Abstract 5637: A cellular model of acquired resistance to rilotumumab (AMG 102) in glioblastoma.

Abstract

Abstract Acquired drug resistance is a long-standing problem of cancer therapeutics. The issue has become even more vexing with the development of highly selective agents; resistance to gefitinib or erlotinib is acquired frequently in lung adenocarcinomas. Thus, anticipating acquired resistance and understanding its basis may help us develop strategies to prevent or circumvent occurrence. Through its receptor tyrosine kinase Met, hepatocyte growth factor (HGF) regulates mitogenesis, motogenesis, and morphogenesis during development and adulthood. HGF/Met signaling also contributes to cancer progression in many malignancies, including glioblastoma. Rilotumumab is a fully human neutralizing monoclonal antibody against HGF tested in multiple Phase 2 clinical trials, including mono therapy in renal cell carcinoma and glioblastoma, as well as combination trials in gastric, colorectal, small cell lung cancers and castrate resistant prostate cancer. To generate a cellular model of acquired resistance to rilotumumab, an HGF/Met dependent human glioblastoma-derived cell line (U87-MG) was grown in rilotumumab (600 nM) for 120 days. Growth rate, HGF secretion, Met content and Met activation state were 10-fold, 10,000-fold, 10-fold and 80-fold higher than the parental cell values, respectively. The HGF and MET coding sequences were normal in both parental and resistant cells. Quantitative PCR studies to determine mRNA levels of all HGF isoforms revealed a dramatic increase in full-length HGF transcript. CGH array studies indicated amplification within both HGF and MET genes. Xenograft studies confirmed that tumor growth was resistant to rilotumumab, however the resistant cell line and tumors remained sensitive to a highly selective Met tyrosine kinase inhibitor, suggesting that resistance was achieved via increased HGF/Met signaling rather than mutation or activation of alternate pathways. Microarray expression analysis demonstrated transcript profiles that were consistent with HGF/Met pathway activation. Thus the molecular basis of acquired resistance in this model differs from those prevalent in: [1] lung cancers treated with EGFR inhibitors and medulloblastomas treated with hedgehog inhibitors, where most cases acquire secondary mutations in the targeted kinase; [2] breast cancers treated with HER2 inhibitors, where PTEN loss, p27 downregulation, and activation of other receptors are primary causes; or [3] malignant melanoma treated with BRAF inhibitors, where increased signaling via multiple pathways lead to PI3K- and/or MEK-mediated reactivation of the MAPK pathway. In addition to the importance of HGF/Met pathway activity in selecting glioblastoma patients for HGF-targeted therapeutics, our results suggest that monitoring Met pathway activity could provide early indications of acquired resistance to these agents, and that Met kinase inhibitors may still be efficacious when resistance occurs. Citation Format: Fabiola Cecchi, Karen Rex, Joanna Schmidt, Daniel Baker, Michael A. Damore, Angela Coxon, Teresa L. Burgess, Donald P. Bottaro. A cellular model of acquired resistance to rilotumumab (AMG 102) in glioblastoma. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 5637. doi:10.1158/1538-7445.AM2013-5637