Abstract
An experimental design approach is presented for individualized therapy in the special case where the prior information is specified by a nonparametric (NP) population model. Here, a NP model refers to a discrete probability model characterized by a finite set of support points and their associated weights. An important question arises as to how to best design experiments for this type of model. Many experimental design methods are based on Fisher information or other approaches originally developed for parametric models. While such approaches have been used with some success across various applications, it is interesting to note that they largely fail to address the fundamentally discrete nature of the NP model. Specifically, the problem of identifying an individual from a NP prior is more naturally treated as a problem of classification, i.e., to find a support point that best matches the patient’s behavior. This paper studies the discrete nature of the NP experiment design problem from a classification point of view. Several new insights are provided including the use of Bayes Risk as an information measure, and new alternative methods for experiment design. One particular method, denoted as MMopt (multiple-model optimal), will be examined in detail and shown to require minimal computation while having distinct advantages compared to existing approaches. Several simulated examples, including a case study involving oral voriconazole in children, are given to demonstrate the usefulness of MMopt in pharmacokinetics applications.
Similar content being viewed by others
References
Andes D, Marchillo K, Stamstad T, Conklin R (2003) In vivo pharmacokinetics and pharmacodynamics of a new triazole, voriconazole, in a murine candidiasis model. Antimicrob Agents Chemother 47(10):3165–3169
Atkinson AC, Donev AN, Tobias RD (2007) Optimum experimental designs, with SAS. Oxford University Press, New York
Baek Y (2006) An interior point approach to constrained nonparametric mixture models. PhD Thesis, University of Washington
Bayard DS (1991) A forward method for optimal stochastic nonlinear and adaptive control. IEEE Trans Autom Control 36(9):1046–1053
Bayard DS, Schumitzky A (2010) Implicit dual control based on particle filtering and forward dynamic programming. Int J Adapt Control Signal Process 24(3):155–177
Bayard DS, Jelliffe R, Schumitzky A, Milman M, Van Guilder M (1995) Precision drug dosage regimens using multiple model adaptive control: theory and application to simulated vancomycin therapy. In: Selected topics in mathematical physics, Prof. R. Vasudevan Memorial volume. World Scientific Publishing Co., Madras
Bayard DS, Jelliffe R, Neely M (2013) Bayes risk as an alternative to Fisher information in determining experimental designs for nonparametric models. In: PODE 2013 workshop: population optimum design of experiments, Lilly, UK, 15 June 2013
Bayard DS, Jelliffe R, Neely M (2014) Experiment design based on Bayes risk and weighted Bayes risk with application to pharmacokinetic systems. In: PODE 2014 workshop: population optimum design of experiments, Roche, Basel, 11 September 2014
Bender CM, Orszag SA (1978) Advanced mathematical methods for scientists and engineers. McGraw-Hill Book Company, New York
BestDose software package. http://www.lapk.org. Accessed 1 Jan 2016
Blackmore L, Rajamanoharan S, Williams BC (2008) Active estimation for jump Markov linear systems. IEEE Trans Autom Control 53(10):2223–2236
Box REP, Lucas HL (1959) Design of experiments in non-linear situations. Biometrika 46:77–90
Chaloner K, Verdinelli I (1995) Bayesian experimental design: a review. Stat Sci 10(3):273–304
D’Argenio DZ (1981) Optimal sampling times for pharmacokinetic experiments. J Pharmacokinet Biopharm 9(6):739–756
D’Argenio DZ (1990) Incorporating prior parameter uncertainty in the design of sampling schedules for pharmacokinetic parameter estimation experiments. Math Biosci 99:105–118
D’Argenio DZ, Schumitzky A, Wang X (2009) ADAPT 5 user’s guide. Biomedical simulation resource. University of Southern California
