Immediate versus deferred chemotherapy after radical cystectomy in patients with pT3-pT4 or N+ M0 urothelial carcinoma of the bladder (EORTC 30994): an intergroup, open-label, randomised phase 3 trial.

BACKGROUND: Patients with muscle-invasive urothelial carcinoma of the bladder have poor survival after cystectomy. The EORTC 30994 trial aimed to compare immediate versus deferred cisplatin-based combination chemotherapy after radical cystectomy in patients with pT3-pT4 or N+ M0 urothelial carcinoma of the bladder. METHODS: This intergroup, open-label, randomised, phase 3 trial recruited patients from hospitals across Europe and Canada. Eligible patients had histologically proven urothelial carcinoma of the bladder, pT3-pT4 disease or node positive (pN1-3) M0 disease after radical cystectomy and bilateral lymphadenectomy, with no evidence of any microscopic residual disease. Within 90 days of cystectomy, patients were centrally randomly assigned (1:1) by minimisation to either immediate adjuvant chemotherapy (four cycles of gemcitabine plus cisplatin, high-dose methotrexate, vinblastine, doxorubicin, and cisplatin [high-dose MVAC], or MVAC) or six cycles of deferred chemotherapy at relapse, with stratification for institution, pT category, and lymph node status according to the number of nodes dissected. Neither patients nor investigators were masked. Overall survival was the primary endpoint; all analyses were by intention to treat. The trial was closed after recruitment of 284 of the planned 660 patients. This trial is registered with ClinicalTrials.gov, number NCT00028756. FINDINGS: From April 29, 2002, to Aug 14, 2008, 284 patients were randomly assigned (141 to immediate treatment and 143 to deferred treatment), and followed up until the data cutoff of Aug 21, 2013. After a median follow-up of 7.0 years (IQR 5.2-8.7), 66 (47%) of 141 patients in the immediate treatment group had died compared with 82 (57%) of 143 in the deferred treatment group. No significant improvement in overall survival was noted with immediate treatment when compared with deferred treatment (adjusted HR 0.78, 95% CI 0.56-1.08; p=0.13). Immediate treatment significantly prolonged progression-free survival compared with deferred treatment (HR 0.54, 95% CI 0.4-0.73, p<0.0001), with 5-year progression-free survival of 47.6% (95% CI 38.8-55.9) in the immediate treatment group and 31.8% (24.2-39.6) in the deferred treatment group. Grade 3-4 myelosuppression was reported in 33 (26%) of 128 patients who received treatment in the immediate chemotherapy group versus 24 (35%) of 68 patients who received treatment in the deferred chemotherapy group, neutropenia occurred in 49 (38%) versus 36 (53%) patients, respectively, and thrombocytopenia in 36 (28%) versus 26 (38%). Two patients died due to toxicity, one in each group. INTERPRETATION: Our data did not show a significant improvement in overall survival with immediate versus deferred chemotherapy after radical cystectomy and bilateral lymphadenectomy for patients with muscle-invasive urothelial carcinoma. However, the trial is limited in power, and it is possible that some subgroups of patients might still benefit from immediate chemotherapy. An updated individual patient data meta-analysis and biomarker research are needed to further elucidate the potential for survival benefit in subgroups of patients. FUNDING: Lilly, Canadian Cancer Society Research.

Personalised cancer medicine.

The evolving field of personalised medicine is playing an increasingly important role in cancer prevention, diagnosis, prognosis and therapeutics. Its importance in clinical management is demonstrated by the recent introduction into routine clinical practice of various individualised, molecularly targeted therapies with increased efficacy and/or reduced toxicity. The identification of cancer predisposition genes, such as the BRCA genes in breast cancer, permits screening programmes to identify patients "at-risk" of developing cancer and helps them make decisions on individual risk-modification behaviours. Personalised medicine also plays an increasingly important role in cancer treatment. It is increasingly clear that there are molecularly distinct subtypes of various common cancers, with different therapeutic approaches required for each subtype, for example, the use of the monoclonal antibodies (trastuzumab and cetuximab) in HER2-positive breast cancer and wild-type KRAS colorectal cancer; tyrosine kinase inhibitors (imatinib, gefitinib, erlotinib and crizotinib) in chronic myeloid leukaemia, gastrointestinal stromal tumours and non-small-cell lung cancer and intracellular agents (vemurafenib and olaparib) in metastatic malignant melanoma and ovarian, breast and prostate cancer. The efficacy of various targeted therapies in such disparate tumours suggests that we are entering an era in which treatment decisions will be based on tumour molecular abnormality profile or "signature," rather than tumour tissue type or anatomical site of origin, improving patient prognosis and quality of life. This mini review focuses on the role of personalised medicine in cancer prevention and treatment as well as its future direction in oncology.

