The application of population approaches to drug development is recommended in several US Food and Drug Administration (FDA) guidance documents. Population pharmacokinetic (and pharmacodynamic) techniques enable identification of the sources of inter- and intra-individual variability that impinge upon drug safety and efficacy. This article briefly discusses the 2-stage approach to the estimation of population pharmacokinetic parameters, which requires serial multiple measurements on each participant, and comprehensively reviews the nonlinear mixed-effects modelling approach, which can be applied in situations where extensive sampling is not done on all or any of the participants.
Certain preliminary information, such as the compartment model used in describing the pharmacokinetics of the drug, is required for a population pharmacokinetic study. The practical design considerations of the location of sampling times, number of samples/participants and the need to sample an individual more than once should be borne in mind. Simulation may be useful for choosing the study design that will best meet study objectives.
The objectives of the population pharmacokinetic study can be secondary to the objectives of the primary clinical study (in which case an add-on population pharmacokinetic protocol may be needed) or primary (when a stand-alone protocol is required). Having protocols for population pharmacokinetic studies is an integral part of ‘good pharmacometric practice’.
Real-time data assembly and analysis permit an ongoing evaluation of site compliance with the study protocol and provide the opportunity to correct violations of study procedures. Adequate policies and procedures should be in place for study blind maintenance. Real-time data assembly creates the opportunity for detecting and correcting errors in concentration-time data, drug administration history and covariate data.
Population pharmacokinetic analyses may be undertaken in 3 interwoven steps: exploratory data analysis, model development and model validation (i.e. predictive performance). Documentation for regulatory purposes should include a complete inventory of key runs in the analyses undertaken (with flow diagrams if possible), accompanied by articulation of objectives, assumptions and hypotheses. Use of diagnostic analyses of goodness of fit as evidence of reliability of results is advised. Finally, the use of stability testing or model validation may be warranted to support label claims.
The opinions expressed in this article were revised by incorporating comments from various sources and published by the FDA as ‘Guidance for Industry: Population Pharmacokinetics’ (see the FDA home page http://www.fda.gov for further information).
KeywordsAdis International Limited Exploratory Data Analysis Population Pharmacokinetic Model Population Pharmacokinetic Study Standardise Prediction Error
Unable to display preview. Download preview PDF.
- 4.Steimer JL, Vozeh S, Racine-Poon A, et al. The population approach: rationale, methods, and applications in clinical pharmacology and drug development [chapter 15]. In: Welling PG, Balant LP, editors. Pharmacokinetics of drugs (Handbook of experimental pharmacology). Berlin-Heidelberg: Springer-Verlag, 1994:404–51.Google Scholar
- 6.Ette EI, Miller R, Gillespie WR, et al. The population approach: FDA experience. In: Aarons L, Balant LP, Danhof M, et al., editors. The population approach: measuring and managing variability in response, concentration and dose. Brussels: Commission of the European Communities, European Cooperation in the Field of Scientific and Technical Research, 1997: 271–5.Google Scholar
- 11.Prevost G. Estimation of a normal probability density function from samples measured with non-negligible and non-constant dispersion [in French]. Internal Report 6–77, Adersa-Gerbios, 2 avenue du 1er mai, F-91120 Palaiseau, France. Palaiseau: Adersa-Gerbios, 1984.Google Scholar
- 12.Racine-Poon A, Smith AMF. Population models. In: Berry DA, editors. Statistical methodology in pharmaceutical sciences. New York: Dekker, 1990: 139–62.Google Scholar
- 13.Beal SL, Sheiner LB. Estimating population pharmacokinetics. CRC Critical Rev Biomed Eng 1982; 8: 195–222.Google Scholar
- 16.Phases of an investigation. Code of Federal Regulations. 21 Pt 312. Section 21.Google Scholar
- 19.E7 studies in support of special populations: geriatrics. ICH Guidance. Bethesda (MD): Federal Register. National Archives and Records Administration; 1994 Aug 2 (59 FR 39398).Google Scholar
- 20.Steimer JL, Mentre F, Mallet A. Population studies for evaluation of pharmacokinetic variability: why? how? when? In: Aiache JM, Hirtz J, editors. Experimental pharmacokinetics. 2nd European Congress on Biopharmaceutics and Pharmacokinetics. Vol. 2. Paris: Lavoisier, 1996: 40–9.Google Scholar
- 25.Ette EI, Sun H, Ludden TM. Design of population pharmacokinetic studies. Proceedings of the American Statistics Association (Biopharmaceutics Section); 1994: 487–92.Google Scholar
- 27.Johnson NE, Wade JR, Karlsson MO. Comparison of some practical sampling strategies for population pharmacokinetic studies. J Pharmacokinet Biopharm 1996; 24(6): 245–72.Google Scholar
- 28.Sun H, Ette EI, Ludden TM. On error in the recording of sampling times and parameter estimation from repeated measures pharmacokinetic data. J Pharmacokinet Biopharm 1996; 24(6): 635–48.Google Scholar
- 30.Hale M, Gillespie WR, Gupt SK, et al. Clinical simulation: streamlining your drug development process. Appl Clin Trials 1996; 5: 35–40.Google Scholar
- 33.Grasela TH, Antal EJ, Fiedler-Kelley J, et al. An automated drug concentration screening and quality assurance program for clinical trials. Drug Info J. In press.Google Scholar
- 34.Fiedler-Kelly JD, Foit DJ, Knuth DW, et al. Development of a real-time, therapeutic drug monitoring system, delavardine registration trials. Pharm Res 1996: 13 Suppl.: S454.Google Scholar
- 35.Rombout F. Good pharmacokinetic practice (GPP) and logistics a continuing challenge. In: Aarons L, Balant LP, Danhof M, et al., editors. The population approach: measuring and managing variability in response, concentration and dose. Brussels: Commission of the European Communities, European Cooperation in the Field of Scientific and Technical Research, 1997: 183–93.Google Scholar
- 37.Donner A. The relative effectiveness of procedures commonly used in multiple regression analysis for dealing with missing values. Am Stat 1982; 36: 378–81.Google Scholar
- 47.Peck C. Population approach in pharmacokinetics and pharmacodynamic: FDA view. Proceedings of the COST B1 Conference; 1991; Manchester. Brussels: European Commission, 1992: 157–68.Google Scholar
- 49.Mandema JW, Verotta D, Sheiner LB. Building population pharmacokinetic-pharmacodynamic models. In: D’Argenio DZ, editor. Advanced pharmacokinetic and pharmacodynamic systems analysis. New York: Plenum Press, 1995: 69–86.Google Scholar
- 55.Mentre F, Ebelin ME. Validation of population pharmacokinetic/pharmacodynamic analyses: review of proposed approaches. In: Aarons L, Balant LP, Danhof M, et al., editors. The population approach: measuring and managing variability in response, concentration and dose. Brussels: Commission of the European Communities, European Cooperation in the Field of Scientific and Technical Research, 1997: 147–60.Google Scholar
- 58.Beal SL. Validation of a population model. E-mail to NONMEM usersnet participants. 1994 Feb 2 (http://www.phar.com/nonmeml/nmo/topic006.html).
- 60.Gelman A, Carlin JB, Stern HS, et al. Bayesian data analysis. New York: Chapman and Hall, 1995.Google Scholar