[PDF][PDF] Are metformin doses used in murine cancer models clinically relevant?

NS Chandel, D Avizonis, CR Reczek, SE Weinberg… - Cell metabolism, 2016 - cell.com
NS Chandel, D Avizonis, CR Reczek, SE Weinberg, S Menz, R Neuhaus, S Christian…
Cell metabolism, 2016cell.com
An important mechanism of action of metformin as an antidiabetic drug involves inhibition of
hepatic gluconeogenesis, but the molecular basis for this is controversial. A recent
perspective (He and Wondisford, 2015) argued that direct action of the drug on AMPK is of
key importance and occurs at clinically relevant concentrations of $70 μM. However, that
review cited conflicting studies, which concluded either that AMPK-independent
mechanisms are important in metformin action, or that AMPK activation plays a role but is …
An important mechanism of action of metformin as an antidiabetic drug involves inhibition of hepatic gluconeogenesis, but the molecular basis for this is controversial. A recent perspective (He and Wondisford, 2015) argued that direct action of the drug on AMPK is of key importance and occurs at clinically relevant concentrations of $70 μM. However, that review cited conflicting studies, which concluded either that AMPK-independent mechanisms are important in metformin action, or that AMPK activation plays a role but is secondary to energetic stress resulting from inhibition of oxidative phosphorylation by the drug. As work to define the molecular mechanisms of metformin in diabetes has proceeded, the hypothesis that biguanides may be ‘‘repurposed’’for additional applications has received increased attention. This followed retrospective pharmacoepidemiologic studies that showed reduced cancer risk and/or improved cancer prognosis among diabetics treated with metformin relative to those treated with other drugs. Although some of these studies have been criticized on methodologic grounds, they nevertheless led to laboratory studies that revealed clear antineoplastic activity and provided a rationale for clinical trials (Buzzai et al., 2007). More than 100 trials designed to detect clinical utility of metformin in oncology are now ongoing. There is also experimental evidence that metformin extends lifespan in model organisms, sparking interest in the possibility that these studies have clinical relevance. Not surprisingly in view of controversies regarding the molecular mechanisms of metformin action in diabetes, there is uncertainty regarding the mechanisms relevant to its putative antineoplastic and antiaging effects, and it cannot be assumed that these are identical to those involved in suppression of gluconeogenesis in diabetes (Foretz et al., 2014). One possibility is that the drug acts indirectly by altering the host endocrine milieu in a manner that inhibits the growth of a subset of cancers, for example, by lowering insulin levels. While this is possible, the magnitude of such changes is small and may or may not be sufficient to alter neoplastic behavior. Another possibility relates to direct actions of biguanides on cancer cells or cells at risk for transformation, for example, by inhibiting oxidative phosphorylation (Wheaton et al., 2014). However, it is unclear if clinically practical metformin doses are sufficient for such mechanisms to operate. Most in vitro assays reveal that millimolar (mM) levels of metformin are required to have antiproliferative effects. At first glance, this implies that direct clinical antineoplastic activity is unlikely, as diabetic patients receiving typical doses of 1.5–2.5 g of metformin per day ($30 mg/kg) have plasma levels in the 10 μM range (Graham et al., 2011). However, sensitivity to metformin is known to vary greatly with nutrient availability, and in vitro culture conditions are unlikely to mimic the in vivo microenvironment in this respect, as glucose, serine, and glutamine, each of which reduces metformin sensitivity, are likely present in media at higher concentration than in poorly vascularized tumors (Gravel et al., 2014). Thus, we suggest that the requirement for millimolar metformin concentrations in vitro may not be useful for predicting the drug concentration required for activity in vivo or clinically. Nevertheless, it is important to determine the in vivo concentration of metformin in mouse models that demonstrates antineoplastic activity in order to determine if this level is achievable in patients.
We administered metformin to C57BL/6J mice at a dose of 250 mg/kg in drinking water for 2 weeks and used mass spectrometry to measure metformin …
cell.com