Chemicals, reagents, and preparations

A detailed list of the used chemicals and reagents as well as a description of the respective preparations (buffer solution, stock solutions, calibration standards, quality control samples as well as blank pig whole blood and serum) can be found in the Supplementary Information (SI).

In vivo studyAnimals

Analogous to previous studies (Walle et al. 2021, 2024), the in vivo experiments conducted in the present study were performed in accordance with the German legislation on protection of animals and the National Institutes of Health Guide for the Care and Use of Laboratory Animals (permission number: 44/2019).

Sixteen domestic male pigs of the Swabian Hall strain were used. The body weight (BW) of the pigs varied between 44 and 64 kg. In accordance to previous studies (Schaefer et al. 2018, 2019; Walle et al. 2021; Doerr et al. 2021; Nordmeier et al. 2022a, b), the animals had free access to tap water and daily standard chow up to 12 h before the start of the experiment. Then, they were kept fasting with still free access to water.

Surgical procedures

All surgical procedures were in accordance with already published previous studies (Schaefer et al. 2018, 2019, 2020; Walle et al. 2021; Doerr et al. 2021; Nordmeier et al. 2022a, b) and are described in detail in the SI.

Study design

The study included two different routes of administering the drug, i.v. or inhalative. First, six pigs received an i.v. dose of 200 µg/kg BW of cumyl-5F-P7AICA. For preparation of a solution with a concentration of 5 mg/mL, the SC was first diluted in ethanol. Following, to obtain the required dose of 200 µg/kg BW, the respective volume of the solution was withdrawn, fortified with 1 mL Polysorbat 80 for solubilization, and filled up with 0.9% sodium chloride to a final volume of 10 mL. Subsequently, the final solution was administered intravenously via the jugular vein over 30 s. Then, the venous catheter was washed for 30 s using 10 mL of 0.9% sodium chloride in order to remove possible retained substance (t = 0 min). After the washing step, blood samples were drawn 1, 2, 5, 10, 15, 30, 45, 60, 90, 120, 180, 240, 300, and 360 min after administration. Additionally, a control sample was taken before the administration of cumyl-5F-P7AICA.

Additional ten pigs received a 200 µg/kg BW dose of cumyl-5F-P7AICA via inhalative administration. For this purpose, cumyl-5F-P7AICA was initially dissolved in ethanol to obtain a stock solution of 5 mg/mL. The required volume to obtain a 200 µg/kg BW dose was diluted with ethanol to receive a final volume of 2 mL. The applied setup as well as the subsequent inhalative administration of cumyl-5F-P7AICA were in accordance to previous studies (Walle et al. 2021, 2024). Briefly, the prepared solution was nebulized and administered inhatively. For nebulization, the M-neb flow + ventilation ultrasonic nebulizer MN-300/7 (Nebutec, Elsenfeld, Germany) was used, applying the inspiration-triggered mode (< 0.2 mL/min). Blood samples were drawn prior to the administration as well as 1, 2, 5, 6, 7, 8, 9, 10, 15, 30, 45, 60, 90, 120, 180, 240, 300, and 360 min after the start of nebulization.

To obtain serum specimens, the blood specimens sampled during the experiment were centrifuged at 1476×g for 15 min. Blood and serum samples were stored at − 20 °C until analysis.

Sample preparation

For qualitative and quantitative determination of cumyl-5F-P7AICA and its N-pentanoic acid (NPA) metabolite in pig blood and serum specimens, a solid phase extraction using Strata C18 endcapped cartridges (200 mg/3 mL; Phenomenex LTD, Aschaffenburg, Germany) was performed, following the procedure previously successfully applied for other SCs by Schaefer et al. (2015, 2016, 2018). If measured concentrations were above the calibration range, samples were diluted 1:10 and analyzed again. A detailed description of the sample preparation and method validation can be found in the SI.

Liquid chromatography (LC)-quadrupole time of flight (TOF)–mass spectrometry (MS) apparatus

The settings of the LC-quadrupole TOF–MS used for detection and quantification of the substances in pig blood and serum samples were in accordance with a recent study (Walle et al. 2024) and can be found in detail in the SI.

Non-compartmental analysis

A non-compartmental analysis (NCA) was performed using the Software R (Version 4.3.0, The R Foundation for Statistical Computing, Vienna, Austria) and R package ‘PKNCA’ (Version 0.10.2) (Denney et al. 2015). Mean values and standard deviations (SD) were calculated using the available parent drug and metabolite concentration measurements in serum and whole blood. The areas under the curve (AUCs) were derived from the concentration–time profiles using the linear up/log down method.

Population TK modeling

A population (pop) TK model was developed using non-linear mixed-effects modelling techniques, facilitated by the software NONMEM (Version 7.4.3, ICON Development Solutions, Ellicott City, MD, USA). This approach enables the concurrent estimation of population medians for the model parameters alongside inter-individual (IIV) and residual variability. The model development process comprised three sequential phases: (I) Initial establishment of a TK model for cumyl-5F-P7AICA serum concentration after i.v. administration, involving exploration of various structural models (1-, 2-, 3- and 4 compartment models) and different elimination kinetics (i.e. linear and saturable processes); (II) subsequent integration of the metabolite formation into the parent model through an additional clearance rate from parent to metabolite, considering diverse structural models, metabolite formation and elimination kinetics; and (III) eventual extension of the model to incorporate parent and metabolite profiles following pulmonary administration by evaluating different absorption models. BW was incorporated as an exponential covariate on all clearance and volume of distribution parameters with an exponent of 0.75 to facilitate allometric scaling to human subjects (Schaefer et al. 2018).

For parameter estimation, the first-order conditional estimation algorithm with interaction was used. Model selection was based on visual inspection of goodness-of-fit plots (Karlsson and Savic 2007), precision of parameter estimates in the form of relative standard errors (Upton and Mould 2014), visual predictive checks (VPCs) and the objective function value (OFV) provided by NONMEM. Here, a nested model was considered superior if the difference of OFVs was > 3.84 points (chi2, p < 0.05, 1 df). For the VPC, 1000 simulations of the dataset were performed including random effects with the final model. Based on the simulation results, median serum concentration–time profiles and 90% prediction intervals were calculated and compared with the observed serum concentration. The software R (Version 4.3.0, The R Foundation for Statistical Computing) was used for the generation of the NONMEM dataset and graphics.

Prediction of human exposure

The final pig model was upscaled to humans, using a reference BW of 70 kg and following allometric principles (Schaefer et al. 2018). Simulated scenarios included single-dose administrations of 0.5, 2 and 14 mg as well as multiple dose application of 2 mg every 60 min, for both i.v. and pulmonary application and with a fixed inhalation duration of 10 min. Each scenario underwent 1000 simulations including random effects. Subsequently, median simulated serum concentration–time profiles were plotted along with their corresponding 90% prediction intervals.