Anticancer drugs are rarely curative as single agents, thus most treatment regimens utilize a combination of agents. Selection of rational combinations based upon presumed mechanisms of action is an active area of research. Nonetheless, treatment regimens may have built up over time by addition of new agents to existing standards of care. To establish a framework for extending the understanding of combination therapy, we have initiated an additional approach: the systematic testing of all pairwise combinations of anticancer drugs approved the Food and Drug Administration (FDA) in the NCI-60 panel of human tumor cell lines.
About 100 small molecule drugs are approved for cancer treatment worldwide. Combinations are tested at 3 or more concentrations of each agent, and single agents are tested on the same plates. About 100 pairs of drugs have been tested in the NCI-60 in a pilot phase. Some combinations of drugs show better than single agent activity in nearly all cell lines tested. Other combinations show more restricted benefit, with some cell lines showing enhanced growth inhibition or cell kill, while in other cell lines the combination is antagonistic or identical to the more active single agent.
The NCI-60 panel of cell lines has been extensively molecularly characterized, with publicly available data including gene mutation, DNA copy number, DNA methylation, and expression of mRNA, protein and microRNA. The patterns of cell line sensitivity for a particular drug combination can be used to probe these molecular characterization data, generating hypotheses about potential predictive markers and/or mechanisms of action. The most promising combinations identified from NCI-60 testing are being tested in vivo to determine the therapeutic index. The combination screening data, along with tools to analyze the data, will be made available to the public through the NCI website.