Gemcitabine Hydrochloride With or Without Vismodegib in Treating Patients With Recurrent or Metastatic Pancreatic Cancer

This randomized phase I/II trial studies gemcitabine hydrochloride and vismodegib to see how well they work compared with gemcitabine hydrochloride alone in treating patients with recurrent or metastatic pancreatic cancer. Drugs used in chemotherapy, such as gemcitabine hydrochloride, work in different ways to stop the growth of tumor cells, either by killing the cells or by stopping them from dividing. Vismodegib may slow the growth of tumor cells. It is not yet known whether giving gemcitabine hydrochloride together with vismodegib is more effective than gemcitabine hydrochloride alone in treating patients with pancreatic cancer.

PRIMARY OBJECTIVES: l. To compare the progression-free survival of advanced pancreatic cancer patients treated with the combination of gemcitabine plus GDC-0449 (vismodegib) versus gemcitabine plus placebo. SECONDARY OBJECTIVES: I. To compare overall survival of advanced pancreatic cancer patients treated with the combination of gemcitabine plus GDC-0449 versus gemcitabine plus placebo. II. To compare the objective response rate of advanced pancreatic cancer patients treated with the combination of gemcitabine plus GDC-0449 versus gemcitabine alone. III. To determine the toxicity experienced by pancreatic cancer patients treated with the combination of gemcitabine plus GDC-0449. IV. To determine the activity, in an exploratory analysis, of gemcitabine plus GDC-0449 in patients who progress on gemcitabine plus placebo. TERTIARY OBJECTIVES: I. To determine if tumor immunohistochemical expression patterns of proteins in the Hh pathway, including sonic hedgehog (Shh), indian hedgehog (Ihh), patched tumor suppressor gene (PTCH), smoothened protein (SMO), and GLI1 and 2, within pancreatic tissue obtained at the time of curative intent surgery predict response and prognosticate outcome of patients treated with or without GDC-0449 at the time of relapse. II. To determine the prognostic ability (relapse free survival, \[RFS\]) of these biologic markers for resected patients in an archival cohort of patients undergoing resection. III. To determine expression pattern of pancreatic CSC markers, including CD44, CD24, CD133, aldehyde dehydrogenase (ALDH) and epithelial specific antigen (ESA) by immunohistochemistry (IHC) on these archival tissues in relation to Hh pathway markers and correlate these with clinical outcomes. IV. To determine whether high baseline serum Shh, as well as changes in serum Shh during treatment, predict treatment efficacy and/or prognosticate clinical outcome. V. To determine the frequency of mutation of Hh pathway genes, PTCH, SMO, SuFU, and if the presence or absence of mutations are correlated with clinical outcome. VI. To determine the frequency of amplification of Hh pathway genes, gene copy number by quantitative polymerase chain reaction (qPCR) of GLI1 and SMO in those tumors that have high protein expression as seen by IHC. Gene amplification will be correlated with clinical outcome. VII. To determine if there is a correlation of K-ras mutation, and MET and RON expression, amplification, or mutation status with Hh pathway abnormalities, CSC markers, and clinical outcomes. VIII. To determine if baseline contrast perfusion imaging volume transfer constant (Ktrans) within primary and liver metastatic lesions as measured on a 256-detector computed tomography (CT) scanner predicts objective response rates, and other clinical endpoints including progression-free survival (PFS), to treatment with gemcitabine and GDC-0449/placebo. (University of Chicago ONLY) IX. To determine if treatment with Gemcitabine and GDC-0449 improves tumor perfusion, as measured by Ktrans, over the course of treatment by serial CT scans every 2 cycles, compared to tumors treated with Gemcitabine and placebo. (University of Chicago ONLY) X. To determine if improved tumor perfusion with GDC-0449 treatment (if observed) improves objective response rates and other clinical endpoints including PFS. (University of Chicago ONLY) OUTLINE: This is a multicenter, safety lead-in study (part I) followed by a randomized study (part II). An initial 6 patients are enrolled in the part I portion of the study. If no dose-limiting toxicities occur in these patients, subsequent patients are enrolled in the part II portion of the study. PART I (safety lead-in study): Patients receive gemcitabine hydrochloride intravenously (IV) over 30 minutes on days 1, 8, and 15 and hedgehog antagonist GDC-0449 orally (PO) once daily (QD) on days 1-28. Courses repeat every 28 days in the absence of disease progression or unacceptable toxicity. PART II (randomized study): Patients are stratified according to disease status (recurrent after surgery vs metastatic) and Eastern Cooperative Oncology Group (ECOG) performance status (0 vs 1). Patients are randomized to 1 of 2 treatment arms. ARM I: Patients receive gemcitabine hydrochloride 1000 mg/m\^2 IV over 30 minutes on days 1, 8, and 15 and placebo PO QD on days 1-28. Courses repeat every 28 days in the absence of disease progression or unacceptable toxicity. At the time of disease progression, patients are unblinded and may crossover to arm II. ARM II: Patients receive gemcitabine hydrochloride 1000 mg/m\^2 IV over 30 minutes on days 1, 8, and 15 and hedgehog antagonist GDC-0449 PO QD on days 1-28. Courses repeat every 28 days in the absence of disease progression or unacceptable toxicity. Tumor tissue, blood, serum, and plasma samples are collected periodically for biomarker and other analyses. After completion of study treatment, patients are followed periodically.