|
|
|
|
|
|
|
IX. Population Toxicokinetics |
|
|
|
|
|
|
|
|
The use of pharmacokinetic screens and population approaches to measure kinetic parameters using minimal sampling techniques has many advantages in drug development [105,106] and has been advocated by the FDA to improve information from large scale clinical trials [107]. There has been some reluctance to wholeheartedly embrace this idea, since many companies are not sure if the benefits outweigh the investment in time, money, and personnel. Many medical directors see this as an additional burden increasing the clinical kinetic workload. Not only will it be necessary to undertake pair-wise kinetics comparisons in disease groupelderly, hepatic, renal disease, etc.but in addition, any new interactions that may be become apparent by the population analysis. However, this method can be used with all the advantages and few of the problems in preclinical toxicokinetics. In these safety studies, there is a need to measure exposure by estimating peak levels and the area under the curve. In dogs or primates, serial samples can be taken from the animals in the study, but this is not often possible from rodents. To overcome this difficulty, parallel or satellite groups are kept under the same conditions as the main study group but are only used for blood sampling and subsequent drug analysis. This considerably increases the number of animals and the number of kinetic analyses. Under these circumstances population kinetics can be used effectively [108] by taking only one sample from each of many animals but at different times (composite sampling), covering the whole of the dosing interval. These data can be analyzed together to define the average plasma level and the toxicokinetic profile for each dose with precision estimates by using population analytical methods [109,110]. These allow calculation of all measures of exposure without the inherent need to previously know exactly the profile or kinetics for each dosage in each species. The advantages are clear: in rats, there is no requirement for a satellite group, and in mice, the number of animals is greatly reduced. Similarly, the number of samples analyzed can be reduced 24-fold. In terms of management decisions, this approach is clearly cost effective, and should gain wider acceptance in the future. Indeed, minimal sampling techniques can be taken further to reduce the analytical load. We have found in our laboratories that, during carcinogenicity studies, when the drug is administered in the diet, only three samples are necessary to adequately describe the AUC with less than 10% error [111]. When the amount of food ingested by the animals over the previous 24 hours is taken as a correction factor, only one sample in the morning is necessary to obtain good estimates of exposure throughout the previous day. This method seems to have a wide applicability for a range of drugs with different half-lives [112]. Preliminary discussions with the Medicines Control Agency (MCA) in the UK would |
|
|
|
|
|