Kamau Pierre 1, Katy Norman 1, Jason Mastaitis 1, Joe Grimsby 1, Jeanne Mendell 1, Lutz Harnisch 1, Alejandro Perez-Pitarch 1, Vincent Hurez 1
1 Regeneron Pharmaceuticals (Tarrytown, United States)
Introduction: Incretin-mimetic drugs (IMDs) such as glucagon-like peptide-1 (GLP-1) and gastric inhibitory peptide (GIP) receptor agonists have revolutionized obesity treatment, establishing new standards of care over traditional approaches. An emerging concern of these IMDs is the concomitant loss of skeletal muscle mass with body weight (BW) reduction, which can consequently compromise metabolic health[1]. Regeneron is currently developing novel monotherapies and combinations to improve the quality of weight loss and ultimately promote long-term health benefits in patients afflicted with obesity. Complete 26-week results from the Phase 2 COURAGE clinical trial in adult patients with obesity (NCT06299098) demonstrate that the addition of trevogrumab (anti-myostatin), with or without garetosmab (anti-activin A), to semaglutide preserved lean mass over monotherapy[2]. Employing a multiscale modeling approach, we have developed a comprehensive QSP platform that firstly characterizes overweight/obesity progression in adulthood, and consequently captures BW, body composition, and metabolic responses to lifestyle and pharmacological intervention strategies.
Methods: An 8-compartment QSP model was developed to mechanistically describe homeostatic metabolic processes spanning across multiple scales (molecular, cellular, tissue/organ, and whole-body) as well as their dynamic responses to external inputs such as diet, physical activity, and therapeutic drug administration. The model accounts for both the macronutrient intake of proteins, carbohydrates, and fats, and the core biochemical reactions that occur following their digestion, absorption, and systemic distribution. Modeled mechanisms include representations of triglyceride esterification, protein synthesis, adaptive thermogenesis, glycogenesis, fuel substrate oxidation, and adenosine triphosphate (ATP) dynamics. Constrained by the first law of thermodynamics, the described generation of ATP via macronutrient oxidation facilitated the coupling of mass and energy balances, which consequently allowed for simultaneous predictions of mass, metabolite concentrations, and energy expenditure trajectories. BW and its components, fat mass (FM) and lean mass (LM), are functions of glycogen, protein, fat, water, and bone mineral masses. Leveraging extensive in vitro and in vivo datasets, the model was calibrated to capture physiologically representative tissue concentrations of biomarkers (e.g., glucose, HbA1c, and triglycerides), metabolic and oxidation rates, body composition, and weight dynamics. Model interpolations of sex-specific datasets from healthy subjects and those with obesity at both baseline and following lifestyle perturbations were conducted. Utilizing published models of semaglutide’s pharmacokinetics[3] and its inhibitory pharmacodynamic effect on energy intake[4], the model was validated against STEP 1 clinical trial data[5]. The mechanisms of action of trevogrumab and garetosmab were also incorporated into the platform, and model outputs were calibrated to healthy volunteer Phase I data[6]. Responses of a Virtual Population (VPop), generated via an adaptation of the developed acceptance-rejection algorithm[7], were then used to predict the COURAGE trial outcomes on BW, FM, and LM changes.
Results: The metabolism platform well-characterized steady state/pre-intervention physiology, metabolic dysfunction driven by aging and chronic increased food intake, and obesity disease progression. Calibrated model outputs were consistent with experimental data, capturing a diversity of metabolic metrics such as adipose, liver, and muscle resting metabolic rates; macronutrient oxidation rates; respiratory quotients (RQ); and BW, FM, and LM dynamics across varying caloric intakes, diet compositions, and physical activity levels. The model’s predictive power was validated as it recapitulated key clinical outcomes of the STEP 1 and COURAGE trials. The simulated population level responses of the VPop, which reflected the distributions of COURAGE baseline demographics and clinical characteristics, were aligned with the reported changes across treatment arms. In silico combination of semaglutide with trevogrumab enhanced FM loss and prevented approximately 50% of semaglutide-induced loss of LM. Mean VPop percent changes (over baseline) for combination with 200 mg trevogrumab were -4.64% and -18.30% for LM and FM, respectively; an improvement over predicted -7.96% (LM) and -16.13% (FM) monotherapy responses.
Conclusions: Our QSP model describes a network of interconnected metabolic pathways that regulate mass and energy balances by integrating signals across various tissues and organs. Attributed to its mechanistic representations, the platform enables the elucidation of key biochemical drivers of metabolic dysfunction, the interpretation of treatment response heterogeneity, and the deconvolution of mechanisms driving response. This modeling framework, therefore, provides a useful tool for probing the therapeutic effects of a variety of assets, in mono- or combo-regimens, within Regeneron’s obesity portfolio. Future applications may include supporting clinical trial design optimization and informing patient population selection.
References:
1. Mechanick, J.I., et al., Strategies for minimizing muscle loss during use of incretin‐mimetic drugs for treatment of obesity. Obesity Reviews, 2025. 26(1): p. e13841.
2. Mosenzon, O., et al., Anti-Myostatin (trevogrumab) with and without anti-Activin a (garetosmab) preserves lean mass and increases loss of fat mass in people with obesity treated with semaglutide, in 61st EASD Annual Meeting. 2025.
3. Petri, K.C.C., et al., Semaglutide s.c. Once-Weekly in Type 2 Diabetes: A Population Pharmacokinetic Analysis. Diabetes Therapy, 2018. 9(4): p. 1533-1547.
4. Bosch, R., E.J.G. Sijbrands, and N. Snelder, Quantification of the effect of GLP‐1R agonists on body weight using in vitro efficacy information: An extension of the Hall body composition model. CPT: Pharmacometrics & Systems Pharmacology, 2024. 13(9): p. 1488-1502.
5. Wilding, J.P.H., et al., Weight regain and cardiometabolic effects after withdrawal of semaglutide: The STEP 1 trial extension. Diabetes, Obesity and Metabolism, 2022. 24(8): p. 1553-1564.
6. Trotter, D.G., et al., GDF8 and activin A are the key negative regulators of muscle mass in postmenopausal females: a randomized phase I trial. Nature Communications, 2025. 16(1): p. 4376.
7. Allen, R.J., T.R. Rieger, and C.J. Musante, Efficient Generation and Selection of Virtual Populations in Quantitative Systems Pharmacology Models. CPT: Pharmacometrics & Systems Pharmacology, 2016. 5(3): p. 140-146.
Reference: PAGE 34 (2026) Abstr 11901 [www.page-meeting.org/?abstract=11901]
Poster: Drug/Disease Modelling - Endocrine