Effect of metformin on breast cancer metabolism

There is a great and growing interest in the anti-cancer effects of the diabetes drug metformin as evidenced by the 50 clinical trials worldwide investigating the role of metformin in cancer treatment that are actively recruiting patients. Metformin’s potential anti-cancer effects were first brought to light following a series of epidemiological studies showing that treatment with metformin, as opposed to other diabetes drugs, correlated with a reduction in the incidence of cancer in these patients. A subsequent retrospective clinical study of diabetic patients has suggested potential synergy with chemotherapy and several published in vitro and in vivo studies have clearly demonstrated evidence of an anti-cancer effect. However there is still not a consensus as to the mode of action by which metformin achieves this. Many in vitro studies of both cancer and non-cancer cells have shown that metformin can inhibit complex 1 of the mitochondrial respiratory transport chain and subsequently activate the energy sensing kinase, AMPK. The downstream catabolic effects subsequent to AMPK activation would be expected to be antiproliferative in nature in the cancer setting. However, it is also postulated that metformin derives its anticancer effect instead from downregulation of the 'host's' insulin axis, an indirect effect. Hence, there were 3 aims to this thesis project: to gain greater insight into its anticancer mechanism of action, to study pathways that might be targeted to synergise with metformin and to set up a 'window of opportunity' clinical study to develop potential biomarkers of its anticancer and metabolic effect and further investigate its mechanism of action in vivo. This thesis describes the novel finding that in vitro and in vivo metformin treatment results in the marked accumulation of fatty acid triglyceride in cancer cells and that this phenomenon is secondary to the inhibition of fatty acid oxidation. It is also shown that etomoxir, another inhibitor of fatty acid oxidation, has an additive effect with metformin both in terms of inhibition of fatty acid oxidation, triglyceride fatty acid accumulation and the antiproliferative effect. In vitro work also showed the importance of glutamine in maintaining cancer cell growth for metformin treated cells and evidence of marked synergy with several compounds that inhibit the steps of the glutaminolysis and reductive carboxylation pathways, and also glutamine uptake. Lastly, this thesis describes a clinical study in which 2 weeks of metformin treatment was given to women with early breast cancer prior to any other treatment being initiated. Pharmacodynamic assays were carried out pre- and post-metformin, the study tests including breast core biopsies assayed for changes in gene expression using next generation sequencing, immunohistochemistry and metabolomics. Dynamic 18[F]-FDG PET-CT scans were also carried out along with measurement of 'host' serum metabolic markers. Preliminary gene sequencing data from the clinical study demonstrates striking changes in the expression of genes encoding for the units of the mitochondrial electron transport chain (including complex 1) and the regulatory enzymes of glycolysis, glutaminolysis and fatty acid oxidation consistent with a direct metabolic effect on breast cancer cells by metformin. However, insulin secretion was also lowered by metformin in this cohort of non-diabetic breast cancer patients, suggesting that downregulation of the host's insulin axis may also play a role.