Metformin failed to improve invasive disease-free survival or overall survival when used as adjuvant treatment in patients with early breast cancer, irrespective of estrogen or progesterone receptor status.
Metformin (Fortamet) failed to improve invasive disease-free survival (IDFS) or overall survival (OS) when used as adjuvant treatment in patients with early breast cancer, irrespective of estrogen or progesterone receptor (PR) status, according to data from the phase 3 CCTG MA.32 trial (NCT01101438) presented during the 2021 San Antonio Breast Cancer Symposium.
The primary analysis showed that outcomes in patients with estrogen receptor (ER)–positive and PR-positive disease was similar between the metformin and placebo arms (HR, 1.01; 95% CI, 0.84-1.21; P = .93).
“Metformin does not improve IDFS, OS, or other breast cancer [BC] outcomes in moderate/high risk estrogen receptor (ER)/PR-positive or ER/PR-negative breast cancer patients and should not be used as breast cancer treatment in those groups,” Pamela J. Goodwin, MD, MSc, FRCPC, lead investigator and Professor of Medicine at the University of Toronto, said during the presentation.
The research team attempted to determine if metformin was capable of inducing better tumor responses based on the association of obesity with poor breast cancer outcomes and the agent’s ability to promote weight loss and lower insulin levels.
Metformin was given at 850 mg twice per day for 5 years versus placebo, with a 4-week ramp up of once daily. Patients were eligible for this study if they had a diagnosis of invasive breast cancer within 1 year and negative margins following surgery. Tumors had to be staged as T1c to T3 and N0 to N3, with T1cN0 requiring additional adverse features. All patients must have received standard breast cancer therapy and be not diabetic.
The primary analysis was in patients with ER/PR–positive breast cancer, with 1268 patients in the metformin group and 1265 in the placebo group. They had median ages of 52 (range, 25-74) and 53 (range, 25-74) years, respectively, and 62.1% and 60.2% were postmenopausal. A total of 52.5% and 54.2%, respectively, had T2 tumor stage, and most disease was grade 2 or grade 3. Additionally, 16.5% and 17.4% had HER2-positive disease, with 97% receiving trastuzumab (Herceptin).
IDFS events were noted in 18.5% and 18.3% of patients receiving metformin and placebo, respectively. Rates of distant, local/regional, and contralateral invasive recurrence as well as new primary cancer were similar between groups.
In total, 131 patients (10.3%) in the metformin arm died compared with 119 (9.9%) in the placebo arm (HR, 1.10; 95% CI, 0.86-1.41; P = .47). Cause of death in the metformin arm included breast cancer (n = 99; 7.8%), other primary malignancies (n = 15; 1.2%), and cardiovascular disease (n = 4; 0.3%) among other factors (n= 13; 1.0%) with rates being similar to those in the placebo group.
A total of 21.7% of patients in the metformin arm experienced grade 3 or higher adverse effectscompared with 18.7% in the placebo arm. Symptoms include nausea, vomiting, bloating, and diarrhea.
Results involving patients with ER/PR-negative breast cancer showed that futulity was established at 29.5 months of follow-up with 172 IDFS events. At the 96-month follow-up, there were 245 IDFS events in 1116 patients, with no benefit noted for the use of metformin on IDFS (HR, 1.01; 95% CI, 0.79-1.30; P = .92) or OS (HR, 0.89; 95% CI, 0.64-1.23; P = .46).
The exploratory analysis looked at patients who had HER2-positve breast cancer. Interestingly, patient who had at least 1 C allele of a prespecified ATM associated rs11212617 snp had a higher pathologic complete response rate with metformin than those who had no allele. Goodwin said any C allele is associated with a metformin benefit on glucose control in diabetes.
A total of 620 HER2-poistive cases were analyzed, with 99.4% receiving chemotherapy and 96.5% receiving trastuzumab. There were 99 IDFS events, and 47 OS events. When looking at the entire HER2-positive population, patients had fewer IDFS events when taking metformin (HR, 0.64; 0.43-0.95; P = .026) compared with the placebo arm. There were also fewer patient deaths with metformin (HR for OS, 0.53; 95% CI, 0.30-0.98; P = .038).
There was a significant interaction between snp and IDFS with metformin (HR, 0.51; 95% CI, 0.31-0.83; P = .007), whereas those with the AA genotype had no benefit of metformin (HR, 1.32; 95% CI, 0.58-2.96; P = .51). This association extended to OS with the snp being statistically significant for metformin benefit (HR, 0.35; 95% CI, 0.17-0.73; P = .003) vs the AA genotype (HR, 2.15; 95% CI, 0.56-8.36; P = .26).
“Exploratory analyses in HER2-positive breast cancer suggested a beneficial effect of metformin on IDFS and OS, notably in patients with at least one C allele of the rs11212617 snp. These observations should be replicated in future research,” concluded Goodwin.