Accelerated methotrexate, vinblastine, doxorubicin, and cisplatin (AMVAC) as neoadjuvant chemotherapy for patients with muscle-invasive transitional cell carcinoma of the bladder.

BACKGROUND: Meta-analysis data demonstrate a 5% absolute survival benefit for neoadjuvant chemotherapy (NAC) using cisplatin-based combination regimens in the radical treatment of muscle-invasive bladder cancer (MIBC). However, there are no randomized, controlled trial data on the optimum regimen. Accelerated methotrexate, vinblastine, doxorubicin, and cisplatin (AMVAC) is a dose-intense regimen that has the potential to minimize delays to definitive, potentially curative therapy. A retrospective analysis is presented of the efficacy and toxicity of AMVAC as NAC in patients with MIBC and its impact on the patient pathway. METHODS: Eighty consecutive patients with MIBC were treated with AMVAC as NAC by 2 UK multidisciplinary uro-oncology teams. Three or 4 cycles of AMVAC (methotrexate 30 mg/m(2) , vinblastine 3 mg/m(2) , doxorubicin 30 mg/m(2) , and cisplatin 70 mg/m(2) ) were given at 2-week intervals, with granulocyte colony-stimulating factor support, prior to either radical surgery or radical radiotherapy. RESULTS: All planned cycles of chemotherapy were completed, without dose reduction or delay in 84% of patients. All 80 patients subsequently received their planned definitive therapy. Grade 3/4 toxicities were seen in 26% of the 42% of patients for whom toxicity data are available, including 12% grade 3/4 neutropenia. Pathological complete response to AMVAC was seen in 43% of 60 surgical patients. Objective radiological local response was seen in 83% of 57 evaluable patients. Two-year disease-free and overall survival were 65% and 77%, respectively. CONCLUSIONS: AMVAC is safe and appears to be a well-tolerated and effective NAC regimen for MIBC. It minimizes delays to definitive treatment and produces excellent pathological and radiological response rates. It is an appropriate comparator for future randomized trials.

Development of a low-seroprevalence, αvß6 integrin-selective virotherapy based on human adenovirus type 10.

Oncolytic virotherapies (OV) hold immense clinical potential. OV based on human adenoviruses (HAdV) derived from HAdV with naturally low rates of pre-existing immunity will be beneficial for future clinical translation. We generated a low-seroprevalence HAdV-D10 serotype vector incorporating an αvß6 integrin-selective peptide, A20, to target αvß6-positive tumor cell types. HAdV-D10 has limited natural tropism. Structural and biological studies of HAdV-D10 knob protein highlighted low-affinity engagement with native adenoviral receptors CAR and sialic acid. HAdV-D10 fails to engage blood coagulation factor X, potentially eliminating "off-target" hepatic sequestration in vivo. We engineered an A20 peptide that selectively binds αvß6 integrin into the DG loop of HAdV-D10 fiber knob. Assays in αvß6+ cancer cell lines demonstrated significantly increased transduction mediated by αvß6-targeted variants compared with controls, confirmed microscopically. HAdV-D10.A20 resisted neutralization by neutralizing HAdV-C5 sera. Systemic delivery of HAdV-D10.A20 resulted in significantly increased GFP expression in BT20 tumors. Replication-competent HAdV-D10.A20 demonstrated αvß6 integrin-selective cell killing in vitro and in vivo. HAdV-D10 possesses characteristics of a promising virotherapy, combining low seroprevalence, weak receptor interactions, and reduced off-target uptake. Incorporation of an αvß6 integrin-selective peptide resulted in HAdV-D10.A20, with significant potential for clinical translation.

Efficient Intravenous Tumor Targeting Using the αvß6 Integrin-Selective Precision Virotherapy Ad5NULL-A20.

