Abstract P6-12-07: Discordance in cyclin D1 changes after metformin exposure by different protein expression methods: Results from a "Window of Opportunity" trial

Abstract

Abstract Background: Laboratory and population studies demonstrate that metformin offers a beneficial breast cancer (BC) effect. In vitro, metformin has been shown to induce cycle cycle arrest. In a pre-surgical metformin trial of overweight/obese, multi-ethnic BC patients, we reported no difference in tumor proliferation, as measured by ki-67. Reverse Phase Protein Array (RPPA) is a high-throughput antibody-based technique to assess cellular protein activity in signaling networks. The goal of this study was to assess changes in cyclin D1 by RPPA as compared to immunohistochemistry (IHC) in patients treated in a pre-surgical metformin trial. Methods: Metformin 1500mg PO daily (500mg am/1000 mg pm) was administered for 2-4 weeks prior to resection in 35 patients with stage 0-III operable BC, BMI > 25 kg/m2, and no history of diabetes. All tumor analysis was performed on paraffin-embedded tumor tissue. For RPPA, protein was extracted from pre- and post-metformin tissue, denatured by sodium dodecyl sulfate, and printed on nitrocellulose-coated slides. Samples were probed with 160 antibodies associated with various cellular activities, including cyclin D1. For RPPA, the cyclin D1 antibody used was by Santa Cruz (SC-718 rabbit polyclonal). For IHC, the cyclin D1 antibody used was by Ventana (SP4-R rabbit monoclonal). We analyzed changes in protein expression in tumor tissue of study patients with those of untreated historical controls, matched by age, BMI, and tumor characteristics. For RPPA and IHC, paired t-test was used to calculate within-group changes, and two-sample t-tests were used to compare between-group changes in cases and controls (significance: p ≤ 0.05). For RPPA, multiple comparisons were adjusted for by fixing the false discovery rate (FDR) at 25%. Pearson’s correlation coefficient was used for correlation between RPPA and IHC Results: Of the 35 metformin-treated patients, 32 were evaluable. The majority were Hispanic (80%). Metformin was administered for a median of 23 days (range: 8-64). Of the invasive BCs (n=21/35), 80% of patients had HR+/HER2- BC. The 33 historical controls were well-matched. Adjusting for multiple comparisons, there was a statistically significant increase in cyclin D1 by RPPA after metformin vs. control [mean change from baseline after metformin: +0.065 (0.118) vs. mean change from baseline in control: -0.044 (0.152), p=0.002]. There was no change in cyclin D1 by IHC after metformin vs. control [mean change from baseline after metformin: -5.64 (21.5) vs. mean change from baseline in control: -6.37 (14.28), p =0.88]. There was no correlation in cyclin D1 between IHC and RPPA (estimate = 0.09, p=0.31). In vitro assessment of cyclin D1 after metformin use in various cell lines is ongoing. Conclusions: We report no correlation between cyclin D1 between IHC and RPPA. Pre-clinical assessment of changes in the cell cycle after metformin is ongoing. Interpreting results of proteomic findings based off of paraffin-embedded tissue should be done so with caution. Citation Format: Kevin Kalinsky, Hanina Hibshoosh, Tian Zheng, Katherine D Crew, Susan Refice, Sheldon Feldman, Bret Taback, Eileen Connolly, Matthew Maurer, Dawn L Hershman. Discordance in cyclin D1 changes after metformin exposure by different protein expression methods: Results from a "Window of Opportunity" trial [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P6-12-07.