Metformin and biomarkers relevant to neoplasia in nondiabetic patients.

Abstract

e14615 Background: Use of the anti-diabetic drug Metformin has been associated with reduced risk or improved prognosis of several cancers in both lab models and retrospective population studies of diabetics. Plausible mechanisms of action include (1) reduction in serum levels of insulin or inflammatory cytokines and (2) direct effects on target cells due to action of the drug as an inhibitor of mitochondrial respiration, including activation of AMPK. Methods: Non-diabetic subjects (n = 47) consented to bloodwork and sigmoid biopsy at screening colonoscopy, metformin administration (500 mg TID) for 10 weeks, and end-of-study sigmoid biopsy and bloodwork. Serum assays were performed on both fasting and post standard oral glucose load specimens. Metformin levels were measured by mass spectroscopy. Anti-phospho-histone-3 immunostaining was used to quantify proliferation. Results: Metformin reduced fasting levels of c-peptide, insulin, and IL-6 levels by < 10%, a magnitude of change that laboratory studies suggest is unlikely to be biologically significant. However, postprandial insulin levels following metformin were reduced by 18 % relative to baseline postprandial levels, a finding of potential physiologic relevance. Serum metformin level showed considerable variation between subjects (1005 ± 769 ng/ml) and this variation was not correlated with effects of metformin on epithelial proliferation. Proliferation index fell from 5.9 to 3.3, (p<0.03), a change greater than predicted by in vitro studies of direct action of metformin at the concentrations measured in serum. Conclusions: Metformin lowers insulin levels even in non-diabetic subjects, and post prandial insulin levels are more sensitive indicators of metformin action than fasting biomarkers. As high insulin level has been linked to increased cancer risk and/or worse prognosis (eg J Natl Cancer Inst. 2004;96:546), this systemic action suggests clinical applications. However, direct exposure of colonic epithelial cells to high luminal concentrations of metformin may also be relevant and is one of several factors that may explain the greater inhibitory effect of the drug on colonic epithelium proliferation than that previously reported for breast or prostate.