Inclusion Criteria:

- Metastatic pancreatic ductal adenocarcinoma (i.e., Stage 4).

- Histological/cytological confirmation of metastatic tissue (not primary tumor) by a
central pathology laboratory (H&E stain) to ensure sufficient material is available
for later hENT1 analysis.

- Adjuvant chemotherapy/radiotherapy ≥ 6 months prior to randomization.

- Palliative radiotherapy (if administered) ≥ 1 month prior to randomization.

- CT scan ≤30 days prior to randomization

- Performance Status (ECOG) 0 or 1.

- Estimated life expectancy ≥ 12 weeks.

- Age ≥ 18 years.

- Adequate hematological and biological function.

- Written consent on an IRB/IEC-approved Informed Consent Form prior to any
study-specific evaluation.

Exclusion Criteria:

- Prior palliative chemotherapy for pancreatic cancer.

- Radical pancreatic resections (e.g., Whipple procedure) are not allowed < 6 months
prior to randomization. Exploratory laparotomy, palliative (e.g., bypass) surgery,
or other procedures (e.g., stents) are not allowed < 14 days prior to randomization.
In both cases the patient must be sufficiently recovered and stable.

- Symptomatic brain metastases.

- Participation in other investigational drug clinical studies ≤ 30 days prior to
randomization.

- Concomitant treatment with prohibited medications.

- History of allergy to gemcitabine or eggs.

- Presence of any serious or unstable concomitant systemic disorder incompatible with
the clinical study (e.g., substance abuse, uncontrolled intercurrent illness
including active infection, arterial thrombosis, symptomatic pulmonary embolism).

- Any disorder that would hamper protocol compliance.

- Prior nonpancreatic malignancy treated with chemotherapy. Prior malignancies treated
with surgery or radiotherapy alone must be in remission ≥ 3 years. The following
prior malignancies are allowable irrespective of when they occurred: in situ
carcinoma of the cervix, in situ ductal breast cancer, low-grade local bladder
cancer, and nonmelanotic skin cancer.

- Females who are pregnant or breastfeeding.

- Refusal to use adequate contraception for fertile patients (females and males during
the study and for 6 months after the last study treatment). Adequate forms of
contraception are double-barrier methods (condoms or diaphragm with spermicidal jelly
or foam); oral, depot, or injectable contraceptives; intrauterine devices; tubal
ligation.

- Any other reason the investigator considers the patient should not participate in the
study.