A Population PK Approach to Evaluate Plasma and ELF Exposure Changes Following Co-administration of Spectinamide 1599 and Pyrazinamide in Mice

Objectives: Spectinamide 1599, an aminocyclitol derivative with potent anti-TB activity, has demonstrated in vivo synergy with pyrazinamide (PZA) in both acute and chronic aerosol-infected mouse models [ 1]. To investigate the mechanistic basis of this synergistic interaction, we conducted in vivo drug interaction studies to assess potential differences in drug exposure - either for one or both agents - in plasma and epithelial lining fluid (ELF). Our objective was to develop population pharmacokinetic (PK) models for 1599 and PZA in plasma and ELF of healthy mice and to compare drug concentrations between monotherapy and combination groups. Methods: Spectinamide 1599 (200 mg/kg, subcutaneously) and PZA (150 mg/kg, orally) were administered either alone or in combination to BALB/c mice once daily for five consecutive days. On Day 5, plasma and ELF samples were collected at multiple time points (pre-dose to 24 h post-dose) using destructive sampling (n = 3 per time point). Given the sparse sampling and inherent inter-individual variability associated with destructive sampling, a population pharmacokinetic modeling strategy was employed using NONMEM (version 7.5) with the FOCE method. One- and two-compartment models were tested to describe the plasma PK of 1599 and PZA. Covariate analysis was performed to evaluate the impact of co-administration on systemic pharmacokinetics. A joint plasma-ELF model was then constructed, where plasma PK parameters were fixed to previously estimated values and only ELF parameters were estimated. Multiple structural models, including direct response and effect-site equilibrium models, were tested to characterize ELF kinetics and assess potential delays in ELF distribution [ 2]. The final model was selected based on the best fit and was subsequently used to assess the effect of drug co-administration on ELF penetration. Results: Plasma concentration - time profiles of 1599 and PZA were best described by two-compartment and one-compartment models, respectively, each incorporating first-order absorption and linear elimination kinetics. The estimated apparent clearance values were 0.62 L/h/kg for 1599 and 0.76 L/h/kg for PZA, with corresponding volumes of distribution of 0.10 L/kg and 0.76 L/kg, respectively. Covariate analysis indicated that co-administration with PZA did not significantly influence the plasma pharmacokinetics of 1599. However, the presence of 1599 led to a twofold increase in the apparent volume of distribution of PZA. ELF concentrations of both agents were best captured using a direct response (instantaneous equilibrium) model linking plasma and ELF compartments. Notably, co-administration with PZA increased the ELF-to-plasma penetration ratio of 1599 by approximately 90%, suggesting enhanced pulmonary distribution. In contrast, 1599 did not significantly affect the ELF/plasma penetration ratio of PZA, indicating no meaningful difference in PZA exposure at the site of infection between monotherapy and combination groups. Conclusion: Population pharmacokinetic modeling indicates that co-administration with pyrazinamide enhances 1599 exposure in the epithelial lining fluid (ELF), suggesting a pharmacokinetic basis for their observed in vivo synergy.