We previously developed a refined, tumor-selective adenovirus, Ad5NULL-A20, harboring tropism ablating mutations in each major capsid protein, to ablate all native means of infection. We incorporated a 20-mer peptide (A20) in the fiber knob for selective infection via αvß6 integrin, a marker of aggressive epithelial cancers. Methods: To ascertain the selectivity of Ad5NULL-A20 for αvß6-positive tumor cell lines of pancreatic and breast cancer origin, we performed reporter gene and cell viability assays. Biodistribution of viral vectors in mice harboring xenografts with low, medium, and high αvß6 levels was quantified by qPCR for viral genomes 48 h post intravenous administration. Results: Ad5NULL-A20 vector transduced cells in an αvß6-selective manner, whilst cell killing mediated by oncolytic Ad5NULL-A20 was αvß6-selective. Biodistribution analysis following intravenous administration into mice bearing breast cancer xenografts demonstrated that Ad5NULL-A20 resulted in significantly reduced liver accumulation coupled with increased tumor accumulation compared to Ad5 in all three models, with tumor-to-liver ratios improved as a function of αvß6 expression. Conclusions: Ad5NULL-A20-based virotherapies efficiently target αvß6-integrin-positive tumors following intravenous administration, validating the potential of Ad5NULL-A20 for systemic applications, enabling tumor-selective overexpression of virally encoded therapeutic transgenes.

Identification of folate receptor α (FRα) binding oligopeptides and their evaluation for targeted virotherapy applications.

Oncolytic virotherapies (OV) based on human adenoviral (HAdV) vectors hold significant promise for the treatment of advanced ovarian cancers where local, intraperitoneal delivery to tumour metastases is feasible, bypassing many complexities associated with intravascular delivery. The efficacy of HAdV-C5-based OV is hampered by a lack of tumour selectivity, where the primary receptor, hCAR, is commonly downregulated during malignant transformation. Conversely, folate receptor alpha (FRα) is highly expressed on ovarian cancer cells, providing a compelling target for tumour selective delivery of virotherapies. Here, we identify high-affinity FRα-binding oligopeptides for genetic incorporation into HAdV-C5 vectors. Biopanning identified a 12-mer linear peptide, DWSSWVYRDPQT, and two 7-mer cysteine-constrained peptides, CIGNSNTLC and CTVRTSAEC that bound FRα in the context of the phage particle. Synthesised lead peptide, CTVRTSAEC, bound specifically to FRα and could be competitively inhibited with folic acid. To assess the capacity of the elucidated FRα-binding oligopeptides to target OV to FRα, we genetically incorporated the peptides into the HAdV-C5 fiber-knob HI loop including in vectors genetically ablated for hCAR interactions. Unfortunately, the recombinant vectors failed to efficiently target transduction via FRα due to defective intracellular trafficking following entry via FRα, indicating that whilst the peptides identified may have potential for applications for targeted drug delivery, they require additional refinement for targeted virotherapy applications.

Histone deacetylase inhibitor trichostatin A sensitises cisplatin-resistant ovarian cancer cells to oncolytic adenovirus.

Ovarian cancer is often termed a silent killer due to the late onset of symptoms. Whilst patients initially respond to chemotherapy, they rapidly develop chemo-resistance. Oncolytic adenoviruses (OAds) are promising anti-cancer agents engineered to "hijack" the unique molecular machinery of cancer cells enabling tumour-selective viral replication. This allows spread to adjacent cells and amplification of oncolysis within the tumour. OAds represent an excellent opportunity for ovarian cancer therapy via intra-peritoneal delivery, however the efficacy of OAds thus far is limited. Here, we evaluate chromatin (histone) modification in chemo-resistant cells and its relationship to Ad efficacy (wild-type or oncolytic Ad). In contrast to cisplatin-sensitive A2780 cells that show an efficient reduction of cell viability by Ad in the presence of cisplatin, cisplatin-resistant A2780/cp70 cells show diminishing Ad-mediated reduction of cell viability with escalating doses of cisplatin. Histone deacetylase (HDAC)-2 and to a lesser extent HDAC1 were up-regulated in cisplatin-resistant but not cisplatin-sensitive cells. Cisplatin-resistant cells treated with a pan-HDAC inhibitor trichostatin A (TsA) significantly enhanced Ad-mediated reduction of cell viability in the presence of cisplatin. Cells treated with TsA alone did not reduce cell viability suggesting these findings are Ad-dependent. Thus, we identify HDAC inhibition as a potential means to sensitise cisplatin-resistant ovarian cancer cells to virotherapies, an observation that may offer improved outcomes for patients with late stage, chemotherapy-resistant ovarian cancer.

