A Physiologically-Based Pharmacokinetic Model for Ganciclovir and its Prodrug Valganciclovir in adults and children- Online Resource 1
All modeling and simulation was carried out using GastroPlus™ version 9.0 (Simulations Plus, Inc., Lancaster CA). Physiologies, including organ weights, volumes, and blood perfusion rates in individual species, were generated by the program’s internal Population Estimates for Age-Related (PEAR™) Physiology™ for all simulations. The basic physicochemical and biopharmaceutical properties for ganciclovir and valganciclovir are listed in Table1 of the main article. Both ganciclovir and valganciclovir are hydrophilic, poorly permeable molecules. To account for slow diffusion across tissue cell membranes, the permeability-limited tissue model was used for all tissues in the PBPK model. The rate of passive drug diffusion across the cell membrane is given by permeability-surface-area product (PStc) and is driven by concentration gradient between unbound concentrations in extracellular and intracellular tissue compartments (Eq 1.).
Equation 1
In Eq 1., Perm is rate of passive diffusion across the cell membranes in a tissue, PStc is permeability-surface-area product for that tissue, Ce,u and Ci,u are unbound drug concentrations in extracellular and intracellular tissue space, respectively.
Since the total cell surface areas of individual tissues are not known, the application of this tissue model typically requires fitting PStc values against in vivo data. To reduce the number of model parameters to be estimated, a single value of PStc per cell volume (Specific PStc) was used. The total PStc for each individual tissue was calculated from the Specific PStc and total cell volumes in each tissue (Eq 2.).
Equation 2
In Eq 2, PStc is total permeability-surface-area product in a tissue, Vol is total tissue volume, Fvec is extracellular volume fraction in the tissue, and SpecificPStc is PStc per mL of cell volume.For liver and kidney tissues, the total tissue PStc is split between basolateral and apical cell surface.
The plasma concentration-time (Cp-time) profiles after intravenous (IV) administration of ganciclovir were used to parameterize the distribution and renal clearance of ganciclovir. The passive renal filtration was calculated from the fraction unbound in plasma and the glomerular filtration rate (fup*GFR). Carrier-mediated tubular secretion (mediated by OAT1 and MRP/MATE2K transporters in human) was assumed to occur in pre-clinical species also. The ganciclovir Cp-time profile after 10 mg/kg IV administration in monkey was used to fit ganciclovir Specific PStc value and the Vmax values for renal transporters. The same Specific PStc value as fitted against monkey data (3.1E-4 mL/s/mL-cell volume) was then used to simulate ganciclovir tissue distribution in mouse, rat, and dog. Along with the tissue volumes for different animals, the Specific PStc value was used to calculate animal-specific PStc values for all tissues. The ganciclovir PStc values for all tissues and animals are summarized in Table 1of this appendix. The Vmax values for renal transporters were refitted for individual animals against the ganciclovir Cp-time profiles after ganciclovir IV administration (10 mg/kg dose in rat and dog, 10 and 16 mg/kg dose in mouse). The fitted ganciclovir transporter parameters for all animal species are summarized in Table 2of this appendix.
The Cp-time profiles after IV administration of valganciclovir were then used to parameterize the distribution of valganciclovir, its conversion to ganciclovir and biliary secretion of ganciclovir. Monkey data (valganciclovir and ganciclovir Cp-time profiles after IV administration of valganciclovir) was used to fit the valganciclovir Specific PStc value,valganciclovirVmax values for PepT1 and metabolic conversion by esterasesin kidney and liver, ganciclovir Vmax value for MRP in liver, and ganciclovir biliary secretion. The same Specific PStc value as fitted against monkey data (2E-3 mL/s/mL-cell volume) was again used to simulate the valganciclovir tissue distribution in all other animals (mouse, rat, and dog) and the remaining parameters were refitted for individual animals. The final valganciclovirPStc values for all tissues and animals are summarized in Table 1of this appendix. The fitted enzyme and transporter parameters for all animal species are summarized in Table 2of this appendix.
