Omar Zaher 1, Dominik Selzer 1, Helena Loer 1, Frank Filip Steinbauer 1, Christiane Dings 1, Nicola Luigi Bragazzi 1, Felix Neuschwender 1, Nadja Walle 2, Adrian Doerr 2, Benjamin Peters 2, Matthias W.Laschke 3, Michael W.Menger 3, Peter H.Schmidt 2, Markus R.Meyer 4, Nadine Schaefer 2, Thorsten Lehr 1
1 Clinical Pharmacy, Saarland University (Saarbrücken, Germany), 2 Institute of Legal Medicine, Saarland University (Saabrücken, Germany), 3 Institute for Clinical and Experimental Surgery, Saarland University (Saarbrücken, Germany), 4 Department of Experimental and Clinical Toxicology and Pharmacology, Center for Molecular Signaling (PZMS) , PharmascienceHub (PSH) , Saarland University (Saarbrücken, Germany)
Introduction: New psychoactive substances (NPSs) have attracted significant attention from public health authorities due to their growing use as alternatives to conventional recreational drugs. Among these, synthetic cannabinoids (SCs) are of particular concern, having been linked to a growing number of severe adverse events and fatalities[1], [2]. The synthetic cannabinoid 5F-Cumyl-P7AICA (5FC), first reported by the European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) in 2015, belongs to the “new generation” of SCs featuring a 7-azaindole core and a carboxamide moiety. This structural class has raised considerable concern owing to increasing case reports of severe intoxication and deaths associated with their use[3]. Human clinical studies for high-risk molecules are ethically constrained, which highlights the need for alternative methodologies for human exposure assessment. Classical approaches such as allometric scaling and population pharmacokinetic (PopPK) modeling, although widely used[4], are fundamentally empirical and lack the capacity to mechanistically capture interspecies variability in tissue composition, enzyme expression, or metabolic routes, restricting their applicability to tissue-level predictions. Physiologically-based pharmacokinetic (PBPK) modeling provides a mechanistic framework that combines species-specific anatomical and physiological data with drug-specific physicochemical characteristics, allowing simulation of concentration-time profiles in individual tissues and facilitating extrapolation from preclinical species to humans[5].
Objective: This study aimed to develop a whole-body PBPK model for 5F-Cumyl-P7AICA (5FC) and its primary metabolite NPA in pigs, and to translate it to humans for predicting systemic and tissue-specific exposure. Human PBPK predictions were compared against allometrically scaled predictions from a previously established population toxicokinetic (PopTK) model[6] to evaluate the predictive performance of mechanistic PBPK modeling relative to conventional empirical extrapolation.
Methods: The pigs PBPK model was built utilizing the published porcine TK data for intravenous (IV) and inhalation routes using PK-Sim software (version 11.2; www.open-systems-pharmacology.org). 5FC and NPA metabolite serum, whole blood and fat tissue concentrations were used for model optimization and evaluation. Subsequently, the model was scaled to humans to predict systemic and tissue specific exposure.
Results: The porcine PBPK model adequately described the concentration-time profiles of 5FC, with MRD values of 1.40 for the IV route and 1.66 for the inhalation route. For the NPA metabolite, MRD values were 1.94 for the IV route and 2.02 for the inhalation route. Fat tissue concentrations of 5FC following inhalation were well captured (MRD = 1.74). Additionally, the model was able to predict different perimortem tissues concentrations adequately. When translated to humans, the PBPK model yielded higher systemic exposure estimates compared to allometric scaling, particularly for the inhalation route. Moreover, the human PBPK model allowed simulation of drug disposition across individual tissues, capturing organ-specific differences in exposure dynamics, including sustained accumulation in adipose tissue reflective of the lipophilic nature of 5FC.
Conclusion: The developed PBPK model demonstrates reliable predictive performance for 5FC and NPA in pigs and supports interspecies extrapolation to humans. This framework offers a practical tool for forensic toxicological applications, enabling tissue-specific exposure estimation of Synthetic Cannabinoids in humans.
References:
[1] I. C. Santos, D. Maia, R. J. Dinis-Oliveira, and D. J. Barbosa, “New Psychoactive Substances: Health and Legal Challenges,” Psychoactives, vol. 3, no. 2, Art. no. 2, Jun. 2024, doi: 10.3390/psychoactives3020018.
[2] A. Guirguis, “New psychoactive substance: a public health issue,” IJPP, vol. 5, no. 25, pp. 323–325, Sep. 2017, doi: doi.org/10.1111/ijpp.12313.
[3] EUDA, “EU Drug Market: New psychoactive substances — In-depth analysis,” Jun. 2024. [Online]. Available: https://www.euda.europa.eu/publications/eu-drug-markets/new-psychoactive-substances_en
[4] I. Mahmood and J. D. Balian, “Interspecies scaling: predicting clearance of drugs in humans. Three different approaches,” Xenobiotica, vol. 26, no. 9, pp. 887–895, Jan. 1996, doi: 10.3109/00498259609052491.
[5] N. E. Andrea, P. T. Frank, S. Walter, and W. Stefan, “Whole body physiologically-based pharmacokinetic models: their use in clinical drug development,” Expert opinion on drug metabolism and toxicology, vol. 4, no. 9, pp. 1143–1152, 2008, doi: https://doi.org/10.1517/17425255.4.9.1143.
[6] N. Walle et al., “Does a carboxamide moiety alter the toxicokinetics of synthetic cannabinoids? A study after pulmonary and intravenous administration of cumyl-5F-P7AICA to pigs,” Arch Toxicol, vol. 99, no. 2, pp. 633–643, Feb. 2025, doi: 10.1007/s00204-024-03906-z.
Reference: PAGE 34 (2026) Abstr 12303 [www.page-meeting.org/?abstract=12303]
Poster: Drug/Disease Modelling - Absorption & PBPK