Ad5NULL-A20: A Tropism-Modified, αvß6 Integrin-Selective Oncolytic Adenovirus for Epithelial Ovarian Cancer Therapies.

Purpose: Virotherapies are maturing in the clinical setting. Adenoviruses (Ad) are excellent vectors for the manipulability and tolerance of transgenes. Poor tumor selectivity, off-target sequestration, and immune inactivation hamper clinical efficacy. We sought to completely redesign Ad5 into a refined, tumor-selective virotherapy targeted to αvß6 integrin, which is expressed in a range of aggressively transformed epithelial cancers but nondetectable in healthy tissues.Experimental Design: Ad5NULL-A20 harbors mutations in each major capsid protein to preclude uptake via all native pathways. Tumor-tropism via αvß6 targeting was achieved by genetic insertion of A20 peptide (NAVPNLRGDLQVLAQKVART) within the fiber knob protein. The vector's selectivity in vitro and in vivo was assessed.Results: The tropism-ablating triple mutation completely blocked all native cell entry pathways of Ad5NULL-A20 via coxsackie and adenovirus receptor (CAR), αvß3/5 integrins, and coagulation factor 10 (FX). Ad5NULL-A20 efficiently and selectively transduced αvß6+ cell lines and primary clinical ascites-derived EOC ex vivo, including in the presence of preexisting anti-Ad5 immunity. In vivo biodistribution of Ad5NULL-A20 following systemic delivery in non-tumor-bearing mice was significantly reduced in all off-target organs, including a remarkable 107-fold reduced genome accumulation in the liver compared with Ad5. Tumor uptake, transgene expression, and efficacy were confirmed in a peritoneal SKOV3 xenograft model of human EOC, where oncolytic Ad5NULL-A20-treated animals demonstrated significantly improved survival compared with those treated with oncolytic Ad5.Conclusions: Oncolytic Ad5NULL-A20 virotherapies represent an excellent vector for local and systemic targeting of αvß6-overexpressing cancers and exciting platforms for tumor-selective overexpression of therapeutic anticancer modalities, including immune checkpoint inhibitors. Clin Cancer Res; 24(17); 4215-24. ©2018 AACR.

The Wales Cancer Research Centre: improving cancer care together.

The Wales Cancer Research Centre (WCRC), funded by the Welsh Government, through Health and Care Research Wales, is our latest weapon in the ongoing war against cancer. It is Wales' contribution to a global effort to reverse the increasing trend in numbers of newly diagnosed cancers. With the increasing need for new and better cancer treatments, more research is needed, and the WCRC hopes to fulfill some of this demand. WCRC is particularly concerned with the problems of cancer patients in Wales, but also aims to help cancer patients beyond Wales by performing cancer research of world class quality.

Pseudotyped αvß6 integrin-targeted adenovirus vectors for ovarian cancer therapies.

Encouraging results from recent clinical trials are revitalizing the field of oncolytic virotherapies. Human adenovirus type 5 (HAdV-C5/Ad5) is a common vector for its ease of manipulation, high production titers and capacity to transduce multiple cell types. However, effective clinical applications are hindered by poor tumor-selectivity and vector neutralization. We generated Ad5/kn48 by pseudotyping Ad5 with the fiber knob domain from the less seroprevalent HAdV-D48 (Ad48). The vector was shown to utilize coxsackie and adenovirus receptor (CAR) but not CD46 for cell entry. A 20-amino acid peptide NAVPNLRGDLQVLAQKVART (A20) was inserted into the Ad5. Luc HI loop (Ad5.HI.A20) and Ad5/kn48 DG loop (Ad5/kn48.DG.A20) to target a prognostic cancer cell marker, αvß6 integrin. Relative to the Ad5.Luc parent vector, Ad5.HI.A20, Ad5.KO1.HI.A20 (KO1, ablated CAR-binding) and Ad5/kn48.DG.A20 showed ~ 160-, 270- and 180-fold increased transduction in BT-20 breast carcinoma cells (αvß6high). Primary human epithelial ovarian cancer (EOC) cultures derived from clinical ascites provided a useful ex vivo model for intraperitoneal virotherapy. Ad5.HI.A20, Ad5.KO1.HI.A20 and Ad5/kn48.DG.A20 transduction was ~ 70-, 60- and 16-fold increased relative to Ad5.Luc in EOC cells (αvß6high), respectively. A20 vectors transduced EOC cells at up to ~ 950-fold higher efficiency in the presence of neutralizing ovarian ascites, as compared to Ad5.Luc. Efficient transduction and enhanced cancer-selectivity via a non-native αvß6-mediated route was demonstrated, even in the presence of pre-existing anti-Ad5 immunity. Consequently, αvß6-targeted Ad vectors may represent a promising platform for local intraperitoneal treatment of ovarian cancer metastases.

