Hormone-receptor negative, HR(-), phenotype occurs frequently in diabetic women and is one of the worst prognostics in breast cancer (BC). We hypothesized that, by reducing insulin resistance in diabetics with BC, we could modify the HR status to improve BC outcomes. Our study explored for the first time whether specific diabetes mellitus (DM) therapy may be associated with more favorable HR status and improved BC prognosis as well as with lower insulin resistance and inflammation in this population. Diabetics newly diagnosed with BC between 2003 and 2010 (Roswell Park Cancer Institute) were retrospectively reviewed (n 225). HR status, outcomes, comorbidities and drug therapy were abstracted from medical records, the Institute's Breast Surgery Database and Tumor Registry. Follow up began at BC diagnosis and ended with first recurrence and/or death, or last follow-up. Plasma levels of inflammatory cytokines, adipokines, CRP and C-peptide were determined by Luminex in specimens donated at the time of BC diagnosis prior to chemotherapy or surgery. A total of 98 cases, DM+BC, and their matched controls, BC only (n 196) were analyzed. Fully adjusted hazard/odds ratios (HR/OR) and 95% confidence intervals (CI)s representing the association between DM drugs intake, HR status, specific biomarker levels and a defined BC event were computed with Cox proportional hazards model. Kaplan-Meyer analysis with log-rank statistics was also performed. After a median follow up of 30 months, patients receiving insulin sensitizers (IS) were found to have better disease-free survival (log rank X2 16.63, p 0.002), and overall survival (log rank X2 15.11, p 0.004) compared to those receiving exogenous insulin (Ins) or secretagogues (S), drugs stimulating insulin secretion. Intake of S was associated with elevated C-reactive protein, CRP (OR 1.98, CI 0.52-7.55, p 0.03) and C-peptide (OR 4.98, CI 1.03-24.03, p 0.01). As compared to BC only, the use of Ins in diabetics with BC was associated with higher levels (ng/ml) of CRP (3.62 vs 2.71, F 2.00, p 0.1), TNFa (6.98 vs 5.46, F 1.11, p 0.35), IL-8 (6.27 vs 4.74, F 1.36, p 0.25) and respectively lower leptin (1403.32 vs 1701.91, F 1.96, p 0.1) and C-peptide (322.97 vs 547.83, F 6.83, p 0.0001); adiponectin levels did not change. As compared to IS, S intake was suggestively associated with HR(-) status, although this not significant given the low sample size; OR 6.79, CI 0.65-70.97, p 0.28 for the estrogen receptor (ER) and OR 6.45, CI 1-41.58, p 0.15 for the progesterone receptor (ER). ER negative phenotype occurred more frequently in Ins receivers (OR 3.27) and S (OR 1.73) than IS (OR 0.24), when compared to BC with no DM, p 0.16. Addition of IS to the S regimen lowered this risk (OR 1.53 vs 1.73). Our data suggests that better outcomes may be achieved if IS are used for DM management in BC patients. We further speculate that our findings may also have an impact on prevention of anti-estrogen resistance.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr LB-329. doi:1538-7445.AM2012-LB-329

©2012 American Association for Cancer Research
2012