Patterns of relapse in poor-prognosis germ cell tumors in the GETUG 13 trial: Implications for assessment of brain progression.

Abstract

365 Background: GETUG 13 investigated personalized chemotherapy based on tumor marker decline in patients with poor-prognosis GCT and demonstrated that a dose-dense regimen improves progression-free survival in patients with an unfavorable decline (Fizazi K, ASCO 2013). We investigated the pattern of relapse for patients included in GETUG 13. Methods: We conducted an analysis of relapse events in patients from GETUG 13 and an unfavorable decline of tumor markers (n=203). Baseline procedures before inclusion in the trial comprised a thoraco-abdomino-pelvic CT scan and an MRI of the brain: 22 patients (10%) had evidence of brain metastases at presentation. Results: With a median follow-up of 4.1 years (0.3 ; 8.8 years) when the analysis was performed, a progression event was observed in 94/203 patients (46%). First event consisted in a marker progression only in 41 patients (43%), a radiographic progression only in 29 patients (31%), a mix progression on both markers and imaging in 11 patients (12%), and death in 13 patients (14%). In patients with radiographic progression only, brain was the predominant site (n=16/29, 55%). Among all patients who experienced a radiographic progression (as first and subsequent progression event, n=58), brain only was the site of progression in 26 patients (45%): 12/30 (40%) in patients treated with BEP and 14/28 (50%) in those treated with dose-dense chemotherapy. Conclusions: Brain metastases develop often, early, and frequently as the only site of relapse in the course of poor-prognosis GCT. This raises the question of early detection and treatment of brain metastases in these patients, for example by integrating a systematic brain MRI after 2-3 months of chemotherapy.