A Multicenter, Open-Label, Randomized, Phase II Trial of Adjuvant Dasatinib Plus Gemcitabine Versus Single-Agent Gemcitabine in Patients With Resected Pancreatic Adenocarcinoma

Inclusion Criteria:

- Written informed consent before beginning any protocol specified procedures

- Histologically proven pancreatic adenocarcinoma

- Any T, any N, M0 disease that has had all gross disease resected (R0 or R1 resection)

- ECOG Performance status index 0 or 1

- Absolute Neutrophils >= 1.5 x 10^9/L

- Platelets >= 100 x 10^9/L

- Hemoglobin >= 10 g/dL

- Total bilirubin =< 2.0 x UNL; subjects with Gilbert's syndrome, confirmed by
genotyping or invader UGTIA1 molecular assay before study entry must have total
bilirubin < 3 x UNL

- ASAT (SGOT) and ALAT (SGPT) =< 2.5 x UNL

- Alkaline Phosphatase =< 5 x UNL

- Creatinine < 1.5 x UNL

- Serum Na, K+, Magnesium, Phosphate and Calcium >= LNL

- First study treatment must be given within 60 days after surgery and within 7 days
after randomization

- Patients must be accessible for treatment and follow-up and compliant with study
procedures

- Negative pregnancy test (urine or serum) within 7 days before first study treatment
for all women of childbearing potential, whom also must implement adequate
non-hormonal contraceptive measures during study treatment and for at least 3 months
after the last dose of study therapy

- Ability to take oral medication (dasatinib must be swallowed whole)

Exclusion Criteria:

- Prior or concurrent systemic anticancer therapy (immunotherapy, hormonal therapy,
biological therapy, or chemotherapy) for pancreatic cancer

- Prior or concurrent radiation therapy for pancreatic cancer

- Pregnant or lactating patients

- M1 pancreatic cancer

- Concurrent congestive heart failure, unstable angina pectoris, or M1 within the 6
months before first study treatment

- Uncontrolled hypertension or high-risk uncontrolled arrhythmias

- Any history of clinically significant ventricular arrhythmias (such as ventricular
tachycardia, ventricular fibrillation, or torsades de pointes)

- Diagnosed or suspected congenital long QT syndrome

- Prolonged QTc interval on pre-entry electrocardiogram (> 450 msec)

- History of significant neurologic or psychiatric disorders including psychotic
disorders, dementia or seizures that would prohibit the understanding and giving of
informed consent

- Past or current history of neoplasm other than pancreatic adenocarcinoma, except for:
curatively treated non-melanoma skin cancer; in situ carcinoma of the cervix; other
cancer curatively treated and with no evidences of disease for at least 1 year

- Concurrent treatment with other experimental drugs or treatment with investigational
drugs within 30 days of first study treatment

- Currently receiving drugs with known significant CYP 3A4 inhibitory effects (such as
ketoconazole, itraconazole, troleandomycin, erythromycin, diltiazem, verapamil,
ritonavir, indinavir)

- Concurrent administration with inducers of CYP 3A4 may result in a lower exposure to
dasatinib and are therefore not allowed (e.g., phenytoin, carbamazepine, rifampicin,
phenobarbital, pentobarbital, or St John's Wort)

- Known allergy reactions to dasatinib or gemcitabine or excipients used in the study

- History of significant bleeding disorders unrelated to cancer, including: diagnosed
congenital bleeding disorders (e.g., Von Willebrand's disease); diagnosed acquired
bleeding disorder within 1 year (e.g., acquired anti-factor VIII antibodies); ongoing
or recent (=< 3 months) significant gastrointestinal bleeding

- Patients currently taking drugs that are generally accepted to have a risk of causing
Torsades De Pointes including: quinidine, procainamide, disopyramide; amiodarone,
sotalol, ibutilide, dofetilide; erythromycins, clarithromycin; chlorpromazine,
haloperidol, mesoridazine, thioridazine, pimozide; cisapride, bepridil, droperidol,
methadone, arsenic, chloroquine, domperidone, halofantrine, levomethadyl,
pentamidine, sparfloxacin, lidoflazine

- Concurrent treatment with intravenous bisphosphonates; prior treatment should be
stopped at least 2 weeks before first dose of study treatment

- Concurrent medical condition which may increase the risk of toxicity, including
pleural or pericardial effusion or any grade

- Active uncontrolled infection requiring parenteral antimicrobials