Abstract 1371: Systems biology analysis to identify biomarkers for lenvatinib in the preclinical cancer cell line panels

Abstract

Abstract Objectives: Lenvatinib mesilate (lenvatinib) is the selective inhibitor of VEGFR1-3, and other proangiogenic and oncogenic pathway-related RTKs including fibroblast growth factor receptors (FGFR1-4), the platelet-derived growth factor receptor (PDGFR) α, KIT, and RET. Lenvatinib have been tested in several clinical trials, such as thyroid cancer and hepatocellular carcinoma in Ph3, but biomarkers to predict lenvatinib activity were not established. We performed a preclinical systems biology approach to identify biomarkers predictive of lenvatinib sensitivity in endometrial cell lines. Methods: We performed gene expression profiling using Affimetrix Human exon 1.0 ST array and targeted exon sequencing of 1,163 genes using Illumina sequencer in 19 endometrial cell lines. In vitro anti-proliferation activity (IC50) and in vivo anti-tumor activity of lenvatinib were determined in the panel. Besides, we did siRNA kinase library screening to classify the cell lines into two groups by unsupervised hierarchical clustering. We identified genes significantly over-expressed (p value <0.01, based on t-test) or enriched in mutations (p<0.1, based on Fishers test) between clusters. Correlation analysis between expression and in vivo anti-tumor activity was performed using t-test (p<0.05) and spearman correlation (p<0.01). Correlation between gene mutation and lenvatinib sensitivity was calculated by Wilcox-test (p<0.1) Pathway enrichment was performed by Ingenuity Pathway Analysis software. Results: Based on the clustering of siRNA data cell lines were grouped into clusters. These groups correlated with in vitro and in vivo lenvatinib sensitivity. Based on the clusters we derived genes with significantly altered expression between the groups (t-test p<0.01) or with significantly enriched mutations in one of the clusters (Fisher's test p<0.1). Additionally we performed correlation analysis between gene expression and mutation and in vivo sensitivity to lenvatinib. Pathway enrichment analysis of these signatures identified pathways associated with in vitro and in vivo lenvatinib sensitivity. We defined a set of 306 genes as potential biomarkers based pre-clinical data. Around one third of genes showed significant correlation with the clinical outcome of a Ph2 trial of lenvatinib in patients with advanced endometrial cancer. Additionally, several pathways identified based on the gene signature associated with in vitro or in vivo data, such as ERK/MAPK and Angiopoietin signaling, were commonly involved in sensitivity to lenvatinib in clinical data. Conclusion: Pre-clinical systems biology analysis identified candidate biomarkers correlating with in vitro or in vivo lenvatinib activity in 19 endometrial cell lines. A large fraction of genes from the pre-clinical analysis showed significant correlation with clinical outcome of a Ph2 trial of lenvatinib in endometrial cancer. Citation Format: Zoltan Dezso, Mitsuhiro Ino, Yukinori Minoshima, Osamu Tohyama, Naoko Hata Sugi, Sergei Agoulnik, Yoshiya Oda, Yasuhiro Funahashi. Systems biology analysis to identify biomarkers for lenvatinib in the preclinical cancer cell line panels. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1371. doi:10.1158/1538-7445.AM2015-1371