Limits...
Using Bayesian-PBPK modeling for assessment of inter-individual variability and subgroup stratification.

Krauss M, Burghaus R, Lippert J, Niemi M, Neuvonen P, Schuppert A, Willmann S, Kuepfer L, Görlitz L - In Silico Pharmacol (2013)

Bottom Line: Inter-individual variability in clinical endpoints and occurrence of potentially severe adverse effects represent an enormous challenge in drug development at all phases of (pre-)clinical research.Moreover, clinically relevant homogeneous subpopulations can be mechanistically identified.The large scale PBPK model separates physiological and drug-specific knowledge which allows, in combination with Bayesian approaches, the iterative assessment of specific populations by integrating information from several drugs.

View Article: PubMed Central - PubMed

Affiliation: Bayer Technology Services GmbH, Computational Systems Biology, Leverkusen, 51368 Germany ; RWTH Aachen, Schinkelstr, Aachen Institute for Advanced Study in Computational Engineering Sciences, Aachen, 2, 52062 Germany.

ABSTRACT

Purpose: Inter-individual variability in clinical endpoints and occurrence of potentially severe adverse effects represent an enormous challenge in drug development at all phases of (pre-)clinical research. To ensure patient safety it is important to identify adverse events or critical subgroups within the population as early as possible. Hence, a comprehensive understanding of the processes governing pharmacokinetics and pharmacodynamics is of utmost importance. In this paper we combine Bayesian statistics with detailed mechanistic physiologically-based pharmacokinetic (PBPK) models. On the example of pravastatin we demonstrate that this combination provides a powerful tool to investigate inter-individual variability in groups of patients and to identify clinically relevant homogenous subgroups in an unsupervised approach. Since PBPK models allow the identification of physiological, drug-specific and genotype-specific knowledge separately, our approach supports knowledge-based extrapolation to other drugs or populations.

Methods: PBPK models are based on generic distribution models and extensive collections of physiological parameters and allow a mechanistic investigation of drug distribution and drug action. To systematically account for parameter variability within patient populations, a Bayesian-PBPK approach is developed rigorously quantifying the probability of a parameter given the amount of information contained in the measured data. Since these parameter distributions are high-dimensional, a Markov chain Monte Carlo algorithm is used, where the physiological and drug-specific parameters are considered in separate blocks.

Results: Considering pravastatin pharmacokinetics as an application example, Bayesian-PBPK is used to investigate inter-individual variability in a cohort of 10 patients. Correlation analyses infer structural information about the PBPK model. Moreover, homogeneous subpopulations are identified a posteriori by examining the parameter distributions, which can even be assigned to a polymorphism in the hepatic organ anion transporter OATP1B1.

Conclusions: The presented Bayesian-PBPK approach systematically characterizes inter-individual variability within a population by updating prior knowledge about physiological parameters with new experimental data. Moreover, clinically relevant homogeneous subpopulations can be mechanistically identified. The large scale PBPK model separates physiological and drug-specific knowledge which allows, in combination with Bayesian approaches, the iterative assessment of specific populations by integrating information from several drugs.

No MeSH data available.


Related in: MedlinePlus

Identification and assignment of patient subgroups by monitoring the logarithmic mean for each patient. A density estimation of the logarithmic mean values supported the identification of specific patient subgroups. The logarithmic mean values of the transporter activities for MRP2 and OATP1B1 were calculated from the subsample of the posterior and the kernel densities were quantified. Since the density for OATP1B1 provided the separation of the patient logarithmic mean values into two groups, single values were also plotted with symbols. Additionally, they were colored related to their specific genotype.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4230716&req=5

Fig7: Identification and assignment of patient subgroups by monitoring the logarithmic mean for each patient. A density estimation of the logarithmic mean values supported the identification of specific patient subgroups. The logarithmic mean values of the transporter activities for MRP2 and OATP1B1 were calculated from the subsample of the posterior and the kernel densities were quantified. Since the density for OATP1B1 provided the separation of the patient logarithmic mean values into two groups, single values were also plotted with symbols. Additionally, they were colored related to their specific genotype.

Mentions: First, we performed a Shapiro-Wilk test for normal distribution (Shapiro & Wilk, 1965) of the logarithmic mean values of the 200 samples of every patient, since protein expression has to be log-normally distributed in homogenous groups of patients (Sigal et al., 2006; Spencer et al., 2009). The results supported the hypothesis of lognormal distribution for MRP2 (p>0.75) and gave a strong indication of rejection of the hypothesis for OATP1B1 (p<0.1). Visual inspection of the estimated kernel densities (Bowman & Azzalini, 1997) of the logarithmic mean values (Figure 7) supported this, since two groups of patients were monitored for OATP1B1 but the density of MRP2 is clearly normally distributed. Thus, with regard to OATP1B1 the patient mean values were analyzed individually to examine which patient can be assessed to which group (Figure 7). A clear separation into two groups of four and six patients, respectively, was found. It should be noted that this separation of the OATP1B1 transporter activity was not an implicit property of the model structure but emerged as a result during the Bayesian-PBPK approach.Figure 7


Using Bayesian-PBPK modeling for assessment of inter-individual variability and subgroup stratification.

