Abstract B232: Preclinical evaluation of AMG479 a fully human insulin-like growth factor receptor-1 (IGFR1) antibody in ovarian cancer cells

Abstract

Abstract Background: Dysregulation of the insulin-like growth factor receptor-1 (IGFR1) signaling pathway has been implicated in the development and progression of ovarian cancer. Here we explore the therapeutic potential of AMG479 a fully human monoclonal antibody that targets the IGFR1 for the treatment of ovarian cancer. Methods: We examined the effects of AMG479 on tumor cell growth in a panel of 40 characterized human ovarian cancer cell lines. Molecular markers for response prediction were studied using gene expression profiling, multiplexed phospho-protein ELISA, array comparative genomic hybridization, and mutational analyses. Associations between biomarkers and growth inhibition were analyzed using Spearman's rho correlation, and absolute levels were compared using the Mann-Whitney U test. Results: Concentration-dependent antiproliferative effects of AMG479 were seen in nearly all ovarian cancer cell lines tested but varied significantly between individual cell lines. In vitro sensitive cell lines demonstrated significantly higher expression levels of IGFR1 (r = 0.36, p = 0.024), IGFBP5 (r =0.45, p =0.005), but lower levels of insulin receptor (r = -0.28, p =0.096). Eighteen of the 40 (45%) ovarian cancer cell lines demonstrated PI3K, KRAS, BRAF, or PTEN mutations. Cell lines without these mutations in the downstream signaling pathway of IGFR1 were significantly more sensitive to AMG479 when compared to those cell lines with downstream mutations (p = 0.026). Response to AMG479 was significantly correlated with its ability to inhibit (pre- versus post-treatment) phospho-p70S6K (r = 0.35, p = 0.028), and phospho-GSK-3 (r = 0.43, p = 0.005), but not with its ability to inhibit phospho-IGFR1 (r = 0.21, p = 0.183). Gene copy number gains or LOH of PTEN or PI3K, although common, were not associated with in vitro response to AMG479. Conclusions: AMG 479 is a promising candidate for therapeutic use in IGFR1 dependent ovarian cancers. The present correlative study will help to identify biomarkers that may predict sensitivity to IGFR1 inhibition in ovarian cancer. PTEN, PI3K, or KRAS mutations may inversely correlate with sensitivity. Citation Information: Mol Cancer Ther 2009;8(12 Suppl):B232.