Background: PC is a major health concern worldwide and a deadly disease due to its high metastatic behavior. In recent years, incremental progresses have been made with the use of new chemotherapy (CT) regimen in the metastatic setting. Thus, both PRODIGE 4/ACCORD11 (Conroy T, et al. 2011) and MPACT (Von Hoff D, et al. 2013) phase III trials established 2 new standard-of-care in the first line treatment of metastatic PC (mPC), demonstrating a survival benefit over gemcitabine monotherapy, with the use of FFX or GA. In the phase I/II GABRINOX trial (Assenat E, et al. ESMO Open 2021), we reported that sequential GA followed by FFX provided a high overall response rate (ORR) (64.9%) and a promising median progression-free survival (PFS, 10.5 months) and median overall survival (OS, 15.1 months), together with acceptable toxicity and remarkably low severe neurotoxicity rate (gr.3: 5.3%). To follow up these encouraging results in a controlled study, we aimed at comparing our experimental GABRINOX regimen to control FFX in the GABRINOX-2 randomized phase 2 trial (NCT05065801). Our primary objective is PFS, and our secondary objectives are tolerance, ORR, disease-control rate, OS, and Quality-of-life. Methods: Main inclusion criteria were as follows: Patients (pts) in good condition (ECOG PS ≤ 1), aged from 18 to 75 yo, with histologically or cytological proven mPC and at least one measurable metastatic target. Pts should have not been treated with (adjuvant) chemotherapy in the last 6 months. Eligible pts are randomized (ratio 1:1) either in the standard FFX group or in the experimental GABRINOX group where a GA (gemcitabine 1000 mg/m² and Nab-paclitaxel 125 mg/m², day 1-8-15) cycle alternates after a 2-weeks rest with a FFX cycle. To detect an increase in median PFS from 6.4 to 10.5 months (HR = 0.61) with a 80% power and a 5% α risk, 130 events are required among a total population of 210 pts. PFS was defined as the length of time between randomization and the onset of 1^st documented progression (RECIST 1.1 criteria) or death. The study of quality of life will use the EORTC QLQ-C30 and QLQ-PAN26 self-reported questionnaires at baseline and every 2 months up to 12 months and then at 16, 20 and 24 months. Circulating DNA tests will be carried out at baseline and every 2 months until progression. All numerical variables will be expressed as medians and 95% CI, while PFS and OS will be estimated using Kaplan-Meier method. Multivariate analyses will use Cox proportional hazard model. Enrolment started in late 2021 and 3 patients were included so far. Clinical trial information: NCT05065801.