Sequential versus single high-dose chemotherapy in patients with relapsed or refractory germ cell tumors: long-term results of a prospective randomized trial.

PURPOSE To evaluate the long-term survival rates in patients with relapsed or refractory germ cell tumors (GCTs) after single or sequential high-dose chemotherapy (HDCT). PATIENTS AND METHODS Between November 1999 and November 2004, 211 patients with relapsed or refractory GCT were randomly assigned to treatment with either one cycle of cisplatin 100 mg/m(2), etoposide 375 mg/m(2), and ifosfamide 6 g/m(2) (VIP) plus three cycles of high-dose carboplatin 1,500 mg/m(2) and etoposide 1,500 mg/m(2) (CE, arm A) or three cycles of VIP plus one cycle of high-dose carboplatin 2,200 mg/m(2), etoposide 1,800 mg/m(2), and cyclophosphamide 6,400 mg/m(2) (CEC, arm B) followed by autologous stem-cell reinfusion. Long-term progression-free survival (PFS) and overall survival (OS) 6 years after random assignment of the last patient were compared by using the log-rank test. RESULTS Overall, 108 and 103 patients were randomly assigned to arms A and B, respectivelyl. The study was stopped prematurely because of excess treatment-related mortality in arm B (14%) compared with that in arm A (4%; P = .01). As of December 2010, nine (5%) of 211 patients were lost to follow-up; 94 (45%) of 211 are alive and 88 (94%) of 94 patients are progression free. Five-year PFS is 47% (95% CI, 37% to 56%) in arm A and 45% (95% CI, 35% to 55%) in arm B (hazard ratio [HR], 1.16; 95% CI, 0.79 to 1.70; P = .454). Five-year OS is 49% (95% CI, 40% to 59%) in arm A and 39% (95% CI, 30% to 49%) in arm B (HR, 1.42; 95% CI, 0.99 to 2.05; P = .057). CONCLUSION Patients with relapsed or refractory GCT achieve durable long-term survival after single as well as sequential HDCT. Fewer early deaths related to toxicity translated into superior long-term OS after sequential HDCT.