Jakob Kolar1, Igor Locatelli1, Iztok Grabnar1
1University of Ljubljana, Faculty of Pharmacy
Objectives The phosphodiesterase type-5 inhibitor sildenafil is widely used for the treatment of erectile dysfunction and pulmonary hypertension. As it crosses the blood-brain barrier (BBB), it may have neuroprotective and neurorestorative effects and is therefore a potential drug for central nervous system disorders (1). In addition, it is a repurposing drug candidate for the treatment of Alzheimer’s disease as it reduces tau phosphorylation and has protective effects in iPSC derived neurons of Alzheimer’s patients (2). Our aim was to establish and validate a physiologically based pharmacokinetic (PBPK) model for sildenafil and its main active metabolite N-desmethyl sildenafil (DMS) (3). The purpose of the model was to predict the unbound sildenafil concentrations in the interstitial fluid of the adult brain so that it can be used for dose optimisation in further sildenafil repurposing studies. Methods The PBPK model was developed in PK-Sim, version 12.0 (Open Systems Pharmacology). Twenty clinical studies (dosage range 10 mg to 200 mg) with sildenafil plasma measurements were used for modelling sildenafil, four of which were used for sildenafil model building. Fifteen of these studies included DMS plasma measurements, three of which were used for DMS model building. Virtual individuals and populations with RT PCR expressions and administration protocols were created based on reported clinical studies. A Weibull dissolution with a dissolution time of 5 minutes and a dissolution shape of 0.92 was assumed for tablet formulations. The model included sildenafil as a substrate for CYP 3A4, CYP 2C9 and the P glycoprotein as well as DMS as a substrate for CYP 3A4. The model was validated using the mean relative deviation (MRD) of all predicted plasma concentrations and the geometric mean fold error (GMFE) of all predicted areas under the curve from time 0 to infinity (AUC0?8) and maximum plasma concentrations (Cmax) for sildenafil and DMS. Values = 2 were considered indicative of an adequate model. Furthermore, the ratios between the unbound interstitial concentration in the brain (Cbrain interstitial, unbound) and the unbound plasma concentration (Cplasma, unbound) were determined at different time points after administration (0.5 h, 1.0 h, 1.5 h, 2.0 h and 2.5 h) for each study including oral administration of 25 mg to 200 mg sildenafil. The mean ratio was calculated and compared with the mean ratio observed in M. fascicularis (4). The ratios were validated using the MRD of all predicted ratios. Results The optimised lipophilicity values were 3.03 and 2.45 log units for sildenafil and DMS, respectively. The blood-to-plasma ratio for sildenafil was set at 0.64 (5). The model assumed that sildenafil is mainly metabolised to DMS, therefore sink metabolites were not considered in the model. The values of the catalytic constants of CYP 3A4 and CYP 2C9 were optimised to 34.44 min-1 or set to 0.72 min-1 (3) , respectively, and the Michaelis Menten constants (Km) were set to 15.0 µM (6) and 27.0 µM (3), respectively. The maximum transport rate of P glycoprotein was estimated at 114 pmol/min/pmol (7, 8) with a Km value of 23.2 µM (7) and a reference concentration of 0.077 µM (multiplied by a factor of 3.57 in the intestinal mucosa) (9). The optimised specific clearance of DMS by CYP 3A4 was 4.09 min-1. The performance of the model was confirmed as the values for MRD, AUC0?8 and Cmax GMFE were = 2, namely 1.89, 1.22 and 1.47 for sildenafil and 1.99, 1.35 and 1.61 for DMS, respectively. All predicted ratios between Cbrain interstitial, unbound and Cplasma, unbound were above 0.1, i.e. above the threshold for drug crossing the BBB (4), therefore the PBPK model predicted sildenafil crossing BBB. The predicted average ratio ± standard deviation was 0.56 ± 0.07 and observed mean average ratio ± standard deviation was 0.63 ± 0.15. The MRD value of all predicted ratios was 1.24, thus confirming the predictive performance of the model. Conclusions A PBPK model for sildenafil and DMS was successfully developed and validated. It predicted that sildenafil crosses BBB with concentration ratios similar as observed in vivo. This model can be used to predict plasma concentrations and unbound interstitial concentration in the brain, which may be responsible for the potential neuroprotective and neurorestorative effects of sildenafil.
[1] Xiong Y, Wintermark P. The Role of Sildenafil in Treating Brain Injuries in Adults and Neonates. Front Cell Neurosci. 2022 May 10;16:879649. [2] Gohel D, Zhang P, Gupta AK, Li Y, Chiang CW, Li L, Hou Y, Pieper AA, Cummings J, Cheng F. Sildenafil as a Candidate Drug for Alzheimer’s Disease: Real-World Patient Data Observation and Mechanistic Observations from Patient-Induced Pluripotent Stem Cell-Derived Neurons. J Alzheimers Dis. 2024;98(2):643-657. [3] Hyland R, Roe EG, Jones BC, Smith DA. Identification of the cytochrome P450 enzymes involved in the N-demethylation of sildenafil. Br J Clin Pharmacol. 2001 Mar;51(3):239-48. [4] Gómez-Vallejo V, Ugarte A, García-Barroso C, Cuadrado-Tejedor M, Szczupak B, Dopeso-Reyes IG, Lanciego JL, García-Osta A, Llop J, Oyarzabal J, Franco R. Pharmacokinetic investigation of sildenafil using positron emission tomography and determination of its effect on cerebrospinal fluid cGMP levels. J Neurochem. 2016 Jan;136(2):403-15. [5] Revatio, INN-sildenafil – European Medicines Agency, https://www.ema.europa.eu/en/documents/scientific-discussion/revatio-epar-scientific-discussion_en.pdf [6] Ku HY, Ahn HJ, Seo KA, Kim H, Oh M, Bae SK, Shin JG, Shon JH, Liu KH. The contributions of cytochromes P450 3A4 and 3A5 to the metabolism of the phosphodiesterase type 5 inhibitors sildenafil, udenafil, and vardenafil. Drug Metab Dispos. 2008 Jun;36(6):986-90. [7] Choi MK, Song IS. Characterization of efflux transport of the PDE5 inhibitors, vardenafil and sildenafil. J Pharm Pharmacol. 2012 Aug;64(8):1074-83. [8] Verscheijden LFM, Koenderink JB, De Wildt SN, Russel FGM. Differences in P-glycoprotein activity in human and rodent blood–brain barrier assessed by mechanistic modelling. Arch Toxicol. 2021 Sep;95(9):3015–29. [9] Hanke N, Gómez-Mantilla JD, Ishiguro N, Stopfer P, Nock V. Physiologically Based Pharmacokinetic Modeling of Rosuvastatin to Predict Transporter-Mediated Drug-Drug Interactions. Pharm Res. 2021 Oct;38(10):1645-1661.
Reference: PAGE 33 (2025) Abstr 11509 [www.page-meeting.org/?abstract=11509]
Poster: Drug/Disease Modelling - Absorption & PBPK