Personalised chemotherapy based on tumour marker decline in poor prognosis germ-cell tumours (GETUG 13): a phase 3, multicentre, randomised trial.

backgroundPoor prognosis germ-cell tumours are only cured in about half of patients. We aimed to assess whether treatment intensification based on an early tumour marker decline will improve progression-free survival for patients with germ-cell tumours.

methodsIn this phase 3, multicentre, randomised trial, patients were enrolled from France (20 centres), USA (one centre), and Slovakia (one centre). Patients were eligible if they were older than 16 years, had evidence of testicular, retroperitoneal, or mediastinal non-seminomatous germ cell tumours based on histological findings or clinical evidence and highly elevated serum human chorionic gonadotropin or alfa-fetoprotein concentrations that matched International Germ Cell Cancer Consensus Group poor prognosis criteria. After one cycle of BEP (intravenous cisplatin [20 mg/m(2) per day for 5 days], etoposide [100 mg/m(2) per day for 5 days], and intramuscular or intravenous bleomycin [30 mg per day on days 1, 8, and 15]), patients' human chorionic gonadotropin and alfa-fetoprotein concentrations were measured at day 18-21. Patients with a favourable decline in human chorionic gonadotropin and alfa-fetoprotein continued BEP (Fav-BEP group) for 3 additonal cycles, whereas patients with an unfavourable decline were randomly assigned (1:1) to receive either BEP (Unfav-BEP group) or a dose-dense regimen (Unfav-dose-dense group), consisting of intravenous paclitaxel (175 mg/m(2) over 3 h on day 1) before BEP plus intravenous oxaliplatin (130 mg/m(2) over 3 h on day 10; two cycles), followed by intravenous cisplatin (100 mg/m(2) over 2 h on day 1), intravenous ifosfamide (2 g/m(2) over 3 h on days 10, 12, and 14), plus mesna (500 mg/m(2) at 0, 3, 7 and 11 h), and bleomycin (25 units per day, by continuous infusion for 5 days on days 10-14; two cycles), with granulocyte-colony stimulating factor (lenograstim) support. Centrally blocked computer-generated randomisation stratified by centre was used. The primary endpoint was progression-free survival and the efficacy analysis was done in the intention-to-treat population. The planned trial accrual was completed in May, 2012, and follow-up is ongoing. This study is registered with ClinicalTrials.gov, number NCT00104676.

findingsBetween Nov 28, 2003, and May 16, 2012, 263 patients were enrolled and 254 were available for tumour marker assessment. Of these 51 (20%) had a favourable marker assessment, and 203 (80%) had an unfavourable tumour marker decline; 105 were randomly assigned to the Unfav-dose-dense group and 98 to the Unfav-BEP group. 3-year progression-free survival was 59% (95% CI 49-68) in the Unfav-dose-dense group versus 48% (38-59) in the Unfav-BEP group (HR 0·66, 95% CI 0·44-1·00, p=0·05). 3-year progression-free survival was 70% (95% CI 57-81) in the Fav-BEP group (HR 0·66, 95% CI 0·49-0·88, p=0·01 for progression-free survival compared with the Unfav-BEP group). More grade 3-4 neurotoxic events (seven [7%] vs one [1%]) and haematotoxic events occurred in the Unfav-dose-dense group compared with in the Unfav-BEP group; there was no difference in grade 1-2 febrile neutropenia (18 [17%] vs 18 [18%]) or toxic deaths (one [1%] in both groups). Salvage high-dose chemotherapy plus a stem-cell transplant was required in six (6%) patients in the Unfav-dose-dense group and 16 (16%) in the Unfav-BEP group.

interpretationPersonalised treatment with chemotherapy intensification reduces the risk of progression or death in patients with poor prognosis germ-cell tumours and an unfavourable tumour marker decline.

fundingInstitut National du Cancer (Programme Hospitalier de Recherche Clinique).