This randomized phase II trial studies how well metformin hydrochloride works in preventing esophageal cancer in patients with Barrett esophagus. Chemoprevention is the use of certain drugs to keep cancer from forming. The use of metformin hydrochloride may keep esophageal cancer from forming.

PRIMARY OBJECTIVES:

I. To compare the percent change in the mean pS6K1 immunostaining from baseline in mucosal Barrett esophagus (BE) biopsies among patients assigned to 2,000 mg metformin hydrochloride once daily (QD) versus placebo as determined from Barrett mucosal biopsy samples obtained pre- and post-intervention.

SECONDARY OBJECTIVES:

I. To evaluate adverse events associated with the two intervention arms.

TERTIARY OBJECTIVES:

I. To assess the effects of metformin hydrochloride 2,000 mg QD versus placebo on the changes in pS6K1 using traditional IHC categories.

II. To assess the effects of metformin hydrochloride 2,000 mg QD versus placebo on absolute change in pS6K1.

III. To assess changes in serum markers (metformin hydrochloride, fasting insulin, HOMA-IR, IGF-1, IGF-2, IGFBP-1, IGFBP-3, fasting leptin, and fasting adiponectin) as determined from serum samples obtained pre- and post-intervention.

IV. To assess changes in proliferation (Ki-67) and apoptosis (cleaved caspase 3) as determined from Barrett mucosal biopsy samples obtained pre- and post-intervention.

V. To assess changes in molecular mediators of the insulin pathway (p-IRS-1, p-AKT^Serine 473) as determined from Barrett mucosal biopsy samples obtained pre- and post-intervention.

VI. To assess changes in relative activity of AMPK (phosphorylated AMPK/total AMPK ratio) and molecular mediators of AMP kinase (p-mTOR, pS6K1^Serine 235) as determined from Barrett mucosal biopsy samples obtained pre- and post-intervention.

VII. To assess changes in Programmed Cell Death 4 expression and miR-21 as determined from Barrett mucosal biopsy samples pre- and post-intervention.

VIII. To establish a biospecimen repository archive for future correlative studies.

OUTLINE: This is a multicenter study.

Patients are stratified according to nonsteroidal anti-inflammatory drugs use (regular vs no regular), body mass index (≥ 30 kg/m² vs < 30 kg/m²), gender (male vs female), and length of Barrett (2.00 to 4.99 cm vs ≥ 5.00 cm). Patients are randomized to 1 of 2 treatment arms.

Arm I: Patients receive extended-release metformin hydrochloride orally (PO) once daily (QD) on week 1, and twice daily (BID) on weeks 2-12 (every morning [QAM] and every evening [QPM] on week 3) in the absence of unacceptable toxicity or disease progression.

Arm II: Patients receive extended-release placebo PO QD on week 1 and BID on weeks 2-12 (QAM and QPM on week 3) in the absence of unacceptable toxicity or disease progression.

Blood, tissue, and mucosal tissue samples are collected at baseline and after completion of study treatment for pS6K1 analysis and other serum, mucosal, and molecular markers studies by IHC, ELISA, western blotting, and high-performance liquid chromatography (HPLC) methods.

After completion of study treatment, patients are followed up for 30 days.

Inclusion Criteria:

• Histologically confirmed diagnosis of Barrett esophagus, with no dysplasia, indeterminate for dysplasia, or low-grade dysplasia as defined by the presence of specialized columnar epithelium on histology and >= 2 cm of involvement on endoscopy

• Adequate Barrett mucosa, which is defined as >= 1 out of 4 research samples (i.e., >= 25%) with >= 50% intestinal metaplasia in biopsies required to satisfy the endpoints of the study

• No history of esophageal carcinoma or other cancer(s) (except for non-melanoma skin cancers)

• No erosive esophagitis or ulcerative esophagitis, unless treatment with a proton pump inhibitor (PPI) results in healed erosions or ulcers prior to entry endoscopy