Draper NR, Hunter WG (1967) The use of prior distributions in the design of experiments for parameter estimation in non-linear situations. Biometrika 54(1–2):147–153
Duda RO, Hart PE, Stork DG (2001) Pattern classification, 2nd edn. Wiley, New York
Fedorov VV (1972) Theory of optimal experiments. Academic, New York
Fedorov VV, Leonov SL (2013) Optimal design for nonlinear response models. CRC Press, New York
Gibaldi M, Perrier D (1982) Pharmacokinetics, 2nd edn. Marcel Dekker, Inc., New York
Goodwin GC, Payne RL (1977) Dynamic system identification: experiment design and data analysis. Academic, New York
Hennig S, Nyberg J, Fanta S, Backman JT, Hoppu K, Hooker AC, Karlsson MO (2012) Application of the optimal design approach to improve a pretransplant drug dose finding design for ciclosporin. J Clin Pharmacol 52(3):347–360. doi:10.1177/0091270010397731 (Epub 4 May 2011)
Hu C, Lovejoy WS, Shafer SL (1994) Comparison of some control strategies for three-compartment PK/PD models. J Pharmacokinet Biopharm 22(6):525–550
Huurneman LJ, Neely M, Veringa A, Prez FD, Ramos-Martin V, Tissing WJ, Alffenaar JWC, Hope W (2016) Pharmacodynamics of voriconazole in children: further steps along the path to true individualized therapy. Antimicrob Agents Chemother 60(4):2336–2342
Jelliffe R, Schumitzky A, Bayard D, Van Guilder M, Leary RH (2002) The USC*PACK programs for parametric and nonparametric population PK/PD modeling. Population Analysis Group in Europe, Paris
Karthaus M, Lehrnbecher T, Lipp HP, Kluge S, Buchheidt D (2015) Therapeutic drug monitoring in the treatment of invasive aspergillosis with voriconazole in cancer patients: an evidence-based approach. Ann Hematol 94(4):547–556
Katz D, D’Argenio DZ (1984) Discrete approximation of multivariate densities with application to Bayesian estimation. Comput Stat Data Anal 2(1):27–36
Kendall M, Stuart A (1979) The advanced theory of statistics, vol 2, 4th edn. Charles Griffin and Company, London
Kistos CP (2013) Optimal experimental design for non-linear models: theory and applications. Springer, New York
Landaw EM (1980) Optimal experiment design for biologic compartmental systems with applications to pharmacokinetics. PhD Dissertation, University of California, Los Angeles
Leary R, Jelliffe R, Schumitzky A, Van Guilder M (2001) An adaptive grid non- parametric approach to pharmacokinetic and dynamic (PK/PD) population models. In: Proceedings of the 14th IEEE symposium on computer-based medical systems, 2001, CBMS 2001. IEEE, p 389–394
Lindley DV (1972) Bayesian statistics, a review. Society for Industrial and Applied Mathematics (SIAM), Philadelphia
Lindsay B (1983) The geometry of mixture likelihoods: a general theory. Ann Stat 11:86–94
Luong ML, Al-Dabbagh M, Groll AH, Racil Z, Nannya Y, Mitsani D, Husain S (2016) Utility of voriconazole therapeutic drug monitoring: a meta-analysis. J Antimicrob Chemother 71(7):1786–1799
Mallet A (1986) A maximum likelihood estimation method for random coefficient regression models. Biometrika 73:645–656
Merle Y, Mentre F (1995) Bayesian design criteria: computation, comparison, and application to a pharmacokinetic and a pharmacodynamic model. J Pharmacokinet Biopharm 23(1):101–125
Neely M, Rushing T, Kovacs A, Jelliffe R, Hoffman J (2010) Voriconazole pharmacokinetics and pharmacodynamics in children. Clin Infect Dis 50(1):27–36
Neely M, Margol A, Fu X et al (2015) Achieving target voriconazole concentrations more accurately in children and adolescents. Antimicrob Agents Chemother 59:3090–3097
Pronzato L, Pazman A (2013) Design of experiments in nonlinear models: asymptotic normality, optimality criteria and small-sample properties. Springer, New York
Pronzato L, Walter E (1985) Robust experiment design via stochastic approximation. Math Biosci 75:103–120
Raiffa H, Schlaifer R (2000) Applied statistical decision theory. Wiley Classics Library, Wiley, New York
Ryan EG, Drovandi CC, McGree JM, Pettitt AN (2015) A review of modern computational algorithms for Bayesian optimal design. Int Stat Rev 84(1):128–154
