SAN ANTONIO—Metformin, the biguanide agent taken by millions of people with diabetes to control their disease, created a splash at ASCO's 2012 Annual Meeting, with workshops and seminars on the drug “repurposed” as an anticancer agent (OT, 5/25/12).
The mood was far more subdued, though, here at the CTRC-AACR San Antonio Breast Cancer Symposium, where metformin research was featured in only five presentations, in a poster-discussion session.
Previously, metformin had been found in small studies to reduce the overall risk of cancer incidence by 31 percent, decrease cancer-related mortality, and enhance the response to neoadjuvant chemotherapy in people with type 2 diabetes. A strong relationship between hyperinsulinemia, insulin resistance, and breast cancer has been established.
But the Discussant for the poster session, Michael N. Pollak, MD, Professor in the Departments of Medicine and Oncology and Director of the Division of Cancer Prevention at McGill University, said that follow-up pharmacoepidemiological studies have been inconsistent, and that flaws have been found in some prior studies that claimed to show reduced cancer risk or improved outcome in patients with diabetes treated with metformin.
“We cannot assume those papers are definitive, and there are discussions about whether this is related to flaws [in the earlier studies], or whether this was an early true signal,” said Pollak, whose research program includes a focus on insulin-like growth factors in the pathophysiology of neoplasia.
Pollak cited a December paper in Diabetes Care (2012;2665-2673) that questioned the findings of retrospective studies that showed dramatically lowered cancer risk with metformin treatment and speculated that time-related biases could have exaggerated the drug's potential beneficial effects.
Recent small trials are also yielding disappointing data, Pollak noted in an email exchange after the meeting. “My view is that it is too early to be sure, but few would say the early results are actually encouraging.” Researchers are awaiting the results of larger trials, as well as results of ongoing laboratory studies investigating strategic drug combinations involving biguanides besides metformin not yet in clinical trials.
“There are tantalizing clues that justify the investigation of antineoplastic activities of biguanides,” he wrote in a recent review in Cancer Discovery (2012;2:778-790). “The complexity of their biologic effects requires further translational research to guide clinical trial design.”
At the poster discussion, he said investigation of metformin for cancer prevention and treatment is “at the end of the beginning,” quoting the title of his review paper.
The moderator of the poster-discussion session, Adrian Lee, PhD, Professor in the Department of Pharmacology/Chemical Biology, who has a research interest in insulin-like growth factors and breast cancer, was also optimistic, and said the paucity of presentations at the meeting did not indicate any waning of interest in metformin.
“These [papers] are the first trials to come out, window-of-opportunity trials to see how metformin works, a movement toward the clinical setting,” he said in an interview before the session. He said researchers are ready to discuss the “bigger picture of what's next, what combinations we can use metformin with, and in what clinical setting.”
In his discussion at the meeting, Pollak reviewed recent developments in metformin research. Among other points, he said metformin at conventional antidiabetic doses is safe and that no Phase I trials are required for cancer research, but researchers now recognize that optimum metformin concentrations in laboratory models are higher than those achieved by conventional antidiabetic dosing, opening the possibility that exposure levels in models may be unachievable in the clinic.
In addition, while metformin has potential as an antineoplastic agent, it may not be “pharmacologically optimized” for that use, he said, predicting that derivatives of metformin with superior antineoplastic activity will be the subject of future research. One candidate he mentioned was the IGF1 insulin kinase-receptor inhibitor BMS-536924, which has been found to be more effective than metformin in the treatment of experimental insulin-responsive breast cancer.
Reports Raise More Questions
The papers reported in the poster discussion (PD03-01, PD03-02, PD03-03, PD03-04, PD03-05) illustrated the changing nature of metformin research at the moment. Researchers from the European Institute of Oncology reported that use of metformin before surgery did not significantly affect Ki67 overall in non-diabetic patients, but that it showed significantly different effects according to insulin resistance, particularly in luminal B tumors.
More importantly, though, Pollak said the study showed an effect of placebo on apoptosis, suggesting either that there is a measurement technical issue, or that the surgery itself has an effect, or there is an issue of exactly how the biopsies were obtained. “If we see that those effects were associated with placebo, we have to be careful interpreting the metformin effects,” he said.
In another study, researchers from the University of Sheffield, UK, confirmed the concept that metformin works in vivo via upreglation of tumor pAMPK and downregulation of pAkt and proliferation, suggesting a mechanistic effect for metformin. But an analysis conducted by researchers at the Ontario Cancer Institute and colleagues, of tumor cell signaling in response to neoadjuvant metformin in women with early stage breast cancer, showed that while metformin reduced the phosphorylation of Akt and lowered tumor insulin-resistance expression, it did not induce AMPK phosphorylation.
Pollak said that preclinical models have demonstrated more clearly AMPK activation by metformin than the clinical studies presented at the San Antonio meeting do, another note of caution for investigators.
Researchers from UCLA Geffen School of Medicine reported that metformin analogs were more effective than metformin at inhibiting cell proliferation and inducing apoptosis of ER-positive and triple-negative breast cancer cells in vitro.
And a study from the College of Physicians and Surgeons at Columbia University did not identify a significant difference in tumor proliferation, as measured by Ki-67 tumor levels, before or after metformin use in overweight and obese patients prior to breast cancer surgery. This was in contrast to prior presurgical trials that did show changes in markers of tumor proliferation, the researchers said.