It is an FDA requirement that the“first in human" dose be based on pre-clinical animal model efficacy and safety testing to ensure a safe entry into Phase I clinical trials. Pre-clinical safety and efficacy models range from mouse to non-human primates. Interspecies scaling of pharmacokinetic parameters is therefore important for predicting drug doses in human clinical trials, although it continues to be less than optimal. Understanding the disposition of the compound in different species through in vitro and in vivo experiments is necessary to ensure appropriate species are selected for human estimates. Data for three imaging agents and a pharmacological stress agent (Oncology tumor agent (DPC-A80351), Thrombus agent (DMP-444), Infection agent (RP-517) Pharmacological stress agent (DPC-A78445-00)) that entered clinical trials and an imaging agent being developed (RP845), were assessed for scaling accuracy. Initially, pharmacokinetic data from animal models were used to extrapolate to human though body weight allometric scaling. Subsequently, the impact of adjusting for plasma protein binding and the impact of metabolic stability in the different models were examined. Allometric scaling of animal pharmacokinetic parameters (clearance (CL), half-life (t½) and volume of distribution (Vdss)) achieved a prediction of the human pharmacokinetic parameter within 13 to 109% of the observed values. This prediction was further improved by adjusting for plasma protein binding of the drug, and achieved an estimate within 5 to 57% of the clinically observed values. Since the parent compound was the dominant species (>95%) in the circulation, metabolic stability was not used as a correction factor. Weight based allometric scaling was further examined for an atherosclerotic plaque targeted radiopharmaceutical imaging agent, RP845-Tc-99m, currently in development. Pharmacokinetic parameters were determined in mouse, rat and rabbit followed by allometric scaling to predict the non-human primate values. Differences between predicted versus observed non-human primate Cl, t½ and Vdss were 40%, 52% and 8%, respectively. Correcting for plasma protein binding improved the prediction for Cl and t½ to within 12 and 3 %, respectively. The Vdss prediction, however became less accurate (38% difference). Since blood clearance is the major parameter in predicting human dose, the improvement from 40% to 12% was important. The plasma protein binding adjusted animal data was then used with allometric scaling to predict human CL, t½ and Vdss. The predicted values were 7.6 mL/min/kg, 70.6 minutes and 0.87 L/kg respectively. Based on the predicted human blood clearance and the dose required to image atherosclerosis in a rabbit model, the estimated human dose would be unacceptably high. This demonstrates how allometric scaling can be used in research projects to assess clinical feasibility. The impact of metabolism differences influencing the reliability of various species to predict for man was highlighted by DPC-A78445-00. DPC-A78445-00 is being developed as an alternative to exercise in myocardial perfusion imaging for the evaluation of coronary artery disease. DPC-A78445-00 was rapidly metabolized to the carboxylic acid by mouse and rat blood in vitro and in vivo, however longer stability was observed in the dog. In vitro human blood data was consistent with the dog, suggesting that mouse and rat would not be representative species. DPC-A78445-00 plasma protein binding was at a similar, moderate level in rat, dog and human plasma and metabolism by hepatocytes was similar in dog and human. Phase I human clinical trial testing determined the area under the blood concentration-time curve (AUC) and clearance predicted by the dog were within 32% of the human values. Overall, body weight based allometric scaling of pharmacokinetic parameters from animal models, when corrected for plasma protein binding, yielded reliable predictions of the human pharmacokinetics (within 50%) for radiopharmaceutical imaging agent. However, although predictive scaling from animal data can give insight into feasibility of compounds working in human, it is important to identify species differences with respect to metabolic stability. This allometric scaling method provides an additional tool to better predict doses in human for novel Medical Imaging agents.
| Identifer | oai:union.ndltd.org:wpi.edu/oai:digitalcommons.wpi.edu:etd-dissertations-1442 |
| Date | 10 January 2006 |
| Creators | Onthank, David C |
| Contributors | Joseph C. Bagshaw, Advisor, Hellen Vassallo, Committee Member, Simon Robinson, Committee Member, Robert Siegler, Committee Member |
| Publisher | Digital WPI |
| Source Sets | Worcester Polytechnic Institute |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Doctoral Dissertations (All Dissertations, All Years) |
Page generated in 0.0028 seconds