Background: The ACCORD 11 and the MPACT trials recently showed superiority of FIO and NG over gemcitabine alone, respectively. However, both trials had strict inclusion criteria. The aim of this study was to determine the proportion of patients (pts) with MPA who would be potentially eligible for first-line palliative intent chemotherapy with FIO and NG in routine clinical practice. Methods: 473 consecutive pts who presented with MPA and initiated palliative chemotherapy with gemcitabine from 2000 to 2011 at the BC Cancer Agency were identified using the provincial pharmacy database. Clinicopathological variables and treatment outcomes were retrospectively collected and compared among groups. Eligibility criteria for each regimen were in accordance with the criteria as described in the respective pivotal phase III trials. Results: In total, median age was 66 years (range 34–89) and 258 (55%) were men. Only 24.7% of the pts would be eligible for FIO as compared to 45.2% for NG. The main reasons for ineligibility for FIO were ECOG performance status (PS) ≥ 2 (56.5%), age > 76 years (19.0%), and bilirubin > 1.5 times the upper limit of the normal range (ULN) (18.6%). The main reasons for ineligibility for NG were bilirubin > ULN (24.5%), ECOG PS ≥ 3 (14.6%), and cardiac dysfunction (13.8%). Median overall survival (OS) for the entire cohort treated with gemcitabine was 5.8 months (95% CI 5.4-6.2). On univariate analyses, eligible pts for FIO had longer median OS than ineligible pts (8.6 vs 4.7 months, p<0.001). Eligible pts for NG also had longer median OS than those deemed ineligible (6.7 vs 4.9 months, p=0.008). After accounting for ECOG PS in the multivariate model, eligibility for either FIO or NG no longered predicted for better OS. Conclusions: In ourpopulation-based analysis, almost twice as many pts would be eligible for NG when compared to FIO, mostly due to ECOG PS. The longer OS observed in the FIO-eligible population likely reflects the exclusion of ECOG PS 2 pts.