Study design and setting
This study is designed as a multi-center, double-blind, placebo-control, randomized controlled trial to be performed in nondiabetic patients with a recent history of undergoing colorectal polypectomy. The study will take place at the Division of Gastroenterology, Yokohama City University Hospital, and its 5 affiliate hospitals. The coordinating office shall be at Yokohama City University Hospital, with the registration, randomized allocation and data collection to be conducted at this site.
Ethical considerations and registration
The study protocol is in compliance with the Declaration of Helsinki [23
] and the Ethics Guidelines for Clinical Research published by the Ministry of Health, Labour, and Welfare, Japan [24
]. We obtained approval for this study from the Ethics committee of Yokohama City University Hospital on July 8th
2011. The protocol and informed consent forms were approved by the institutional ethics committee at each of the participating institutions. This trial has been registered in the University hospital Medical Information Network (UMIN) Clinical Trials Registry as UMIN000006254. Written informed consent for participation in the study will be obtained from all the participating patients. The trial results will be reported in conformity with the Consolidated Standards of Reporting Trials (CONSORT) 2010 guidelines [25
All adult patients visiting the hospital for polypectomy will be recruited for the study.
The inclusion criteria are as follows:
1) No colorectal polyps present after the polypectomy
2) Age 40 to 80 years as on the date of informed consent
3) Willingness to provide written informed consent
The likelihood of development of colorectal polyps in the young is low and the diagnosis history of polyps in the young is usually related to familial adenomatous polyposis or hereditary non-polyposis colorectal cancer; on the other hand, the elderly have various complications. This was the rationale for our setting the age criterion for inclusion in this study as 40 to 80 years.
The exclusion criteria are as follows:
1) History of diabetes mellitus (use of medication and/or HbA1c over 6.5%)
2) History of regular use (defined as at least once per week) of NSAIDs and/or aspirin
3) History of bowel surgery
4) History of malignant disease (excluding carcinoma in adenoma, carcinoma in situ that has already been resected)
5) History of heart failure, renal failure, liver cirrhosis or chronic hepatic failure
6) History of familial adenomatous polyposis
7) History of hereditary non-polyposis colorectal cancer
8) History of inflammatory bowel disease
9) Pregnancy or possibility of pregnancy
10) Patients judged as inappropriate candidates for the trial by the investigators
All eligible patients will be allocated randomly to one of two groups, the metformin group and the placebo group. Endoscopists, doctors at the follow-up outpatient clinics and patients will be blinded to the allocation. Patients in the metformin group shall receive oral metformin at 250 mg per day, and those in the placebo group shall receive oral placebo tablet. At the end of 1 year of administration of metformin/placebo, colonoscopy will be performed to evaluate the polyp formation.
The primary endpoint shall be the prevalence of colorectal polyps and number of polyps after 1-year's intervention. The endoscopic examinations and polypectomies will be performed using Olympus colonoscopes (model H260AZI). The day before the endoscopy, each patient will be instructed to consume a low-residue diet and shall receive 5 mg of oral sodium picosulfate. On the day of the endoscopy, the patients shall receive 2000 ml of polyethylene glycol (PEG). If the feces are not sufficiently clear, an additional 1000 - 2000 ml of PEG may be given to ensure sufficient bowel cleaning. At the time of the polypectomies, the endoscope shall be inserted into the cecum, and the entire colorectum will be carefully observed as the endoscope is pulled back. If any polyps are detected, polypectomy will be performed. At the end of 1 year of administration of metformin/placebo, the same endoscopists will perform the repeat endoscopic examinations. If a polyp(s) is detected at the repeat colonoscopy (after treatment for one year), a biopsy will be performed. A total of 6 endoscopists from Yokohama City University Hospital and the 5 affiliate hospitals will perform the polypectomies and endoscopic examinations. All procedures will be recorded on DVD, and all the polyps will be photographed. The number of polyps in each patient will first be counted by the operators during the performance of the colonoscopy. To further ensure validity, the number of polyps will be counted again through observation of the recorded DVD by 3 blinded expert endoscopists (H.T, H.E, and E.S). If these expert endoscopists judge the colonoscopic examination as having been inadequate in any case, that case will be excluded. The biopsied polyps will be evaluated by expert pathologists (Y.N and S.Y).