Phase I and pharmacokinetic study of intravenous irinotecan plus oral ciclosporin in patients with fuorouracil-refractory metastatic colon cancer.

PURPOSE: To assess the safety and toxicity profile of escalating doses of intravenous irinotecan, in combination with a fixed dose of oral ciclosporin (Cs) and to determine the pharmacokinetic profile of irinotecan and its metabolites. PATIENTS AND METHODS: Patients with fluorouracil-refractory metastatic colorectal cancer received escalating doses of intravenous irinotecan from 40 to 125 mg/m(2) every 2 weeks in combination with a fixed dose of oral Cs (5 mg/kg bid for 3 days). Pharmacokinetic analysis of plasma irinotecan and its metabolites SN38 and SN38G was performed during paired cycles with and without Cs. RESULTS: Thirty-seven patients were treated. Dose-limiting toxicity of grade 4 neutropenia was seen at an irinotecan dose of 125 mg/m(2). There was no grade 4 diarrhea, and only one patient experienced grade 3 diarrhea. Toxicities caused by Cs were generally mild. Pharmacokinetic studies demonstrated that irinotecan clearance was reduced from 13.4 to 5.8 L/h/m(2) and area under the curve (AUC)(0-tn) was increased 2.2-fold by the coadministration of Cs. Similar significant increases in AUC(0-24h) were seen for both SN38 and SN38G (2.2-fold and 2.3-fold, respectively) in the presence of Cs. Antitumor activity was seen at every irinotecan dose level. CONCLUSION: The maximum tolerated irinotecan dose and recommended dose for phase II studies is 100 mg/m(2) every 2 weeks. Dose-limiting diarrhea was not seen during this study, supporting the hypothesis that pharmacokinetic modulation of irinotecan by Cs may improve its therapeutic index. Further studies using this combination are warranted.

Incorporation of Peptides Targeting EGFR and FGFR1 into the Adenoviral Fiber Knob Domain and Their Evaluation as Targeted Cancer Therapies.

Oncolytic virotherapies based on adenovirus 5 (Ad5) hold promise as adjunctive cancer therapies; however, their efficacy when delivered systemically is hampered by poor target cell specificity and preexisting anti-Ad5 immunity. Ovarian cancer represents a promising target for virotherapy, since the virus can be delivered locally into the peritoneal cavity. Both epidermal growth factor receptor (EGFR) and fibroblast growth factor receptor 1 (FGFR1) are overexpressed in the majority of human tumors, including ovarian cancer. To generate adenoviral vectors with improved tumor specificity, we generated a panel of Ad5 vectors with altered tropism for EGFR and FGFR, rather than the natural Ad5 receptor, hCAR. We have included mutations within AB loop of the viral fiber knob (KO1 mutation) to preclude interaction with hCAR, combined with insertions in the HI loop to incorporate peptides that bind either EGFR (peptide YHWYGYTPQNVI, GE11) or FGFR1 (peptides MQLPLAT, M*, and LSPPRYP, LS). Viruses were produced to high titers, and the integrity of the fiber protein was validated by Western blotting. The KO1 mutation efficiently ablated hCAR interactions, and significantly increased transduction was observed in hCAR(low)/EGFR(high) cell lines using Ad5.GE11, while transduction levels using Ad5.M* or Ad5.LS were not increased. In the presence of physiological concentrations of human blood clotting factor X (hFX), significantly increased levels of transduction via the hFX-mediated pathway were observed in cell lines, but not in primary tumor cells derived from epithelial ovarian cancer (EOC) ascites samples. Ad5-mediated transduction of EOC cells was completely abolished by the presence of 2.5% serum from patients, while, surprisingly, incorporation of the GE11 peptide resulted in significant evasion of neutralization in the same samples. We thus speculate that incorporation of the YHWYGYTPQNVI dodecapeptide within the fiber knob domain may provide a novel means of circumventing preexisting Ad5 immunity that warrants further investigation.

