Durvalumab Alone or With Tremelimumab in Refractory Germ Cell Tumors

Objective response-rate

Background:
 
 The prognosis of pts who have failed multiple chemotherapy (CT) regimens is quite dismal.
 PD-L1 is frequently expressed by immunohistochemistry (IHC) in germ cell tumors (GCT). D is a
 monoclonal antibody (mAb) that inhibits the binding of PD-L1. T, an anti-CTLA4 mAb, is an
 immunomodulatory therapy. Combination immunotherapy has shown improved activity compared to
 monotherapy. The investigators aimed to investigate the activity of D, alone or in
 combination with T, in chemorefractory GCT.
 
 Trial Design:
 
 This is an open-label, randomized, 3-stage, phase 2 study. Pts who have failed ≥2 prior CT
 regimens (including high-dose CT) will be randomized to receive one of the following: D, 1.5
 g via IV infusion q4w, for up to a total of 12 months (13 doses/cycles) alone or with T, 75
 mg IV q4w, starting on week 0, for up to 4 months (4 doses/cycles). Serum tumor markers,
 computed tomography and fluorodeoxyglucose positron emission tomography (FDG-PET) scans will
 be repeated q8 weeks. The primary endpoint is the objective response-rate (ORR=complete
 response or partial response with normal markers). H0: ORR rate ≤10%, H1: ORR ≥25%, type I
 and II error rates at 10%.
 
 In stage 1, 11 pts will be allocated in each arm. According to Gehan's rule, the trial will
 be terminated whenever no response will be observed. 29 additional pts will be added to each
 arm fulfilling stage 1 criteria. ORR in ≥7 pts will be required. In stage 3, pts from stage
 1-2 of both arms will be retrospectively evaluated for Programmed cell Death Ligand-1(PD-L1)
 IHC. The Ventana PD-L1 IHC assay will be used. In case of negative findings at the end of
 stage 2, if the target benefit is likely to occur only in PD-L1+ pts, further study
 prosecution in accordance with an enrichment strategy will be undertaken.
 
 In particular, predictive power (PP) will be calculated assuming expansion of PD-L1+ cohorts
 up to a maximum of 60 pts. Each arm will be categorized as not promising (PP30%) or
 promising (PP ≥30%). The promising one will enter the stage 3. Should both arms be judged
 promising, the one yielding ≥20% PP advantage will be selected; monotherapy will be preferred
 otherwise. Details on the algorithm to be used for PD-L1 IHC in this study will be finalized
 (EudraCT number 2016-001688-35).