future trial design.
Claudin-6 as a therapeutic target
Claudin-6 (CLDN6) is an oncofetal tight-junction protein normally confined to embryonic tissues and largely absent in adult organs. 73 In several cancers – particularly GCCs, ovarian, endometrial and certain lung tumours – CLDN6 is aberrantly overexpressed and contributes to epithelial proliferation and differentiation. 74 Nearly, all GCCs display CLDN6 expression, making it an attractive therapeutic target.
Multiple CLDN6-directed modalities are under clinical evaluation, including monoclonal antibodies, chimeric antigen receptor-T (CAR-T) cells, bispecific T-cell engagers (BiTEs) and ADCs. 69 The monoclonal antibody ASP1650 was the first tested but showed no objective responses in a phase II trial, with only transient disease stabilisation in 3 of 19 patients. 75
A more promising approach uses CLDN6-CAR-T cells (BNT211-01) in combination with the mRNA vaccine CARVac, which enhances CAR-T persistence by transient CLDN6 expression in antigen-presenting cells. Among 13 refractory non-seminoma patients treated, an ORR of ~60% and disease control rate (DCR) of 85% were achieved. 76 However, cytokine-release syndrome occurred in 65%, cytopenias in 59% and fatal adverse events in ~9%. 77
In the BNT142-01 trial (NCT05262530), an mRNA-encoded anti-CLDN6/CD3 bispecific antibody (RiboMab02.1) induced a dose-dependent immune activation and a 58% DCR in 65 patients (10 GCCs) with manageable toxicity. 78
Two CLDN6-targeting ADCs have entered phase I evaluation: DS-9606a, a pyrrolobenzodiazepine-based ADC that induced PRs in 2 of 7 GCC patients and ⩾90% tumour-marker reductions in most cases, 79 but whose clinical development was recently discontinued; and TORL-1-23, an Monomethyl auristatin E (MMAE)-conjugated ADC that achieved overall response rates of 26%–42% and disease-control rates of 68%–89% in the total study population, including six patients with GCCs. 80
Together, these results identify CLDN6 as a promising therapeutic target in refractory GCCs, though current data remain limited and confirmatory trials are awaited (see Table 3).
Clinical trials investigating novel therapeutic targets, CD30 and CLDN6, enrolling patients with refractory GCTs.
Notes (ClinicalTrials.gov):
-Completed: The study has ended normally, and participants are no longer being examined or treated.
-Terminated: The study has stopped early and will not start again. Participants are no longer being examined or treated.
-Active, not recruiting: The study is ongoing, and participants are receiving an intervention or being examined, but potential participants are not currently being recruited or enrolled.
ADC, antibody drug conjugates; AXL, AXL receptor tyrosine kinase; BiTE, bispecific T-cell engager; BV, Brentuximab vedotin; CAR-NK, chimeric antigen receptor-natural killer cells; CAR-T, chimeric antigen receptor-T; CARVac, CAR T cell-amplifying RNA vaccine; CD137, cluster of differentiation 137; CD3, cluster of differentiation 3; CLDN6, claudin-6; CR, complete response; DCR, disease control rate; GPC3, glypican-3; mAb, monoclonal antibody; mGCT, metastatic germ cell tumour; OR, objective response; ORR, objective response rate; PR, partial response; pts, patients; SD, stable disease; TriTE, trispecific T-cell engager.
Other ADCs targeting refractory GCCs
Given the histology-agnostic approval of trastuzumab deruxtecan for HER2-positive solid tumours, assessing HER2 in GCCs may be relevant. Although no GCC patients were included in DESTINY trials, early IHC studies found HER2 overexpression in ~35% of samples, mainly in post-chemotherapy or mediastinal GCCs, but absent in primary testicular tumours.85,86 Thus, HER2 testing may be justified in selected late-relapses or mediastinal cases.
TROP-2, expressed in about half of GCCs – especially choriocarcinomas and yolk-sac tumours – represents another potential ADC target. The TROP-2-directed ADC datopotamab deruxtecan is under evaluation across tumour types in the TROPION-PanTumour03 trial, though GCCs are not yet included. 87 In addition, emerging descriptive pathology data suggest that Nectin-4 expression may be enriched in specific histologic subtypes, particularly choriocarcinoma, supporting further exploration of Nectin-4-directed ADCs in biomarker-selected cohorts. 88
Future directions and translational priorities
The therapeutic plateau in multiply relapsed, platinum-refractory GCCs highlights the need for biomarker-driven strategies that are tightly linked to tumour biology. A first priority is systematic molecular profiling of viable tumour tissue at relapse, integrating histology (including somatic malignant transformation), immunophenotyping and genomics/epigenomics to identify actionable vulnerabilities and to avoid target-negative enrollment.
From a translational perspective, CLDN6 represents the most compelling near-term target, with early clinical activity across CAR-T, BiTEs and ADC platforms. Key next steps include defining optimal patient selection (e.g. CLDN6 expression thresholds in viable tumour, spatial heterogeneity), improving safety management (cytopenias, CRS, infectious risk) and establishing sequencing concepts with surgery and cytotoxic salvage regimens.
Beyond CLDN6, a rational roadmap should include (i) histology-enriched targeting approaches (e.g. subtype-associated surface antigens), (ii) epigenetic re-sensitisation concepts that leverage the distinctive methylation biology of GCCs and (iii) combination strategies designed to overcome the low neo antigenicity that likely limits checkpoint inhibitor efficacy.
Finally, future trials should incorporate modern design features appropriate for this ultra-rare setting: biomarker enrichment, centralised pathology review, harmonised definitions of platinum resistance, mandatory reporting of surgical resection status and relapse histology and endpoints that capture durable disease control (including post-chemotherapy complete resection rates and long-term NED status). International collaboration and prospective registries will be essential to generate sufficiently powered datasets and to accelerate validation of emerging targets.
Conclusion
Platinum-refractory GCCs remain a therapeutic challenge. While conventional chemotherapy and high-dose regimens still define standard care, novel molecularly targeted and immune-based approaches – particularly CLDN6-directed CAR-T, bispecific and ADC therapies – show early promise. Broader prospective studies and biomarker-driven selection will be key to establishing their clinical role.