The secondary outcomes are (1) the drug safety; adverse events will be monitored by the doctor at every follow-up visit to the outpatient clinic. Adverse events will be graded according to the National Cancer Institute Common Toxicity Criteria for Adverse Events (NCI-CTCAE), version 4.0. Diarrhea, the most frequent adverse event related to use of metformin, will be watched for spontaneous resolution within a few days and/or managed by antiflatulent drugs. If Grade 3 or more severe adverse events appear, the follow-up doctor shall report it (them) to the coordinating office and the case will be withdrawn from the study at that point; (2) laboratory data (fasting blood glucose, fasting blood insulin, HbA1c, total cholesterol, LDL-cholesterol, blood urea nitrogen (BUN), creatinine); (3) the number of ACF. At the time of the 1-year colonoscopy and polypectomy, the lower rectal region from the middle Houston valve to the dentate line will be washed thoroughly with water, sprayed with 0.25% methylene blue, which would be left to stand for 2 min, then again washed thoroughly with water. The number of rectal ACF will be counted with a magnifying endoscope [21
]; (4) physical examination findings (body weight, body mass index (BMI)). Metformin is widely used as an antidiabetic drug that improves insulin resistance. The effect of metformin on insulin resistance and the plasma lipid profile will be evaluated by comparing these parameters measured at the baseline and at 1 year in the metformin group and placebo group. All participants will receive physical examination and laboratory tests at the time of the 1-year endoscopic examination and polypectomy; (5) effects of metformin on the cell-proliferative and apoptotic activities in the rectal epithelium and polyps (if any). Colonic epithelial samples will be obtained from the same trial patients by biopsy at the time of the 1-year colonoscopy and polypectomy. The cell-proliferative activity will be evaluated by staining for the proliferative cell nuclear antigen (PCNA), and the cell-apoptotic activity by the terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling (TUNEL) method; (6) expression levels of protein (AMPK, mTOR/S6K) in the rectal epithelium and polyps (if any) that are thought to be pharmacological targets of metformin. The expression levels of these proteins shall be determined by western blot analysis.
The investigator shall convey the patient's details to the central registration centre via fax. After an eligibility check, the patients will be randomly assigned to receive metformin or placebo at the central registration centre by a computer program, using a minimization method, with stratification by institute, age, gender and BMI. In this way, the patient assignment will be concealed from the investigator. The randomization center will allocate a numbered treatment pack to each patient, which will contain all the drugs or placebos needed to complete a course of the trial treatment for one patient.
Metformin will be purchased from Dainippon Sumitomo Parma Co., Ltd. The placebo (250 mg lactose) will be purchased from Kokando Co., Ltd, Toyama, Japan. All trial drugs will be packaged identically and identified only by number. Subjects will be instructed to take one tablet of the trial drug after breakfast every day, and to visit the hospital every 4 weeks for evaluation of the subjective symptoms and for receiving a new supply of medication. Compliance will be monitored by counting the empty drug packages returned by the patients at every visit to the outpatient clinic.
Sample size estimation
We previously showed that metformin administered at 250 mg/d for 1 month directly suppressed the formation of ACF. In that study, mean number of ACF per patient decreased significantly from 8.78 ± 6.45 (baseline) to 5.11 ± 4.99 (at 1 month, p = 0.007
]. Based on a similar study, Takayama et al. reported that sulindac administration at 300 mg/d for 2 months to post-polypectomy patients suppressed ACF formation, decreasing the number of ACF from 7.70 ± 4.04 (baseline) to 4.00 ± 2.95 (at 2 months, p < 0.001
]. From these reports, we estimated that metformin and sulindac may have equivalent effect on suppression of ACF formation. Moreover, from the same study, Takayama et al. reported that in the post-polypectmoy patients who received 2-months' intervention, the number of polyps (hyperplastic polyp and adenoma) at 1 year after the treatment was significantly lower in the sulindac group in comparison with that in the placebo group (26/48 (54.2%) vs. 15/48 (31.3%), p = 0.025
]. Therefore, we speculated that 1-year's treatment with metformin might yield equivalent suppression of metachronous polyp formation to that with sulindac. To detect the reduction in the number of metachronous polyps in the metformin group using the chi-square test with a two-sided significance level of 5% and a power of 80%, it was found that a sample size of 68 patients per group would be necessary. Assuming a 10% dropout rate, we propose to recruit 75 patients per group, that is, a total of 150 patients.
The prevalence of polyps in each group, the primary endpoint, will be compared between the metformin group and placebo group by the chi-square test. The safety, one of the secondary endpoints, will be similarly compared by the chi-square test. The remaining results in the two groups will be compared by the Mann-Whitney U test or Student's t test. A P value of < 0.05 shall be regarded as indicative of significance. The analysis will be performed using SPSS, version 17.0 (SPSS Inc., Chicago, Il.).
Trial steering committee and data monitoring committee
The Trial Steering Committee and Data Monitoring Committee shall be located at the Department of Clinical Research, Yokohama City University School of Medicine. The committee shall consist of three people: Yutaka Natsumeda, M.D., Satoshi Inoue, M.D., and Yukiharu Yamaguchi, Ph.D. The Management Team will monitor the trial progress status and data by face-to-face and/or telephonic contact with each of the six sites every month.
A flow chart of the study is shown in Figure .