PK/PD-based Development of Anti-infective Agents

The cost of drug development has exploded in recent years and risen to a level that soon will no longer be affordable to society. The public expectation of drug safety and guaranteed therapeutic success has become unrealistic. As a result, the number of new drug approvals can be expected to go down in the near future, a trend that is already noticeable in drug classes with low market potential due to short term therapeutic use, e.g. antibiotics. One reason for the high cost of drug development is the many unnecessary studies, where the results could have been predicted with reasonable certainty. PK/PD modeling is a tool that can be used to collect and integrate all the available information about a drug candidate and its class in order to make rational decisions on studies that will decrease the uncertainty of the compound. It is based on quantitative data on drug exposure and response, and it is particularly well suited to address the question of dose determination and optimization. In the drug development process, it bridges the complete cycle, from discovery to clinical use. The advantage of this approach is to define objective go/no-go decision criteria for the development process, rather than relying on subjective empirical decisions. There is no way that, today, all developing questions can be answered by experimental evidence; modeling and simulation is a powerful alternative approach. This modeling and simulation approach is of particular need in the field of new anti-infective agents, where the rise of resistance has become an international threat to society. However, very few drug companies are currently developing new antibiotics, due to the poor prospects of return on investment. However, the cost of anti-infective drug development can be dramatically lowered by applying pharmacometric concepts and selecting some key experiments based on pharmacokinetic/pharmacodynamic (PK/PD) concepts. Using microdialysis, it is currently possible to measure local exposure at the infection site in both animals and humans. This PK information is much more useful than traditional serum pharmacokinetics. Furthermore, pharmacodynamic activities can be captured more precisely by analyzing time-kill curves, rather than simple minimum inhibitory concentrations (MICs). Examples from various classes of antibiotic drugs will be presented, where these concepts are applied and illustrated. Application of these concepts will help to develop new anti-infective treatments at low cost to combat resistance development with optimum efficacy and safety.