A two-step deconvolution-analysis-informed population pharmacodynamic modeling approach for drugs targeting pulsatile endogenous compounds

Abstract

Pharmacodynamic modeling of pulsatile endogenous compounds (e.g. growth hormone [GH]) is currently limited to the identification of a low number of pulses. Commonly used pharmacodynamic models are not able to capture the complexity of pulsatile secretion and therefore non-compartmental analyses are performed to extract summary statistics (mean, AUC, Cmax). The aim of this study was to develop a new quantification method that deals with highly variable pulsatile data by using a deconvolution-analysis-informed population pharmacodynamic modeling approach. Pulse frequency and pulse times were obtained by deconvolution analysis of 24 h GH profiles. The estimated pulse times then informed a non-linear mixed effects population pharmacodynamic model in NONMEM V7.3. The population parameter estimates were used to perform simulations that show agonistic and antagonistic drug effects on the secretion of GH. Additionally, a clinical trial simulation shows the application of this method in the quantification of a hypothetical drug effect that inhibits GH secretion. The GH profiles were modeled using a turnover compartment in which the baseline secretion, kout, pulse secretion width, amount at time point 0 and pulse amplitude were estimated as population parameters. Population parameters were estimated with low relative standard errors (ranging from 2 to 5%). Total body water (%) was identified as a covariate for pulse amplitude, baseline secretion and the pulse secretion width following a power relationship. Simulations visualized multiple gradients of a hypothetical drug that influenced the endogenous secretion of GH. The established model was able to fit and quantify the highly variable individual 24 h GH profiles over time. This pharmacodynamic model can be used to quantify drug effects that target other endogenous pulsatile compounds.Electronic supplementary materialThe online version of this article (doi:10.1007/s10928-017-9526-0) contains supplementary material, which is available to authorized users.

Modeling Associations Between Latent Event Processes Governing Time Series of Pulsing Hormones

“…The deconvolution model has been applied to these hormones (Johnson, ; [van Esdonk et al. ], ). Thus, this work is more broadly applicable for investigating association hypotheses in these axes as well.…”

The power of modelling pulsatile profiles

“…Based on the used parameters, an individual growth hormone concentration–time profile showed a mean of 7 pulses (range 0–15) during a 14 h timeframe and was highly variable across individuals. The simulated pulse times were included in the dataset as separate columns to be used in NONMEM V7.3 for simulation and parameter estimation [ 13 ].…”

“…Having quantified the pulse times a priori reduces the degrees of freedom in the model and enables the quantification of only the remaining parameters required to fit a pulsatile profile, such as the pulse amplitudes, pulse secretion width, and a continuous zero-order (non-pulsatile) endogenous release. Previous publications have shown that this methodology provides an accurate description of the pulsatile data and is capable of quantifying drug effects on highly variable datasets [ 13 , 14 ]. Furthermore, this deconvolution-analysis-informed modelling approach can be used for more realistic clinical trial simulations that mimic the endogenous secretion pattern, in which the high variability from different sources (differences in pulse times, variability in pulse amplitude, variability in the pharmacokinetics, etc.)…”

Quantification of the endogenous growth hormone and prolactin lowering effects of a somatostatin-dopamine chimera using population PK/PD modeling

“…The pulse frequency, obtained from the deconvolution analysis, of BIM23B065 treated individuals was analyzed for significance (generalized linear model with Poisson distribution, p \ 0.05) compared with the placebo cohort. The pulse frequency and the location of pulses from the deconvolution analysis were converted to a format suitable for population NLME modelling in NONMEM [16,17].…”

“…Interindividual variability (IIV) in the population parameters and between-occasion variability (BOV) between day 7 and day 12 was included following a bottom-up inclusion procedure. Then, the estimated population parameters and variance distributions were fixed to the placebo estimates and model development continued with the full dataset containing both placebo and BIM23B065 treated individuals [17]. Multiple PK/PD relationships, linear and (sigmoidal) maximal effect (E MAX ), driven by the plasma PK of BIM23B065 or via an effect compartment, were tested for significance on the baseline secretion and pulse amplitude parameters during model development [18].…”