Kwan Cheol Pak (1,2), Min Kyu Heo (3), Su Jin Heo (4), Hyeong-Seok Lim (1,2)
(1) Department of Medical Science and Asan Medical Institute of Convergence Science and Technology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea, (2) Department of Clinical Pharmacology and Therapeutics and Asan Medical Institute of Convergence Science and Technology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea, (3) Genexine Inc., Seongnam, Republic of Korea, (4) Handok Inc., Seoul, Republic of Korea
Objectives: Adult growth hormone deficiency (AGHD) is known as clinical syndrome associated with a number of metabolic abnormalities, including abnormal body composition, reduced physical performance, altered lipid metabolism, decreased bone mass, increased insulin resistance, and reduced quality of life[1]. Most of the metabolic abnormalities associated with AGHD can be recovered by growth hormone (GH) replacement therapy. There is no doubt that daily subcutaneous (SC) administration of recombinant human growth hormone (rhGH) is recommended as traditional treatment of growth hormone deficiency (GHD). As the existing therapy for AGHD needs frequent administration of rhGH, it is uncomfortable for patients to adhere to the treatment which can lead to optimal clinical efficacy[2]. GX-H9, as long-acting preparation for improved compliance of treatment, is a recombinant human growth hormone (rhGH) fused to hybrid Fc (hyFc) which shows longer half-life than endogenous or first-generation growth hormone.
The major purposes of this study were followed as: 1) To describe the pharmacokinetic (PK) and pharmacodynamic (PD) profile of GX-H9; 2) To explore appropriate starting dose of GX-H9 and the criteria of dose increase/decrease of GX-H9 adopted in the simulation; 3) To predict the sampling time for IGF-1 SDS (Standardized Insulin-like Growth Factor 1 Score) close to average IGF-1 SDS after dosing of GX-H9 at steady state for the purpose of design of phase 3 clinical trial.
Methods: The serum GX-H9 and IGF-1 concentration data used in this modeling analysis were from the phase 1b and 2 clinical trial[3]. PK-PD modeling and simulation analyses were conducted using NONMEM® (version 7.4.3) subroutine ADVAN 9 and first-order conditional estimation with interaction (FOCE-I) method[4]. To describe the profile of PK-PD of GX-H9, various compartmental linear and nonlinear models and indirect response PD models were assessed based on statistical criteria, graphical goodness of fits. Through stepwise forward selection and backward elimination method, covariate analyses were conducted and inter-occasional variability (IOV) in the PK or PK-PD model was assessed.
IGF-1 SDS was simulated using the PK-PD model based on the predicted serum IGF-1 level, patient-specific age and sex by two groups consisting of older male (I) and younger male, and female who did not take estrogen (II). The target IGF-1 SDS range was -0.5 and 1.5 in a clinical study for Long-Lasting Human Growth Hormone in patients with AGHD[5]. The starting dose and the degree of dose adjustment were determined by simulations (1,000 replicates) where the IGF-1 SDS is expected to be within the target range in majority of patients. In another simulation study (1,000 replicates), we decided a single sampling time point for IGF-1 SDS which is the most predictive of average IGF-1 SDS at steady state.
Results: Two-compartmental linear model with mixture of structural model where majority of the parameters were described by mixture models in NONMEM best described the PK data. Exploratory plots for GX-H9 concentration suggested two, distinctive PK subpopulations. The PD model was implemented using an indirect response model where GX-H9 stimulates the production of IGF-1. PK-PD modeling was conducted sequentially using the individual empirical Bayes PK estimates.
The simulation suggests that the appropriate starting dose was 9.547 mg for group I and 16.46 mg for group II. The appropriate doses for increase and decrease during dose adjustment were respective 12.412 mg and 4.966 mg for group I and 14.322 mg and 5.730 mg for group II. The proper sampling times after the last dosing for the prediction of average IGF-1 SDS at steady state was 112 hour (90% PI, 58 – 169 hour) for group I and 116 hour (90% PI, 47 – 169 hour) for group II, respectively.
Conclusions: We successfully constructed the PK-PD model for GX-H9 and IGF-1 using phase 1b and 2 clinical trials. The sampling time for IGF-1 SDS will help predict the average IGF-1 SDS to adjust doses, making the further clinical trial efficient. Therefore, the clinical trial simulation using the PK-PD model will provide a lot of useful information for designing the future phase 3 clinical trial for GX-H9.
References:
[1] Hazem A, Elamin MB, Bancos I, et al. Body composition and quality of life in adults treated
with GH therapy: a systematic review and meta-analysis. Eur J Endocrinol. 2012;166(1):13-20.
[2] Reed ML, Merriam GR, Kargi AY. Adult growth hormone deficiency – benefits, side effects, and
risks of growth hormone replacement. Front Endocrinol (Lausanne). 2013;4:64.
[3] Ku CR, Brue T, Schilbach K, et al. Long-acting FC-fusion rhGH (GX-H9) shows potential for up to twice-monthly administration in GH-deficient adults. Eur J Endocrinol. 2018 Sep;179(3):169-179.
[4] Beal SL et al. 1989-2011. NONMEM Users Guides. Icon Development Solutions, Ellicott City, Maryland, USA.
[5] https://clinicaltrials.gov/ct2/show/NCT01225666
Reference: PAGE 29 (2021) Abstr 9689 [www.page-meeting.org/?abstract=9689]
Poster: Drug/Disease Modelling - Endocrine