A new study is providing even more evidence that a widely-used anti-diabetes drug may also be effective in fighting breast cancer.

The new study from Harvard Medical School found that combining metformin, sold as Glucophage, with the chemotherapy drug doxorubicin, reduced breast cancer tumours faster than doxorubicin alone, when tested on mice.

What's more, the drug combo also prolonged remission in the mice longer than chemotherapy alone.

The researchers, writing in the journal Cancer Research, say the metformin combo seems to work by targeting the cancer's stem cells and may improve breast cancer outcomes in people.

Evidence is growing that part of the reason why many cancers recur is that current treatments do not target a cancer's stem cells, which can resist chemotherapy and regenerate the various cell types in a tumour. So some researchers have been investigating therapies that selectively target cancer stem cells.

Now, researchers think they've found one that shows promise.

"We have found a compound selective for cancer stem cells," said senior author Kevin Struhl, a professor of biological chemistry and molecular pharmacology at Harvard Medical School. "What's different is that ours is a first-line diabetes drug."

Metformin is a medication widely used to treat type 2 diabetes and is especially effective in obese diabetics who overproduce insulin. It works by preventing the liver from breaking down starch into sugar while also stimulating uptake of sugar by the body's muscles, causing blood sugar and insulin levels to fall.

But the drug has also been found to reduce cancer risk. A number of studies have found that diabetic patients who take metformin have a significantly lower cancer rate, with those taking the highest doses for the longest period of time having the lowest rates.

In this study, the drug seemed to work independently of its ability to improve insulin sensitivity and lower blood sugar levels. Instead, when combined with doxorubicin, metformin appeared to kill human cancer stem cells and non-stem cancer cells.

For the study, researchers introduced human breast cancer stem cells into mice. Among those mice that were pretreated with metformin, the cancer cells failed to form tumours.

In other mice not pre-treated and whose tumours were allowed to take hold for 10 days, the metformin-doxorubicin combination reduced tumour mass more quickly and prevented relapse for longer than doxorubicin alone.

In the two months between the end of treatment and the end of the experiment, tumours regrew in the mice treated with chemotherapy alone, but not in those who received both drugs.

By itself, metformin was ineffective in treating tumours.

It remains to be seen whether the results from Struhl's study will be transferrable to humans with breast cancer. But Dr. Jennifer Ligibel, a medical oncologist and a Harvard Medical School instructor in medicine is about to begin a large-scale phase II trial on humans, with the National Cancer Institute of Canada's Clinical Trials Group.

That trial will compare the effect of adding metformin or a placebo to chemotherapy following surgery for early breast cancer. The study will include both patients with type 2 diabets and high insulin levels, and non-diabetic women.

"This is an exciting study," Ligibel said in response to this latest metformin study on mice.

"There is a lot of interest in studying metformin in breast cancer, but so far we do not have direct evidence that metformin will improve outcomes in patients... That's what this trial is for."