702PSTANDARD CLINICAL PRACTICE OF FOLFIRINOX (FFX) IN ADVANCED/METASTATIC PANCREATIC CANCER (PC) PATIENTS: A CANADIAN RETROSPECTIVE REGISTRY.

ABSTRACT Aim: FFX has been adopted as first-line treatment in patients with good performance status (PS) and liver function. This registry further investigates this regimen in clinical practice and collects disease outcomes in an unselected population. Methods: This retrospective cohort study with PC patients who completed at least one cycle of FFX. Results: 132 evaluable patients with unresectable PC (metastatic 72%, locally advanced disease 28%) enrolled between December 2012 and May 2014 in 3 Canadian centers (TOHCC 53%, SRI 39%, WRH 8%); 55% were male; 42% were 65 years or older; median age was 64 years (range 24-80); 51%/33%/2%/14% had PS 0/1/2/non-evaluable, location 64% head and 36% body/tail (33% had a biliary stent). 36% patients started FFX at standard dose while 20% had more than 1 drug dose adjusted and 44% received no 5FU bolus. Patients received a median of 7.7 (range1-34) cycles of FOLFIRINOX chemotherapy. Relative dose intensity was 69% oxaliplatin, 67% irinotecan, 72% infusional 5FU. Most common grade 3+ toxicities were neutropenia 25%, febrile neutropenia 7.6%, nausea / vomiting 15%, fatigue 11%. Colony stimulating factors (G-CFS) prophylaxis was used in 27% patients (primary 14%, secondary 13%). Mean duration of follow up was 18.6 months. The median PFS and OS were respectively 5.5 mths (95%CI 3.4-6.7) and 7.5 mths (95%CI 6.3-9.5). While PFS shows no relevant variation by the FFX starting dose, OS differs if FFX dose is reduced (standard 11.4 mths 95%CI [6.5-13.1] vs. reduced 7.14 mths, 95%CI [5.5-8.1]; HR 1.67, 95%CI [1.1-2.5], p = 0.02) or if 5FU bolus is cancelled (with 5FUb 8.7 mths 95%CI [6.6-11.9] vs. without 5FUb 6.3 mths 95%CI [5.0-8.2]; HR 1.46, 95%CI [0.9-2.2], p = 0.06). Conclusions: In our study, the majority of patients received FFX with dose modifications. When FFX was delivered as per the ACCORD study, the outcomes were comparable and associated with a reduced incidence of grade 3+ haematological toxicities showing that the toxicity is manageable. Additional analyses are underway to delineate the impact on efficacy and safety of prognostic and predictive factors (e.g. patients' age, PS, G-CSF use, systematic vs. personalized dose adjustments, RDI). Accrual continues to reach the goal of 200 patients. Supported by sanofi-aventis Canada. Disclosure: J.A. Maroun: Advisory board and invited speaker by Sanofi; T.R. Asmis: Consultant/research funding : Sanofi; R. Goodwin: - financial support to attend conference - $1500 ad board honorarium; H. Marginean: - financial support to attend conference - $1500 ad board honorarium. All other authors have declared no conflicts of interest.