Jiyoung Seo1,2, Hyeong-Seok Lim1,2, SeungChan Choi1,2, Hyeeun Kim3, Na-Young Kim3
1Department of Clinical Pharmacology and Therapeutics, Asan Medical Center, University of Ulsan College of Medicine, 2Department of Medical Science and Asan Medical Institute of Convergence Science and Technology, Asan Medical Center, University of Ulsan College of Medicine, 3Hanmi Pharmaceutical Co. Ltd.
Introduction: Obesity has become an increasingly prevalent condition in modern society, significantly impacting both health and overall quality of life. Among the most widely used pharmacological interventions are glucagon-like peptide-1 (GLP-1) receptor agonists, which have demonstrated substantial efficacy in promoting weight loss. However, despite their therapeutic benefits, clinical trials have consistently reported gastrointestinal adverse effects associated with this drug class, including nausea, vomiting, diarrhea, and constipation. Consequently, a comprehensive safety evaluation is crucial in the development of novel GLP-1 receptor agonists to ensure both efficacy and tolerability. Objectives: Efpeglenatide is a novel GLP-1 receptor agonist administered as a once-weekly subcutaneous injection, with demonstrated efficacy in reducing glycated hemoglobin levels and body weight across multiple clinical trials. To assess the safety profile of efpeglenatide, we conducted a pharmacokinetic and pharmacodynamic analysis of adverse events using pooled data from six clinical trials. Given that nausea and vomiting are among the most frequently reported adverse effects of GLP-1 receptor agonists, our analysis specifically focused on characterizing the impact of efpeglenatide on these gastrointestinal symptoms. Method: This analysis incorporates pharmacokinetic data and nausea/vomiting events from six clinical trials, including five conducted in diabetic patients and one in obese patients. Repeated time-to-event modeling was performed using NONMEM (version 7.5) to evaluate nausea and vomiting in a total of 858 participants, including 295 obese individuals. The final model employed an exponential model with multiple covariates. Two distinct models were developed: one incorporating data from all six studies and another exclusively utilizing data from the obesity-focused trial (Study 205), the latter of which was used for predicting nausea and vomiting in obese patients. Among the 678 subjects analyzed, 247 (36%) reported at least one occurrence of nausea or vomiting. Results: The model demonstrated a reasonable fit to the observed data for both the overall patient population and the obese subgroup. Given that Efpeglenatide is intended for obesity treatment, this abstract focuses solely on the obesity-specific model. The concentration-dependent adverse event rate was best described by a log-linear model. The estimated first-order rate constant governing the decline of the non-drug effect over time was 0.0208 hr?¹, corresponding to a half-life of 33.32 hours (approximately 1.5 days). This indicates that the non-drug effect diminishes by half within this period and returns to the baseline hazard approximately 7 days after treatment initiation. The rate of change in drug effect on the log-hazard (?_slope) was estimated at 0.627 1/(hr·ng/mL), with an intercept (?_intercept) of 0.634 ng/mL. This indicates that for every 1 ng/mL increase in the logarithm of the sum of drug concentration and the intercept, the log-hazard ratio increases by 0.627 1/(hr·ng/mL). The intercept, being less than 1, suggests a fixed reduction in log-hazard independent of drug concentration when efpeglenatide is absent. [?_slope*log?(?_intercept )=-0.286] Considering both the intercept and drug concentration, the equivalent drug concentration required to achieve a 1 ng/mL change in log concentration was estimated at 2.986 ng/mL. [e^1- ?_intercept=2.986 ng/mL] This implies that an increase in efpeglenatide concentration by 2.986 ng/mL leads to a 0.627 1/(hr·ng/mL) increase in the log-hazard ratio. To characterize tolerance, an Emax model was applied, using the cumulative area under the concentration-time curve (AUC) as the driving factor. The estimated TC50 (AUC at which 50% tolerance is established) was 109,872 ng·h/mL. At a weekly initial dose without titration, the time required to reach TC50 was approximately four weeks, indicating rapid tolerance development following initial administration. The maximum tolerance effect was estimated to be 0.996, a value close to 1, which suggests that once the maximum tolerance is reached, the drug effect to log hazard is almost completely suppressed and converges to zero. The visual predictive check (VPC) for the obese subject model shows a sharp increase in nausea and vomiting within the first seven days after drug administration and slight increase thereafter, attributed to both the drug effect and a temporary non-drug effect lasting about a week. The model-estimated TC50 value indicated that 50% tolerance to nausea and vomiting was achieved after approximately four weeks with a weekly initial dose of efpeglenatide. Given that the HM-EXC-205 study utilized higher dosing exposure than a weekly initial dosing, TC50 was likely reached earlier than with the weekly initial dose. By twelve weeks, 90% tolerance (TC90) is achieved at the higher doses, leading to a further reduction in nausea and vomiting. Beyond four weeks, drug concentrations stabilize, and the log-linear model predicts minimal further increases in drug-related effects. Consequently, adverse events become rare, aligning with the plateau observed in the cumulative probability curve. In summary, nausea and vomiting peak in the first week, decline significantly after four weeks due to rapid tolerance development, and become infrequent after twelve weeks. Conclusion: The repeated time to event model for nausea and vomiting described and predicted adverse events of efpeglenatide in patients with type 2 diabetes and obesity successfully. The model predicted that efpeglenatide rapidly increases the incidence of nausea and vomiting early after the first dose through approximately four weeks after which tolerance is rapidly established with significant drop in the frequency of adverse events. A simulation study showed that administering an initial dose for four weeks is sufficient to achieve 50% tolerance, which occurs when the cumulative AUC reaches 109,872 hr·ng/mL or higher. Increasing the dose within the first four weeks, rather than maintaining a weekly dose of initial dose or more, did not significantly raise the risk of adverse events after four weeks, with similar outcomes observed across different dosing regimens. Across all simulations, tolerance was almost fully established by twelve weeks, after which the incidence of nausea and vomiting became negligible. The results of this modeling study provide useful information about the relationship between PK and major advance event of nausea / vomiting following efpeglenatide. This information will help determine the appropriate efpeglenatide dose regimens with maximal effect and minimal adverse event.
1. Beal SL, Boeckmann AJ, Sheiner LB. NONMEM Users Guides. 4th ed2017 2017/04. 2. Pharmacokinetic and Pharmacodynamic Data Analysis Concepts and Applications (5th Edition)_by Gabrielsson p.221. 3. Cheng Cai Tang et al. The Journal of Clinical Pharmacology 2024, 64(2) 215–226. 4. Ronald Niebecker et a. CPT Pharmacometrics Syst. Pharmacol. (2019) 8, 230–239;
Reference: PAGE 33 (2025) Abstr 11674 [www.page-meeting.org/?abstract=11674]
Poster: Drug/Disease Modelling - Safety