Does metformin use influence outcome in diabetic women with invasive breast cancer?

Abstract

600 Background: Preclinical and clinical data suggest complex associations between diabetes mellitus (DM) and breast cancer (BC). Emerging evidence suggests that use of the hyperglycemic agent, metformin, may have a beneficial impact on tumor characteristics and outcome in diabetic women with BC. Methods: Retrospective review of institutional databases from 1998-2005 identified 313 diabetic patients who received systemic treatment for invasive BC. Male patients, those with type I DM, gestational or steroid-induced diabetes were excluded. We also excluded those diagnosed with DM >6 months after BC diagnosis. The 164 remaining women with type II DM formed our study cohort. Kaplan-Meier methods were used for comparisons between patients taking metformin and those using other DM management agents. Results: Of 164 women receiving systemic therapy for invasive BC, 76 (46%) were using metformin at cancer diagnosis. Patients using metformin had a longer history of DM, median 5 vs. 3 years (p=0.025), and were more likely to use >1 agent for DM 45% vs. 30% (p=0.002). There was no difference in clinical presentation (palpable mass vs. mammography), mean tumor size, histology, grade, lymphovascular invasion, nodal status, hormone receptor or Her2 status between the two groups. There was also no difference in receipt of neo-adjuvant or adjuvant chemotherapy, hormone therapy or trastuzumab. At a median follow-up of 72.4 months, there was no difference in the rate of local, regional, or distant recurrence, contralateral BC or overall survival. Conclusions: In this sample of patients with Type II DM receiving systemic therapy for primary invasive BC, there was no difference in primary tumor features or short term BC outcomes between those using metformin and those using other DM agents. These data do not support a beneficial effect of metformin on BC outcome. A prospective study with longer follow-up is needed to further clarify this relationship. Metformin (N=76) No metformin (N=88) P value (log-rank test) Median follow-up (mos) 70.1 74.9 5-year event rate Overall survival 92.7% 91.3% 0.80 Recurrences Local 1.3 3.8 0.24 Regional 1.3 2.6 0.84 Distant 8.6 9.8 0.63 Contralateral breast cancer 4.7 9.1 0.23