Metformin may represent a novel antitumor agent, say investigators from the University of Texas M.D. Anderson Cancer Center, Houston.
Metformin may represent a novel antitumor agent, say investigators from the University of Texas M.D. Anderson Cancer Center, Houston. These results were presented at the 2008 American Society of Clinical Oncology (ASCO) Annual Meeting, Chicago.
In a retrospective study, patients with early-stage breast cancer and diabetes who received neoadjuvant systemic therapy and metformin had higher pathologic complete response rates than those not receiving metformin.
Metformin's mechanism of action seems to be through activation of AMP kinase. "AMP kinase is a cellular energy sensor," said Sao Jirlerspong, MD, PhD, a fellow at M.D. Anderson and lead investigator. "Metformin activates AMP kinase in states of nutrient deprivation or low oxygen (stress scenarios). That action phosphorylates a number of protein targets and the end effect is shutdown of cell growth. One of the important pathways it hits is the mTOR pathway."
Another potential mechanism for the antitumor effect of metformin may be the decrease in insulin resistance it produces; insulin may be a growth factor for cancer.
From the M.D. Anderson Breast Medical Oncology database, investigators identified 2,529 women with early-stage breast cancer who received chemotherapy in the neoadjuvant setting (2,374 without diabetes, 68 with diabetes who were not taking metformin, 87 with diabetes who were taking metformin).
Pathologic complete response rate in the patients with diabetes who were taking metformin was 24% compared with 8% in the patients with diabetes who were not taking metformin (p = .03). The complete pathologic response rate was 16% in the women without diabetes.
After adjustments for tumor stage and grade, hormone receptor status, HER2 status, ethnicity, and menopausal status, the odds ratio for achieving a pathologic complete response was 1.32 in the patients with diabetes taking metformin and 0.45 in those not taking metformin (p = .03) compared with the nondiabetic patients.
"We generated some interesting results that deserve further study," said Dr Jirlerspong.