In most patients, bladder cancer metastasis is incurable with current chemotherapy. To try to address this challenge, we have recently employed the novel COXEN algorithm (1) for in silico drug discovery. This approach utilized publicly available gene expression and in vitro sensitivity data for >40k compounds in NCI-60 cell line screen, from which concordant gene expression prediction models from COXEN enabled our discovery of an imidoazacridinone drug, NSC-637993, with high activity in vitro against forty bladder cancer cell lines (2). Examination of additional “hits” provided by COXEN revealed another compound in this class, C1311 (NSC-645809, Symadex) which is undergoing clinical trials in other tumor types.

Our objective was to develop the foundation for use of C1311 in bladder cancer, alone and in rational combination with other clinically relevant agents. Herein, we demonstrate in a panel of 40 human bladder cancer cells that in vitro cytotoxicity profile for C1311 closely correlates that of NSC-637993 and compares favorably to that of standard of chemotherapeutics. Using genome-wide patterns of synthetic lethality of C1311 with ORF knockouts in budding yeast, we determined that combining C1311 with taxanes could provide mechanistically rational combinations for patients that have failed first line therapy with platinum based chemotherapy. To evaluate the biological relevance of these yeast findings in bladder cancer, we evaluated C1311 singly and in doublet combination with paclitaxel in the hollow fiber assay in mice, observing striking efficacy of the agents in vivo. Last, applying COXEN gene expression prediction from 40 bladder cancer cell lines to human patient tumors with known cisplatin-based chemotherapy response data (N=30) (3), we provide evidence that signatures of C1311 sensitivity exist even within patients exhibiting no change or progressive disease on standard of care agents.

Taken together, these findings provide an exemplar for COXEN-based drug discovery and development, while demonstrating the utility of yeast genetics in guidance of rational combinations in vivo. Most importantly, the high activity of these agents in vitro and in vivo calls for clinical trials of these agents, while COXEN provides genomic biomarkers that may be used for patient selection.

(1) Crunkhorn S. Nat Rev Drug Discov 2007;6:782-3.

(2) Lee JK et al. PNAS 2007;104:13086-91.

(3) Als et al. Clin Cancer Res. 2007 Aug 1;13(15 Pt 1):4407-14.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 747.