A landmark study in people with breast cancer has shown that the oral antidiabetic agent metformin does not delay disease recurrence or death overall but may be effective in a subgroup of people with human epidermal growth factor receptor (HER)2-positive tumors.
The phase 3 MA.32 trial, led by Pamela Goodwin from Sinai Health in Toronto, Ontario, Canada, and run by the Canadian Cancer Trials Group under the umbrella of the Breast International Group network, included 3649 participants (mean age 52.4 years, 99.8% women) from Canada, the USA, Switzerland, and the UK who had non-metastatic breast cancer and no history of diabetes.
They were randomly assigned to receive metformin 850 mg twice daily – a dose similar to that given in diabetes – or placebo for 5 years following complete resection of their breast cancer. The study drugs were given alongside standard adjuvant therapy, where indicated.
Goodwin told Inside Precision Medicine that metformin has previously been associated with a reduced risk for breast cancer recurrence, but only in retrospective observational studies with a high risk for bias. It was not clear whether metformin itself was reducing breast cancer risk, or whether it was an indirect effect of treating diabetes, which has also been shown to make breast cancer outcomes worse.
The MA.32 trial set out to address these uncertainties and the findings are reported in JAMA.
During almost 8 years of follow-up, there was no significant difference in the rate of invasive local, regional, or distant recurrences, new primary invasive cancers (breast or nonbreast), or death between the participants who received metformin and those who did not, regardless of their hormone receptor (HR) status.
Among the 2533 HR-positive participants, the incidence rates for invasive disease-free survival (DFS) events were 2.78 and 2.74 per 100 patient–years in the metformin and placebo groups, respectively. The corresponding rates among the 1116 HR-negative patients were 3.58 and 3.60 per 100 patient–years.
However, exploratory analyses among HER2-positive participants who carried at least one C allele (CC or AC) of the rs11212617 single nucleotide variant (SNV) showed that individuals in the metformin group had significantly better invasive DFS rates than those in the placebo groups, at 1.74 versus 3.48 events per 100 patient–years. This translated to a significant 49% lower risk for invasive disease recurrence or death with metformin.
Similar results were observed for overall survival, but there was no significant difference in either outcome with metformin versus placebo in participants who carried the AA genotype of rs11212617.
The C allele of this SNV, which is carried by approximately 70% of people, may be important because it “has been identified by endocrinologists as a predictor of metformin benefit to treat diabetes,” Goodwin explained. “If you have this C allele you’re more likely to get good glucose control and in fact, you may have higher metformin blood levels.”
She also said that “the allele is very close to the ATM gene, which is a gene that is important in breast cancer. So it may have had a very direct effect on breast cancer.”
Of note, there was no differential effect by allele type (either A or C) when the data were analyzed by HR status and the researchers are now carrying out additional analyses to understand the mechanism behind the benefit they observed in HER2-positive C allele carriers.
Another avenue that may warrant further investigation according to Goodwin is the role the PI3 kinase plays in response to metformin. Metformin can directly impact this pathway activating AMP kinase and inhibiting mTOR or indirectly attack it by lowering insulin levels, with both mechanisms leading to reduced cell growth and proliferation.
Interestingly, EGFR-mutated non-small-cell lung cancer (NSCLC), like HER2-positive breast cancer, is driven by mutations in the PI3 kinase pathway and a previous trial in people with EGFR-mutated NSCLC showed that people given metformin had a higher response rate to tyrosine kinase inhibitor therapy than those not given metformin.
“So if you put this all together, what it may be telling us is that metformin may prove to be to have a clinical value or a clinical role in cancers that are driven by that PI3 kinase pathway,” Goodwin remarked.
However, she concluded that “right now, what we can say is in unselected non-diabetic breast cancer patients, metformin is ineffective if the breast cancer is hormone receptor positive or negative.”
“If the breast cancer is HER2-positive, there is evidence in exploratory analyses, which needs replication, that metformin may be beneficial.”