Krauss M, Burghaus R, Lippert J, Niemi M, Neuvonen P, Schuppert A, Willmann S, Kuepfer L, Görlitz L - In Silico Pharmacol (2013)

Identification and assignment of patient subgroups by monitoring the logarithmic mean for each patient. A density estimation of the logarithmic mean values supported the identification of specific patient subgroups. The logarithmic mean values of the transporter activities for MRP2 and OATP1B1 were calculated from the subsample of the posterior and the kernel densities were quantified. Since the density for OATP1B1 provided the separation of the patient logarithmic mean values into two groups, single values were also plotted with symbols. Additionally, they were colored related to their specific genotype.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4230716&req=5

Fig7: Identification and assignment of patient subgroups by monitoring the logarithmic mean for each patient. A density estimation of the logarithmic mean values supported the identification of specific patient subgroups. The logarithmic mean values of the transporter activities for MRP2 and OATP1B1 were calculated from the subsample of the posterior and the kernel densities were quantified. Since the density for OATP1B1 provided the separation of the patient logarithmic mean values into two groups, single values were also plotted with symbols. Additionally, they were colored related to their specific genotype.
Mentions: First, we performed a Shapiro-Wilk test for normal distribution (Shapiro & Wilk, 1965) of the logarithmic mean values of the 200 samples of every patient, since protein expression has to be log-normally distributed in homogenous groups of patients (Sigal et al., 2006; Spencer et al., 2009). The results supported the hypothesis of lognormal distribution for MRP2 (p>0.75) and gave a strong indication of rejection of the hypothesis for OATP1B1 (p<0.1). Visual inspection of the estimated kernel densities (Bowman & Azzalini, 1997) of the logarithmic mean values (Figure 7) supported this, since two groups of patients were monitored for OATP1B1 but the density of MRP2 is clearly normally distributed. Thus, with regard to OATP1B1 the patient mean values were analyzed individually to examine which patient can be assessed to which group (Figure 7). A clear separation into two groups of four and six patients, respectively, was found. It should be noted that this separation of the OATP1B1 transporter activity was not an implicit property of the model structure but emerged as a result during the Bayesian-PBPK approach.Figure 7

Bottom Line: Inter-individual variability in clinical endpoints and occurrence of potentially severe adverse effects represent an enormous challenge in drug development at all phases of (pre-)clinical research.Moreover, clinically relevant homogeneous subpopulations can be mechanistically identified.The large scale PBPK model separates physiological and drug-specific knowledge which allows, in combination with Bayesian approaches, the iterative assessment of specific populations by integrating information from several drugs.

View Article: PubMed Central - PubMed

Affiliation: Bayer Technology Services GmbH, Computational Systems Biology, Leverkusen, 51368 Germany ; RWTH Aachen, Schinkelstr, Aachen Institute for Advanced Study in Computational Engineering Sciences, Aachen, 2, 52062 Germany.

ABSTRACT

Purpose: Inter-individual variability in clinical endpoints and occurrence of potentially severe adverse effects represent an enormous challenge in drug development at all phases of (pre-)clinical research. To ensure patient safety it is important to identify adverse events or critical subgroups within the population as early as possible. Hence, a comprehensive understanding of the processes governing pharmacokinetics and pharmacodynamics is of utmost importance. In this paper we combine Bayesian statistics with detailed mechanistic physiologically-based pharmacokinetic (PBPK) models. On the example of pravastatin we demonstrate that this combination provides a powerful tool to investigate inter-individual variability in groups of patients and to identify clinically relevant homogenous subgroups in an unsupervised approach. Since PBPK models allow the identification of physiological, drug-specific and genotype-specific knowledge separately, our approach supports knowledge-based extrapolation to other drugs or populations.

Methods: PBPK models are based on generic distribution models and extensive collections of physiological parameters and allow a mechanistic investigation of drug distribution and drug action. To systematically account for parameter variability within patient populations, a Bayesian-PBPK approach is developed rigorously quantifying the probability of a parameter given the amount of information contained in the measured data. Since these parameter distributions are high-dimensional, a Markov chain Monte Carlo algorithm is used, where the physiological and drug-specific parameters are considered in separate blocks.

Results: Considering pravastatin pharmacokinetics as an application example, Bayesian-PBPK is used to investigate inter-individual variability in a cohort of 10 patients. Correlation analyses infer structural information about the PBPK model. Moreover, homogeneous subpopulations are identified a posteriori by examining the parameter distributions, which can even be assigned to a polymorphism in the hepatic organ anion transporter OATP1B1.

Conclusions: The presented Bayesian-PBPK approach systematically characterizes inter-individual variability within a population by updating prior knowledge about physiological parameters with new experimental data. Moreover, clinically relevant homogeneous subpopulations can be mechanistically identified. The large scale PBPK model separates physiological and drug-specific knowledge which allows, in combination with Bayesian approaches, the iterative assessment of specific populations by integrating information from several drugs.

No MeSH data available.


Related in: MedlinePlus