electrophoresis, trypsin digestion, peptide extraction, and high-through- put analysis by nanoflow reversed-phase liquid chromatography (na- noRPLC)tandem mass spectrometry (MS/MS). MS analysis provided protein sequence and relative protein quantification in TIF and ascites samples, respectively. Proteins identified by two or more peptide sequences in all patient samples for each respective medium were included in the analysis. Results: Four hundred fifty-nine proteins were exclusively identified in TIF in all patient samples, and 74 proteins were contained in all four ascitic fluid samples. Proteins from these two tumor environ- ments had contrasting function and location. Sixty-six percent of ascitic fluid proteins were extracellular, similar to serum composition. TIF proteins were largely subcellular, 63% cytoplasmic and 22% nuclear in origin (see the figure). Conclusions: We developed a standardized work flow for ovarian cancer biomarker detection in EOC biofluids. TIF contained largely subcellular proteins, which were of lower abundance and ovarian cancer specific, whereas ascitic fluid had a serum-like proteomic profile, with mostly nonspecific extracellular proteins. The distin- guished proteome profile favors TIF as an optimal biomarker source for ovarian cancer screening test development. doi:10.1016/j.ygyno.2010.12.118 112 EZH2 expression correlates with increased angiogenesis in ovarian carcinoma R. Ali-Fehmi 1 , Z. Al-Wahab 1 , A. Semaan 1 , S. Seward 1 , R. Morris 1 , A. Munkarah 1,2 , A. Sood 3 1 Wayne State University, Detroit, MI, 2 Henry Ford Health System, Detroit, MI, 3 University of Texas M.D. Anderson Cancer Center, Houston, TX Objective: We recently discovered EZH2, a polycomb repressor, to have substantially increased expression in tumor endothelial cells. In this study, we examined the clinical and functional significance of EZH2 and its correlation with VEGF and angiogenesis in ovarian carcinoma. Mouse ovarian endothelial cells (MOEC) were transfected with the Renilla luciferase plasmid either with or without the EZH2 promoter construct. Cells were then treated with VEGF (conditioned medium from SKOV3 ovarian cancer cells [SKOV3-CM]). EZH2 mRNA was quantified using real-time RT-PCR. EZH2 protein levels were evaluated with Western blot. One hundred eighty paraffin-embedded epithelial ovarian cancer specimens (collected between 1985 and 2004) with available clinical outcome data were identified. Immu- nohistochemistry (IHC) for EZH2, CD34, and VEGF was performed. For EZH2 and VEGF, the stained slides were scored based on intensity and percentage of cells stained and categorized as high and low expressors. Microvessel density (MVD) was quantified as the number of blood vessels staining positive for CD34 in 10 high-power fields, and the mean was calculated. Clinical parameters and survival data were obtained from the clinical information system and SEER registry. Results: In MOEC, there was a significant increase in EZH2 promoter activity and EZH2 mRNA expression levels in response to VEGF (SKOV3-CM). This increase in EZH2 promoter activity and mRNA was blocked by the VEGFR2-specific antibody DC101. The increase in EZH2 protein levels in response to VEGF was blocked by the anti-VEGFR2 antibody. Using IHC, EZH2 expression was evaluated in 180 cases in both tumor and endothelial compartments (EZH2-T and EZH2-E). Sixty-six and sixty-seven percent of the samples showed high EZH-2 expression in the tumor and endothelial compartments, respectively. High expression of EZH2-T and EZH2-E was significantly associated with high-stage (P < 0.001) and high-grade (P < 0.05) disease. High EZH2-T and EZH2-E expression was also significantly associated with decreased overall survival (median: 2.5 years vs 7.33 years, P < 0.001, for EZH2-T; 2.33 vs 8.33 years, P <0.001, for EZH2-E). Tumors with high VEGF expression were significantly correlated with high EZH2-E expression (P <0.001). Moreover, tumors with high EZH2-E expres- sion also had significantly greater MVD (P < 0.001). Conclusions: These findings suggest that EZH2 may be a regulator of tumor angiogenesis and support the possibility of its being a therapeutic target in ovarian carcinoma. doi:10.1016/j.ygyno.2010.12.119 113 Identification and characterization of CD44+/CD24ovarian cancer stem cell properties and their correlation with survival E. Meng, L. Shevde, B. Long, P. Sullivan, S. McClellan, M. Finan, E. Reed, R. Rocconi University of South Alabama Mitchell Cancer Institute, Mobile, AL Objective: Cancer stem cells are considered to be primarily responsible for cancer self-renewal, invasion, and resistance to therapy. We describe a subpopulation of ovarian cancer cells with the surface marker profile CD44+/CD24that exhibits the cancer stem cell properties of enhanced differentiation, invasion, resistance to therapy and correlation with survival. Fluorescence-activated cell sorting (FACS) was used to sort ovarian cancer cell lines (TOV112D, OV90, SKOV3, ES2) into two phenotypically distinct populations: OCSC (CD44+/CD24) and non-OCSC (all other phenotypes). Each phenotype was evaluated in a serial differentiation experiment to determine the level of differentiation at 24, 48, and 72 hours. Invasion properties were measured using the Matrigel invasion assay. Cells were treated with carboplatin, paclitaxel and paclitaxel+carboplatin and evaluated with the MTS cell survival assay to determine relative resistance to chemotherapy. Ascites derived from 20 patients with advanced-stage ovarian cancer was obtained and evaluated for the proportion of CD44+/CD24cells and its correlation with survival. Results: The proportion of CD44+/CD24cells corresponded to the clinical aggressiveness of each ovarian cancer cell line histologic subtype: in order of increasing aggressiveness, endometrioid, TOV112D––0.5%; papillary serous, SKOV3––66% and OV90––77%; and clear cell, ES2––99%. OCSC demonstrated enhanced progressive S49 ABSTRACTS / Gynecologic Oncology 120 (2011) S2S133