Effect of Metformin Versus Placebo on New Primary Cancers in Canadian Cancer Trials Group MA.32: A Secondary Analysis of a Phase III Randomized Double-Blind Trial in Early Breast Cancer.

JCO Metformin has been associated with lower cancer risk in epidemiologic and preclinical research. In the MA.32 randomized adjuvant breast cancer trial, metformin ( placebo) did not affect invasive disease-free or overall survival. Here, we report metformin effects on the risk of new cancer. Between 2010 and 2013, 3,649 patients with breast cancer younger than 75 years without diabetes with high-risk T1-3, N0-3 M0 breast cancer (any estrogen receptor, progesterone receptor, human epidermal growth factor receptor 2) were randomly assigned to metformin 850 mg orally twice a day or placebo twice a day for 5 years. New primary invasive cancers (outside the ipsilateral breast) developing as a first event were identified. Time to events was described by the competing risks method; two-sided likelihood ratio tests adjusting for age, BMI, smoking, and alcohol intake were used to compare metformin versus placebo arms. A total of 184 patients developed new invasive cancers: 102 metformin and 82 placebo, hazard ratio (HR), 1.25; 95% CI, 0.94 to 1.68; = .13. These included 48 contralateral invasive breast cancers (27 metformin 21 placebo), HR, 1.29; 95% CI, 0.72 to 2.27; = .40 and 136 new nonbreast primary cancers (75 metformin 61 placebo), HR, 1.24; 95% CI, 0.88 to 1.74; = .21. Metformin did not reduce the risk of new cancer development in these nondiabetic patients with breast cancer.