Systemic chemotherapy for patients with bladder cancer--current controversies and future directions.

Many localised, superficial bladder cancers can be effectively controlled. However, disease which has spread to nodes outside the pelvis or to distant organs is generally incurable and systemic therapies, rather than surgery, are appropriate. Combination chemotherapy based around established cytotoxic drugs such as cisplatin has proven benefit in palliating symptoms and prolonging survival in responsive patients with advanced disease. Combination chemotherapies which include newer cytotoxic drugs such as gemcitabine provide the potential for equivalent efficacy with less toxicity than established regimens. Between the extremes of superficial and advanced disease, muscle-invasive bladder cancers have traditionally been treated, with curative intent, by radical surgery or radiotherapy. However, newly published data suggest, for the first time, genuine survival benefits from peri-operative chemotherapy. This article reviews the evidence for cisplatin-based chemotherapy in advanced disease, assesses the potential benefits of newer cytotoxic drugs, discusses the latest evidence pertaining to peri-operative chemotherapy in muscle-invasive disease, and looks forward to potential new biological agents in the systemic therapy of bladder cancer.

Immune recruitment and therapeutic synergy: keys to optimizing oncolytic viral therapy?

Oncolytic viruses consist of a diverse range of DNA and RNA viruses traditionally thought to mediate their effects by exploiting aberrations in tumor pathways, allowing preferential viral replication in, and killing of, tumor cells. Clinical development has progressed to late-phase trials, potentially heralding their introduction into clinical practice. However, despite this promise, the activity of oncolytic viruses has yet to achieve the potential suggested in preclinical models. To address this disparity, we need to recognize the complex interaction among oncolytic viruses, tumor, chemotherapy, and host immune system, and appreciate that direct oncolysis may not be the only factor to play an important role in oncolytic virus-mediated antitumor efficacy. Although key in inactivating viruses, the host immune system can also act as an ally against tumors, interacting with oncolytic viruses under the right conditions to generate useful and long-lasting antitumor immunity. Preclinical data also suggest that oncolytic viruses show synergy with standard therapies, which may offer improved clinical response rates. Here, we explore clinical and preclinical data on clinically relevant oncolytic viruses, highlighting areas of progress, uncertainty, and translational opportunity, with respect to immune recruitment and therapeutic synergy.

High-dose chemotherapy with haematopoietic stem-cell support in patients with poor prognosis, relapsed or refractory germ cell tumours.

OBJECTIVE: To report our experience of high-dose chemotherapy (HDC) with haematopoietic stem-cell support (HSC) in patients with poor risk, relapsed or refractory germ cell tumours (GCTs), as this treatment might offer effective salvage for patients with disseminated GCTs. PATIENTS AND METHODS: We retrospectively reviewed the medical records and database for 33 patients with GCT who were treated with HDC with HSC in our centres. RESULTS: Thirty-three patients were treated with either one or two cycles of carboplatin and etoposide-based HDC with HSC support, between March 1990 and October 2003. Twenty-six patients (79%) had nonseminomatous GCT, six seminoma (18%), and one (3%) a combined seminoma and teratoma. Twenty patients (60%) had previously had a clinical complete response after previous chemotherapy +/- surgery for residual disease. Most patients were treated with HDC for relapsing (49%) or relative refractory disease (30%), but seven (21%) had HDC in the first partial remission. The complete response rate to HDC was 58%. There were two treatment-related deaths (6%). As of April 2005, 18 patients were alive and disease-free with a median (range) follow-up of 72 (0.5-174) months. The 5-year overall and progression-free survival probabilities were 57% and 56%, respectively. The median (range) times to absolute neutrophil count recovery (> or = 500/microL) were 13 (9-24) and 12 (10-15) days, and for platelet count recovery ( > or = 20,000/microL) were 16 (7-50) and 13 (11-17) days, in the first and second cycles, respectively. CONCLUSION: The role of HDC with HSC support in metastatic GCTs remains controversial, and data from randomized controlled trials are needed. Our experience suggests that, in selected patients, this approach might be a useful form of salvage therapy.