Category: Clinical Pharmacology
Purpose: Spectinamides are semisynthetic agents of the naturally occurring antibiotic spectinomycin, and spectinamide 1599 is a lead compound under preclinical development for the treatment of multi-drug resistant tuberculosis. The purpose of this study was to predict the pharmacokinetic profile of the 1599 in human lung tissues based on its physicochemical properties and pharmacokinetic (PK) behavior in rodents.
Methods: The modeling strategy comprised a 3-step approach: In the first step, a physiologically based pharmacokinetic (PBPK) model was applied to predict the pharmacokinetics of 1599 in rats based on its physicochemical parameters, and further optimized based on measured pharmacokinetic parameters. In the second step, the ability of the model to predict the pharmacokinetic profile of 1599 in another species, mice, was assessed by comparing predicted to observed PK data using mouse-specific physiologic parameters. Lastly, the model was extrapolated to humans and a multi-compartment permeability-limited lung model, as described by Gaohua et al. , was used to predict the 1599 concentration-time profiles in different lung compartments.
Physicochemical properties of spectinamide 1599 that served as input for the PBPK model, including fraction unbound in plasma, microsomal intrinsic clearance, blood-to-plasma concentration ratio, and apparent passive permeability, were determined with standard in vitro assays. A single dose of 10 mg/kg of 1599 was administered intravenously to groups of healthy rats (n=6) and mice (n=21), serial blood samples were measured at predefined time points post dose, and 1599 concentrations were measured in the plasma obtained from the blood specimens using a validated LC-MS/MS assay. Pharmacokinetic parameters were calculated from the obtained concentration-time profiles using standard noncompartmental analysis.
The obtained physicochemical parameters were used to build a rat PBPK model in the software tool PK-Sim® 7.4. The simulated PK parameters and concentration-time profiles were compared with the in vivo measured values, and lipophilicity (LogP) and the calculated/predicted permeabilities based on the methodology by Thelen et al. were adjusted to optimize the model for a better fit to the observed data. The optimized model was then used to predict 1599 pharmacokinetics in mice, and the observed concentration-time data and PK parameters were compared to the model predictions to assess model performance across different species. As the model predicted mouse PK reasonably well, it was then used to extrapolate 1599 clearance in humans.
The permeability-limited lung model of Simcyp® v18R1 was then used to predict the concentration-time profile of spectinamide 1599 in mass and fluid compartments of each lung lobe, and the upper and lower airways using human clearance (CLiv) extrapolated from rodents via PK-Sim®, measured fraction unbound in plasma, predicted human blood-to-plasma ratio, volume distribution predicted by Rodgers et al., hepatocyte intrinsic clearance predicted from CLiv, and apparent permeability determined by PAMPA.
Results: The predicted effective molecular weight assessed via PK-Sim® was 464.95 g/mol, base pka values used were 8.69, 6.95, and 3.42. The LogP value was optimized to -1.15 from -2.51 using the parameter identification tool of PK-Sim®. Predicted permeability was 2.04 ×10‑7 cm/min (observed: 2.98 ×10-5 cm/min), predicted specific intestinal permeability (transcellular) was 1.13 ×10-9 cm/min, fraction unbound in rat plasma was 0.563, and predicted blood-to-plasma concentration was 0.72.
The 1599 plasma concentration time-profiles and parameters predicted by the PBPK model in rats and mice were in reasonable agreement with the experimentally observed parameters (Table 1). Based on this agreement, the PBPK model was applied to extrapolate 1599 PK parameters to humans (Table 1), which were subsequently integrated into the permeability-limited human lung model to predict 1599 concentration-time profiles in the pulmonary fluid compartments after an intravenous dose of 10 mg/kg as shown in Figure 1.
Conclusion: We were able to develop a PBPK model for 1599 using physicochemical properties and measured pharmacokinetic parameters and were able to predict 1599 plasma concentration-time profiles in rodents with reasonable agreement with in vivo observed data. We subsequently applied the model to extrapolate 1599 PK parameters to humans and used a permeability-limited lung model to predict 1599 exposure in different lung compartments.
Pradeep Lukka– Memphis, Tennessee
Santosh Wagh– University of Tennessee Health Science Center, Memphis, Tennessee
Zaid Temrikar– Graduate Student , University of Tennessee Health Sciences Center, Memphis, Tennessee
Richard Lee– St Jude Children's Research Hospital, Memphis, Tennessee
Bernd Meibohm– Professor and Associate Dean, University of Tennessee Health Science Center, Memphis, Tennessee