Feedback System Control (FSC) on Cancer Eradication (Chien Sun)
Figure. Average cell viability of WEHIs and MEFs normalized for each iteration through the DE algorithm. A total of 14 iterations were performed. Each iteration has 22 test vectors for WEHIs and 22 test vectors for MEFs. Each cell viability data point is an average of 22 test vectors in its corresponding iteration. Cell viability for each test vector was normalized to the DMSO control for each iteration.
We are using systems control algorithms to identify an anti-cancer agent combination that is toxic to leukemia cells while sparing non-malignant cells. Clinically, anti-cancer drug cocktails have several important advantages over single drug therapies that can include a) lowered concentrations of each drug than if used singularly; b) reduced development of drug resistance; and c) increased drug efficacy through synergistic drug interactions (Sawyers 2007, Catley et al. 2005). The mouse B cell leukemia line WEHI-231 is being tested in combination with mouse embryonic fibroblast (MEF) cells. Six drugs were chosen for study including etoposide, 5-fluorouracil, doxorubicin, docetaxel, vincristine, and rapamycin. Three of the six drugs specifically targets DNA (5-fluorouracil, etoposide, doxorubicin), 2 drugs target microtubules (vincristine, docetaxel), and 1 drug targets the mTOR signaling pathway (rapamycin). Preliminary studies have determined the effective concentration ranges of each of the six drugs in both cell types for testing. The Differential Evolution (DE) algorithm was used in this study with a total of 22 test cases for each generation (Storn and Price, 1997). Cells were treated with drug cocktails and analyzed using flow cytometry after 48 hours. Non-optimized drug cocktails in the initial DE generations resulted in more than 90% cell death of both WEHI and MEF cells. After 15 search generations, optimized drug cocktails were attained that has up to 90% cell death in WEHI Cells and only 5~10% cell death in MEF cells. Future plans include testing an optimized anti-cancer drug cocktail, developed on the feedback-control platform, in vivo using adoptive transfer of B cell leukemia cells into immunodeficient mice, and integration of Phosphoflow and TNPR technologies to elucidate the signalosome network that operates to control leukemia.