• No history of high-grade dysplasia or cancer (confirmed locally by esophagogastroduodenoscopy [EGD] and Pathology reports)

  • No ulcer, plaque, nodule, stricture, or other luminal irregularity within the Barrett segment, unless clinical biopsy produces no evidence of high-grade dysplasia or cancer

• ECOG performance status =< 1

• Hemoglobin >= 10 g/dL

• Leukocytes >= 3,000/mL (>= 2,500/mL for African-American participants)

• Absolute neutrophil count >= 1,500/mL (>= 1,000/mL for African-American participants)

• Platelets >= 100,000/mL

• Total bilirubin =< institutional upper limit of normal (ULN)

• AST (SGOT) and ALT (SGPT) =< 1.5 times institutional ULN

• Creatinine =< institutional ULN

• Willingness to provide tissue samples for research purposes

• No contraindication to esophagogastroduodenoscopy (EGD)

• Willingness, for both men and women, to use adequate contraception (hormonal or barrier method of birth control; surgical intervention; abstinence) prior to study entry and for the duration of study participation

• A negative (serum or urine) pregnancy test done =< 7 days prior to Pre-Registration, for women of childbearing potential only

• No pregnant or nursing women

• No participants with diabetes mellitus

• No history of vitamin B12 deficiency or megaloblastic anemia

• No history of lactic acidosis

• No diseases associated with weight loss: anorexia, bulimia, or nausea

• No history of allergic reactions attributed to compounds of similar chemical or biologic composition to metformin

• No participants with HIV, cirrhosis of any cause, NASH (non-alcoholic steatohepatitis), or hepatitis (auto-immune or infectious)

  • For participants diagnosed with any other hepatic impairment, consult with protocol principal investigator (PI)

• No metabolic acidosis, acute or chronic, including ketoacidosis

• No uncontrolled intercurrent illness including, but not limited to, ongoing or active infection, symptomatic congestive heart failure, unstable angina pectoris, cardiac arrhythmia, or psychiatric illness/social situations that would limit compliance with study requirements; this includes significant medical conditions including renal failure, hepatic failure, sepsis, and hypoxia

• No genetics disorders such as family history of hereditary gastrointestinal polyp disorder (e.g., familial adenomatous polyposis [FAP], hereditary non-polyposis colorectal cancer [HNPCC], Peutz-Jegher disease)

• No chronic alcohol use or a history of alcohol abuse (defined as ingestion of >= 3 drinks per day)

• No kidney disease or renal insufficiency (defined as serum creatinine outside the normal institutional limits)

• Currently on a proton pump inhibitor (PPI) >= 4 weeks (any PPI taken at least once daily is acceptable)

• No medication(s) for weight loss ≤ 2 months prior to Pre-Registration

• No treatment with medications that may increase metformin hydrochloride levels: cationic drugs, e.g., digoxin, amiloride, procainamide, ranitidine, trimethoprim, quinidine, quinine, vancomycin, triamterene, and morphine

• No treatment with other oral hypoglycemic agents

• No participant use of metformin, cimetidine (Tagamet), furosemide (Lasix), or nifedipine (Cardizem), or any other drug contraindicated for use with metformin

• No receipt of any other investigational agents =< 3 months prior to Pre-Registration, except innocuous agents with no known interaction with the study agent (e.g., standard dose multivitamins or topical agents for limited skin conditions), at the discretion of the Protocol Lead Investigator at each Participating Site

• No participants who have undergone ablation or other local therapies (e.g., percutaneous dilatational tracheostomy [PDT], cryotherapy, radiofrequency, argon plasma coagulation [APC], or multipolar electrocoagulation [MPEC])

  • Patients treated with endoscopic mucosal resection [EMR] allowed

• No participants anticipating elective surgery during the study period

• No participants planning to undergo elective radiologic studies involving intravascular administration of iodinated contrast materials