Rodman JH, D’Argenio DZ, Peck CC (2004) Analysis of pharmacokinetic data for individualizing drug dosage regimens. In: Buron ME, Shaw LM, Schentag JJ, Evans WE (eds) Applied pharmacokinetics: principles of therapeutic drug monitoring, 4th edn. Applied Therapeutics, Spokane
Schumitzky A (1991a) Application of stochastic control theory to optimal design of dosage regimens. In: D’Argenio DZ (ed) Advanced methods of pharmacokinetic and pharmacodynamic systems analysis. Plenum Press, New York, pp 137–152
Schumitzky A (1991b) Nonparametric EM algorithms for estimating prior distributions. Appl Math Comput 45(2):143–157
Sheiner LB, Beal SL (1982) Bayesian individualization of pharmacokinetics: simple implementation and comparison with non-Bayesian methods. J Pharm Sci 71:1344–1348
Sheiner LB, Rosenberg B, Melmon KL (1972) Modeling of individual pharmacokinetics for computer-aided drug dosage. Comput Biomed Res 5:441–459
Sheiner LB, Halkin H, Peck C, Rosenberg B, Meimon KL (1975) Improved computer-assisted digoxin therapy: a method using feedback of measured serum digoxin concentrations. Ann Intern Med 82:619–627
Silvey SD (1980) Optimal design: an introduction to the theory for parameter estimation. Chapman and Hall, London
Tod M, Rocchisani J-M (1996) Implementation of OSPOP, an algorithm for the estimation of optimal sampling times in pharmacokinetics by the ED, EID and API criteria. Comput Methods Programs Biomed 50(1):13–22
Tod M, Rocchisani JM (1997) Comparison of ED, EID, and API criteria for the robust optimization of sampling times in pharmacokinetics. J Pharmacokinet Biopharm 25(4):515–537
Tse E, Bar-Shalom Y, Meier L III (1973) Wide-sense adaptive dual control for nonlinear stochastic systems. IEEE Trans Autom Control 18(2):98–108
Vajjah P, Duffull SB (2012) A generalisation of T-optimality for discriminating between competing models with an application to pharmacokinetic studies. Pharm Stat 11(6):503–510
Vozeh S, Steimer J-L (1985) Feedback control methods for drug dosage optimisation: concepts, classification and clinical application. Clin Pharmacokinet 10:457–476
Vozeh S, Berger M, Wenk M, Ritz R, Follath F (1984) Rapid prediction of individual dosage requirements for lignocaine. Clin Pharmacokinet 9:353–363
Walter E, Pronzato L (1987) Optimal experiment design for nonlinear models subject to large prior uncertainties. Am J Physiol Regul Integr Comp Physiol 253:R530–R534
Acknowledgements
Support from NIH Grants GM 068968 and HD 070886 is acknowledged.
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
Appendix: Two-support-point problem with close parameters
Appendix: Two-support-point problem with close parameters
Definition 1
(Two-support-point problem) A general class of two-support-point problems is defined of the form,
where \(a\in \Omega \buildrel \Delta \over =\{a_{1},\,a_{2}\}\) is a random variable taking on values \(a_{1}\) and \(a_{2};\,z(t,\,a_{i}),\, i=1,\,2\) are the support point responses, each assumed to be continuous over the closed interval \(t\in {\mathcal {T}}\buildrel \Delta \over =[t_{A},\,t_{B}];\,y(t_{k})\) is the noisy measurement of \(z(t,\,a)\) taken at discrete time \(t_{k};\,n_{k}\sim N(0,\,1)\) is the measurement noise at time instant \(t_{k};\,\sigma _{k}\) scales \(n_{k}\) to a desired level; and the prior support point probabilities are specified as \(p_{1}\) for \(a_{1}\) and \(p_{2}\) for \(a_{2}\) where \(p_{1}>0, \,p_{2}>0,\) and \(p_{1}+p_{2}=1.\)
The experiment design problem is to find the set U of n optimal sampling times
on a specified time interval \(t\in {\mathcal {T}}\buildrel \Delta \over =[t_{A},\,t_{B}]\) that act to best identify the parameter vector \(a\in \{a_{1},\,a_{2}\}.\)
Definition 2
(Restricted two-support-point problem) A restricted two-support-point problem is defined by Definition 1 under the additional conditions that the noise is independent of time, \(\sigma _{k}=\sigma ;\) there is only one sample to be taken, \(n=1,\,U=\{t_{1}\};\) and the prior support point probabilities are uniformly distributed, i.e., \(p_{1}=p_{2}=0.5.\) Furthermore, it is assumed that the function \(\mu (t,\,a)\) is sufficiently smooth to admit a first-order Taylor expansion in the vicinity of parameter values \(a_{1}\) and \(a_{2}\) for all \(t\in {\mathcal {T}}.\)
