Supplementary information
Population pharmacokinetics and pharmacodynamics of once-daily growth hormone Norditropin® in children and adults
Theodoros Papathanasiou1,*, Henrik Agersø1, Birgitte Bentz Damholt1,**, Michael Højby Rasmussen1, Rasmus Juul Kildemoes1
1. Global Development, Novo Nordisk A/S, 2860 Søborg, Denmark
*Affiliation at the time of the study and the manuscript development: Novo Nordisk A/S. Current affiliation: Department of Pharmacometrics, Novartis, Basel,
Switzerland
**Affiliation at the time of the study and the manuscript development: Novo Nordisk A/S. Current affiliation: Clinical Pharmacology, SNIPR BIOME, Copenhagen, Denmark
Corresponding author: Rasmus Juul Kildemoes, Global Development, Novo Nordisk A/S, Bagsvaerd, Denmark
Email: rvej@novonordisk.com
Journal: Clinical Pharmacokinetics
Supporting information
These supplementary data provide additional information regarding the analyses for growth hormone (GH) and insulin-like growth factor-I (IGF-I) serum concentrations, tables presenting the main model development steps for the PK and PK/PD models, and a summary of the PK and PK/PD model qualifications, including standard
goodness-of-fit plots, and visualizations of the main covariate relationships identified during model development (i.e. body weight).
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
Supplemental Tables
Table S1 Stepwise search for variability and covariate analysis for the base and final PK models, respectively
Random effect and covariate search parameters ΔOFV Random effect (IIV) search for the base model
IIV added on CL/F -544.3
IIV added on V/F -101.2
IIV added on GHBase -751.6
IIV added on Ka -11.3
Separate residual error adults / children -24.1
IIV correlation, Ka-CL and Ka-V -12.6
No further IIVs included. Base model obtained Covariate search for the final model
Body weight added on CL/F -16.3
Body weight added on Ka -5.8
Body weight added on GHBase -5.4
No further covariates fulfilled covariate selection criteria. Final model obtained ΔOFV, difference in objective function value; IIV, inter-individual variability; CL, growth hormone systemic clearance; CL/F, apparent clearance; V,volume of distribution; V/F, apparent volume of distribution; GHBase, baseline growth hormone concentration; IGF-I, insulin-like growth factor-I; Ka, absorption rate constant; PK, pharmacokinetic.
30 31 32 33 34
35 36 37 38
Table S2 Stepwise search for variability and covariate analysis for the base and final PK/PD models respectively
Random effect and covariate search parameters ΔOFV Random effect (IIV) search for the base model
IIV added on Kin -340.3
IIV added on Kout -108.2
IIV added on Emax -120.1
Separate residual error trial -22.1
No further IIVs included. Base model obtained Covariate search for the final model
Body weight added on Emax for children -21.1 No further covariates identified. Final model obtained
ΔOFV, difference in objective function value; IIV, inter-individual variability; Emax, maximum increase in IGF-I production rate; GHD, growth hormone deficiency; Kin, production rate of IGF-I; Kout, first order turn-over of IGF-I; PD, pharmacodynamic; PK, pharmacokinetic.
39 40 41
42 43 44
Supplemental Figures
Fig. S1 GH PK (a) and PD (b) profiles with the final model fit for multiple doses in adults with GHD for the fourth week of GH dosing
Data are presented as geometric mean, with 95% confidence intervals. Lines represent the geometric mean of the population model predictions.
GH, growth hormone; GHD, growth hormone deficiency; IGF-I, insulin-like growth factor-I; PD, pharmacodynamic; PK, pharmacokinetic.
45 46
47 48 49 50 51 52 53
Fig. S2 Standard goodness-of-fit plots for the final population PK model. (a) Observed GH concentrations versus population-predicted GH concentrations. (b) Conditional weighted residuals versus population-predicted GH concentrations. (c) Quantile-quantile (q-q) plot for the conditional weighted residuals. (d) Observed GH concentrations versus individual-predicted GH concentrations. (e) Conditional weighted residuals versus time after the first dose. (f) Distribution of conditional weighted residuals (orange) versus a standard normal distribution (blue)
GH, growth hormone; PK, pharmacokinetic.
54 55 56 57 58 59 60 61 62
Fig. S3 Identified relationship between body weight and (a) the absorption rate constant (Ka), (b) the apparent clearance (CL/F), and (c) the baseline GH
concentrations (PKBase) attributed to endogenous GH production, as estimated with the final PK model
The points represent individual post hoc parameter estimates for children (dark blue) and adults (light blue) with GHD. The lines represent the model predicted relationship for the population.
GH, growth hormone; GHD, growth hormone deficiency; PK, pharmacokinetic.
63 64 65 66 67 68 69 70 71
Fig. S4 Standard goodness-of-fit plots for the final population PK/PD model. (a) Observed IGF-I concentrations versus population-predicted IGF-I concentrations. (b) Conditional weighted residuals versus population-predicted IGF-I concentrations. (c) Quantile-quantile (q-q) plot for the conditional weighted residuals. (d) Observed IGF- I concentrations versus individual-predicted IGF-I concentrations. (e) Conditional weighted residuals versus time after the first dose. (f) Distribution of conditional weighted residuals (orange) versus a standard normal distribution (blue)
IGF-I, insulin-like growth factor-I; PD, pharmacodynamic; PK, pharmacokinetic.
72 73 74 75 76 77 78 79 80
Fig. S5 Identified relationship between body weight and the maximum increase in IGF-I production rate (Emax), as estimated with the final population PK/PD model The points represent individual post hoc parameter estimates for children (dark blue) and adults (light blue) with GHD. The lines represent the model predicted relationship for the population.
GHD, growth hormone deficiency; IGF-I, insulin-like growth factor-I; PD, pharmacodynamic;
PK, pharmacokinetic.
81 82 83 84 85 86 87 88