The in vitro studies showedenzymatic degradation of valganciclovir in mouse whole blood, while no enzymatic degradation was observed in dog or human whole blood (unpublished data). Similar studies were not performed for rat or monkey whole blood, however, based on simulations and observed valganciclovirCp-time profiles it was apparent that similar enzymatic conversion in blood might be happening in rat as well. The final PBPK model assumed the enzymatic conversion of valganciclovir in blood of mouse and rat. The PBPK model in GastroPlus version 9.0 does not allow for enzymatic clearance directly in the blood, therefore lung tissue was selected as a surrogate tissue to account for the additional conversion site.
The fitting of valganciclovir parameters (tissue uptake and conversion to ganciclovir) was driven by valganciclovirCp-time profiles, the biliary clearance of ganciclovir was driven by ganciclovir Cp-time profile after valganciclovir administration. The fitted ganciclovir biliary clearance rate was subsequently included also in the simulations of ganciclovir PK after ganciclovir IV administration, where it, however, did not impact the simulation result. The uptake of ganciclovir from systemic circulation into liver was modeled only by passive transcellular diffusion, which is very low.
After IV administration of ganciclovir, nearly all (98-100% in all four species) ganciclovir was eliminated in urine.
After valganciclovir administration, the major portion of ganciclovir was formed in the liver (78, 63, 81, and 93 % in mouse, rat, dog and monkey, respectively) from where the ganciclovir may be effluxed into systemic circulation or into the bile. The renal elimination of ganciclovir was 54, 71, 97, and 99% of total ganciclovir elimination in mouse, rat, dog, and monkey.
Some of the fitted enzyme and transporter Vmax values showed large interspecies differences. While there is a possibility for differences in enzyme and transporter expression levels in different species, some of these differences may also be related to misspecification of some processes in the model. The fitted Vmax values for PepT1 and valacyclovirase in liver were highest in rat and mouse. This was driven by sharp decrease in valganciclovir concentrations after valganciclovir IV administration in these species. In vitro studies already showed enzymatic degradation of valganciclovir in mouse which was not observed in monkey and it is possible that other tissues may also be contributing as well. Since, due to lack of information about the enzyme expression levels, possible contribution of other tissues was not included in the model, and fitted liver parameters may be compensating for that.
Table2PStc Values (ml/s) for Ganciclovir and Valganciclovir PBPK Models a
Tissue / Mouse (0.032 kg) / Rat (0.2 kg) / Dog (11 kg) / Monkey (6 kg)Ganciclovir
Adipose / 6.18E-4 / 2.00E-3 / 4.56E-1 / 1.66E-1
Brain / 1.30E-4 / 2.42E-4 / 2.05E-2 / 3.26E-2
Heart / 2.73E-5 / 1.88E-4 / 1.65E-2 / 4.00E-3
Kidney b / 5.23E-5 / 3.11E-4 / 5.80E-3 / 2.00E-3
Liver
Liver-Apical c / 3.47E-4
2.56E-1 / 1.30E-3
5.00E-1 / 5.44E-2
1.00 / 2.29E-2
1.68E-1
Lung / 3.86E-5 / 3.20E-4 / 1.82E-2 / 9.70E-3
Muscle / 3.00E-3 / 2.52E-2 / 1.25 / 7.74E-1
Red marrow / 2.81E-4 / 3.93E-4 / 3.92E-2 / 1.42E-2
Repro org. / 3.98E-5 / 4.20E-4 / 3.80E-3 / 6.90E-3
Rest of body / 4.05E-4 / 4.50E-3 / 1.87E-1 / 7.70E-2
Skin / 7.93E-4 / 5.60E-3 / 1.50E-1 / 1.06E-1
Spleen / 2.97E-5 / 1.11E-4 / 6.30E-3 / 1.00E-3
Yellow Marrow / 1.70E-4 / 8.39E-4 / 1.80E-2 / 3.87E-2
Valganciclovir
Adipose / 4.00E-3 / 1.30E-2 / 2.94 / 1.07
Brain / 8.40E-4 / 1.60E-3 / 1.32E-1 / 2.10E-1
Heart / 1.76E-4 / 1.20E-3 / 1.06E-1 / 2.57E-2
Kidney b / 3.37E-4 / 2.00E-3 / 3.72E-2 / 1.26E-2
Liver / 2.20E-3 / 8.70E-3 / 3.51E-1 / 1.48E-1
Lung / 2.49E-4 / 2.10E-3 / 1.17E-1 / 6.26E-2
Muscle / 1.96E-2 / 1.62E-1 / 8.03 / 4.99
Red marrow / 1.80E-3 / 2.50E-3 / 2.53E-1 / 9.18E-2
Repro Org. / 2.57E-4 / 2.70E-3 / 2.44E-2 / 4.47E-2
Rest of body / 2.60E-3 / 2.90E-2 / 1.20 / 4.96E-1
Skin / 5.10E-3 / 3.64E-2 / 9.67E-1 / 6.82E-1
Spleen / 1.92E-4 / 7.13E-4 / 4.03E-2 / 6.40E-3
Yellow Marrow / 1.10E-3 / 5.40E-3 / 1.16E-1 / 2.49E-1
aPStc values were automatically calculated by the program from Specific PStc value (0.002 and 0.00031 mL/s/mL-cell volume for valganciclovir and ganciclovir, respectively) and tissue cell volume (calculated from total tissue volume and extracellular volume fraction) for each physiology.