In what follows, the asymptotic order notation \({\mathcal {O}}(\epsilon ^{k})\) is used to indicate a term that decreases to zero as \(\epsilon ^{k}.\) Specifically, \(\lim _{\epsilon \rightarrow 0}{\mathcal {O}}(\epsilon ^{k})=0\) and there exists a constant M such that [9],
Furthermore, the notational dependence of \(\mu (t,\,a)\) on time t will be suppressed and denoted more simply as \(\mu (a).\)
Theorem 2
The costs associated with the ED, D [evaluated on the mean parameter value \({\overline{a}}={1\over 2}(a_{1}+a_{2})],\) EID and ELD optimal designs for the restricted two-support-point problem of Definition 2 can be written in terms of the response separation \(r(t)=|\mu (a_{2})-\mu (a_{1})|\) for sufficiently small parameter uncertainty \(\Delta a=a_{2}-a_{1}\) as,
Proof of Eq. ( 61 ): Consider the response \(\mu (a)\) as a function of the parameter a. Let \(\mu (a_{2})\) be expressed in terms of the Taylor expansion about \(a_{1}\)
Letting \(\Delta a\buildrel \Delta \over =a_{2}-a_{1}\) yields upon rearranging,
Squaring both sides of (66) gives
Similarly, let \(\mu (a_{1})\) be expressed in terms of the Taylor expansion about \(a_{2}\)
which can be rearranged as
Squaring both sides of (69) yields
Adding one half of (67) to one half of (70) gives
Dividing both sides by \(\sigma ^{2}(\Delta a)^{2}\) gives,
Proof of Eq. ( 62 ): Combining relations \({\overline{a}}={1\over 2}(a_{1}+a_{2})\) and \(\Delta a=a_{2}-a_{1},\) gives
Let \(\mu (a_{2})\) be expressed in terms of a Taylor expansion of \(\mu (a)\) about \({\overline{a}}\)
Rearranging (74) gives \(a_{2}-{\overline{a}}={1\over 2}\Delta a\) which is substituted into (75) to give
Similarly, let \(\mu (a_{1})\) be expressed in terms of a Taylor expansion of \(\mu (a)\) about \({\overline{a}}\)
Rearranging (73) gives \(a_{1}-{\overline{a}}={-}{1\over 2}\Delta a\) which is substituted into (77) to give
Subtracting (78) from (76) and rearranging gives
Squaring both sides of (79) and dividing by \(\sigma ^{2}(\Delta a)^{2}\) gives upon rearranging,
Proof of Eq. ( 63 ): Reciprocating both sides of Eq. (67), using the relation \(\frac{1}{1+\epsilon }= 1-\epsilon +{\mathcal {O}}(\epsilon ^{2})\) for small \(\epsilon ,\) and rearranging gives,
Likewise, reciprocating both sides of Eq. (70) and rearranging gives,
Adding (81) and (82) and multiplying both sides by \(\sigma ^{2}/2\) gives upon rearranging
Proof of Eq. ( 64 ): Multiplying (67) and (70) yields,
Dividing both sides by \(\sigma ^{4}\Delta a^{4}\) gives upon rearranging
Taking \({1\over 2}\ln (\cdot )\) of both sides gives
Corollary 1
As the parameter uncertainty \(\Delta a=a_{2}-a_{1}\) becomes small in the restricted two-support-point problem of Definition 2, the objective functions \(J_{ED},\,J_{D}\) and \(J_{ELD}\) approach a monotonically increasing function of the response separation r(t), while \(J_{EID}\) approaches a monotonically decreasing function of the response separation r(t).
Proof
The result follows by the properties of relations (61), (62), (63), (64) as \(\Delta a\) becomes small, noting that functions 1 / x and \(\ln (x)\) are monotonically decreasing and increasing in x, respectively. \(\square\)
Corollary 1 indicates that as the parameter uncertainty \(\Delta a=a_{2}-a_{1}\) becomes small in the restricted two-support-point problem of Definition 2, the objectives of maximizing \(J_{ED},\,J_{D},\,J_{ELD}\) and minimizing \(J_{EID}\) asymptotically approach the single common objective of maximizing the response separation \(r(t)=|\mu (a_{2},\,t)-\mu (a_{1},\,t)|\) over time \(t\in {\mathcal {T}}.\) These asymptotic properties are examined numerically in the “Two-support-point example” section.
Rights and permissions
About this article
Cite this article
Bayard, D.S., Neely, M. Experiment design for nonparametric models based on minimizing Bayes Risk: application to voriconazole\(^{1}\) . J Pharmacokinet Pharmacodyn 44, 95–111 (2017). https://doi.org/10.1007/s10928-016-9498-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10928-016-9498-5