? DESCRIPTION (provided by applicant): During the process of screening for drugs with therapeutic potential, often times several candidates emerge as prime candidates with similar efficacy and the choice for a lead compound is difficult to make. However, people's lives and billions of dollars may rest on the choice of which compound to bring to clinical investigation and trials. Since any levels of toxicity caused the by drug would contra-indicate its choice as a lead compound, detection of very low-level toxicity could form the basis for ranking such groups of compounds in an objective fashion. Accordingly, the overall goal of the proposal is to build an algorithm relating a novel ultra-sensitive in vitro measure of toxicity to clinical toxicity, whichwill be used to rank candidate drugs being considered as lead compound of clinical trials. We reasoned that even a very slow rate of cell death induced by a chronically administered drug might -- over the years -- kill off a significant proportion of an organ or cell type. However, a change in cell number of only a few percent is below the detection limit of current in vitro tests. We hypothesize that some side effects of drugs occurring in vivo could be revealed in in vitro tests if they were sensitive enough. We have established unique methods to measure cell number and metabolic viability as reflected by continuous measurement of oxygen consumption with the required sensitivity. The key to establishing the utility of our approach will be to demonstrate strong correlation with in vivo safety and toxicology using a relevant panel of benchmark compounds about which prior in vivo data is well known, and that contains a spectrum from highly safe drugs to those with known toxicity. We will optimize methods to measure small changes in oxygen consumption (a measure of cell number) in various selected cell types, and verify the methods by testing with compounds that have varying degrees of toxicity. This will lay the foundation for the eventual construction of an objective algorithm that can be used to rank the potential of lead compounds on the basis of toxicity caused by chronic administration.