bThe same PStc value was used for apical and basolateral kidney membranes.
cApical PStc in liver represents flux of ganciclovir from liver into bile due to active secretion
Table2Transporter and Enzyme Model Parameters for Ganciclovir and Valganciclovir PBPK Models
Species / Human Transporter / enzyme / Location / Type / Vmax(mg/s/g-tissue)a / Km (µg/mL)b
Transporters
Valganciclovir / PepT1 / Liver-Ba, Kidney-Ap / Influx / 15, 15, 2.50E-2, 0.5 / 866
PepT1 / Lungc-Ba / Influx / 3.50E-2, 4.00E-2, -, - / 866
Ganciclovir / MRP / Liver-Ba / Efflux / 9.18E-2, 9.18E-2,
9.18E-2, 9.18E-2 / 1000
MRP / Lungc-Ba / Efflux / 9.18E-2, 9.18E-2, -, -
MRP/MATE2K / Kidney-Ap / Efflux / 2.19E-3, 1.72E-3, 1.70E-3, 1.02 / 1092
OAT / Kidney-Ba / Influx / 7.31E-3, 1.32E-2, 3.75E-5, 1.69E-2 / 229
Enzymes
Valganciclovir / Esterase / Liver, Kidney / ‒ / 10.5, 10.5, 0.035, 0.35 / 947
Esterase / Lungc / 5.25E-2, 5.25E-2, -, - / 947
Abbreviations: Ap = Apical membrane; Ba = Basolateral membrane; Km = Michaelis-Menten constant; MATE = Multi antimicrobial extrusion protein; MRP – multidrug resistance protein; OAT = Organic anion transporter; OCT = organic cation transporter; PepT1 = peptide transporter 1; Vmax = maximum rate of reaction
aThe values are listed for all four animal species in order: mouse, rat, dog, monkey
bThe high value for Km signifies the conversion was not saturated in the range of data examined. Therefore, the combination of parameters Vmax/Km is most meaningful. The Km values for interaction of valganciclovir with PepT1 and esterase were obtained in-house and Km values for interaction of ganciclovir with MATE2K and OAT were used as reported in literature (Tanihara et al. BiochemPharmacol 2007, 74: 359-371;Takeda et al. J Pharm ExpTher2002, 300: 918-924). Experimental Km value for ganciclovir interaction with MRP was not available and was set to high value similar to reported value for MATE2K.
cIn vitro studies showed 15% enzymatic degradation over 60 minutes in whole mouse blood. PBPK model in GastroPlus version 9.0 does not allow for metabolic conversion directly in the blood or plasma and lung was selected as a surrogate tissue to account for this conversion. No hydrolysis was observed in human or dog whole blood. Similar information was not reported for rat or monkey blood, but based on the simulations and observed PK profiles for valganciclovir and ganciclovir, we assumed the enzymatic degradation would occur on rat blood and not in monkey blood.
Figure 1.Observed (points) and simulated (lines) valganciclovir (red) and ganciclovir (blue and green) PK profiles after IV ganciclovir (a,c,e,g) and IVvalganciclovir (b,d,f,h) administration in mouse (a,b), rat (c,d), dog (e,f), and monkey (g,h).Ganciclovir was administered at doses 10 (blue) and 16 (green) mg/kg in mouse and 10 mg/kg in all other animals. Valganciclovir was administered as HCL salt at dose 15.7 mg/kg in all animals.