This is a double-blinded placebo controlled randomized trial examining the effect of metformin in non-diabetic patients with colon cancer on cell growth, immunological and metabolic changes. Patients are randomized to receive metformin 20 days before and 10 days after surgery. Tumor samples are examined for changes in level of cell growth and the composition of tumor cells in the tumor is examined. Blood samples are assessed for immunological markers and insulin resistance is measured. Cell proliferation, migration and adhesion are also examined in vitro by adding plasma obtained from the patients to colon cancer cell lines grown in culture.

Background: Colorectal cancer (CRC) is the third most common cancer worldwide and more than 5000 patients are diagnosed each year in Denmark.

Metformin is the drug of choice for treatment of type 2 diabetes. Several studies indicate that the incidence of colorectal cancer is lower among metformin treated diabetic patients than other diabetic patients and survival after CRC is improved for this group as well.

Metformin lowers plasma glucose in diabetic patients, but studies suggest that metformin also inhibits cancer cell growth.

Tumor cell proliferation and apoptosis can be estimated by determining the expression levels of specific cell cycle related proteins such as Ki67 (proliferation) and cleaved caspase-3 (apoptosis) using immunohistochemistry.

The level of cell proliferation and apoptosis is important for tumor development, but growing evidence suggests that the microenvironment of the tumor and the patient's immune response play important roles as well. The immunoscore has been introduced as a prognostic marker for CRC. The immunoscore is determined by staining whole tumor slides for CD3 and CD8 positive lymphocytes using immunohistochemistry, followed by quantitative assessment and scoring of their densities. A high density is associated with better outcome than a low density. It is possible that metformin can influence the composition of immune cells as well.

Surgery is known to induce a surgical stress response with hormonal and metabolic changes. The stress response leads to an increased insulin resistance and hyperglycemia postoperatively. The degree of insulin resistance and hyperglycemia is correlated with risk of postoperative complications, reoperation, length of stay and death.

Study design: The trial is a randomised, placebo-controlled, double-blinded trial investigating the effect of metformin (intervention group) against placebo (control group) on cell proliferation, metabolic and immunological changes in non-diabetic patients with colon cancer.

Patients are recruited at their visit to the out-patient clinic at Slagelse Hospital when surgery is planned. Patients, who agree to participate, will be randomized to receive metformin or placebo for up to 20 days before their operation and 10 days afterwards. Blood samples will be taken at time of randomization and 4 times more during the study.

The study is divided into 5 sub-studies:

Sub-study 1 - Cell growth on tumor level: The primary outcome is determination of the difference of the level of proliferation after the intervention (time of surgery) adjusted for the level seen at baseline (time of colonoscopy). This is examined by immunohistochemical staining of tumor samples obtained from the colonoscopy and after surgery for Ki67 and cleaved caspase 3.

Sub-study 2 - immunological changes: The immunoscore is measured by immunohistochemical staining of tumor samples for CD3 and CD8 positive lymphocytes. Blood samples are analyzed for immune markers.

Sub-study 3 - cell growth in vitro: Plasma obtained from the metformin- or placebo-treated patients are added to colorectal cancer cells grown in culture. Cell proliferation, migration and adhesion are determined by in vitro studies and differences between the two groups analyzed.

Sub-study 4 - insulin resistance and recovery: Insulin resistance before and after surgery is measured using the homeostatic assessment model (HOMA) from fasting levels of glucose and insulin. Capillary glucose level is measured 4 times a day postoperatively on postoperative day 1 and 2 - before the main meals and before sleeping.

Patient perceived quality of recovery is measured using the Danish version of the validated "Quality of recovery-15" questionnaire.

Sub-study 5 - microbiota: The microbiota of fecal samples will be measured before and after metformin treatment.

Inclusion Criteria:

• Patients with adenocarcinoma of the colon planned for elective curative intended surgery at Slagelse hospital
• Age of 18 or above
• Must be able to understand and sign informed content
• Sufficient amount of representative tumor material from the biopsies taken at the initial colonoscopy must be present

Exclusion Criteria:

• Patients diagnosed with diabetes mellitus
• Patients who are receiving or have received metformin or other oral antidiabetics
• Impaired kidney function (eGFR < 60mL/min)
• Severe liver disease (defined as transaminases above X 3 normal levels)
• Participation in another pharmacological intervention trial
• Predictable poor compliance (for instance not speaking fluent Danish, mentally impaired)
• Presenting with metastatic disease
• Patients undergoing neoadjuvant chemotherapy
• Pregnancy or lactation (fertile women must have a negative serum or urine pregnancy test to participate)
• Fertile women who do not use safe contraception during the study period.
• Allergy to metformin or placebo