Targeting advantage of pulmonary delivery of colistin for treatment of respiratory infections

Colistin has undergone resurgence in clinical use due to the emergence of multidrug-resistant (MDR) Gram-negative bacteria. The inactive prodrug of colistin, colistin methanesulphonate (CMS), has increasingly been administered via the pulmonary and intravenous (IV) route for the management of respiratory infections in cystic fibrosis (CF) and critically-ill patients. Despite this, there is a dearth of information on the pharmacokinetics of CMS and formed colistin following pulmonary and IV administration of CMS which limits the optimisation of inhalational CMS dosing regimens. Therefore the aim of the current project was to characterise the pharmacokinetics of CMS and formed colistin in the lungs and plasma following IV and pulmonary administration of CMS, and in turn, determine the targeting advantage that can be achieved following pulmonary delivery, both in the pre-clinical and clinical setting. Linear pharmacokinetic behaviour was observed for CMS and formed colistin in plasma following IV and intratracheal (IT) instillation of CMS in Sprague-Dawley rats. Pulmonary administration of CMS 14 mg/kg resulted in extensive and prolonged exposure (12 h) to CMS and formed colistin in lung epithelial lining fluid (ELF) when compared to exposure in plasma. Formed colistin concentrations in ELF were maintained well above the minimum inhibitory concentration (MIC) for Pseudomonas aeruginosa and Acinetobacter spp of 1.0 mg/L for the 12 h sampling period. In contrast, ELF exposure of CMS was 1,200-fold lower and formed colistin concentrations unquantifiable following IV administration of the same CMS dose. The extensive exposure of CMS and formed colistin in the ELF following IT instillation was proposed to be due to drug residing in a small volume of lung lining fluid (~82 – 84 µL), potential binding to lung tissue, slow absorption of CMS from the lungs and CMS not available for renal clearance which leads to a greater fraction of the IT CMS dose converted to colistin in the lungs compared to fractional conversion of CMS to colistin in plasma. Building of a population pharmacokinetic model for CMS and colistin confirmed that a greater fractional conversion of CMS to colistin was occurring in the lungs (23%) following IT dosing when compared to in plasma (2.6%) after IV administration. In comparison to plasma, the conversion kinetics of CMS was slower in the lungs which contributed to the extensive exposure of formed colistin in ELF. The greater systemic exposure of formed colistin following IT CMS dosing (2.5- to 3.8-fold higher) when compared to IV administration of CMS was due to the absorption of pre-systemically formed colistin into plasma. Since formed colistin concentrations in ELF were maintained well above the MIC, a significant reduction in the IT CMS dose can be implemented to minimise the systemic exposure to colistin. Similar pharmacokinetics for colistin in ELF and plasma was observed after IT and IV dosing of the active antibacterial moiety, colistin. In CF subjects, pulmonary administration of 2 and 4 million international units of CMS resulted in extensive CMS and formed colistin sputum concentrations that remained above the MIC (1.0 mg/L) for the 12 h sampling period. A proportional increase in formed colistin sputum exposure was evident in the majority of CF subjects with doubling of the nebulised CMS dose. Despite high formed colistin concentrations in the lungs, unquantifiable colistin concentrations in plasma were evident with less than 2% of the CMS nebulised dose recovered in urine. In contrast, following IV infusion of CMS (150 mg of colistin base activity), formed colistin concentrations in sputum were below the MIC throughout the 12 h sampling time. Both nebulised CMS doses were well tolerated in the majority of CF subjects. This thesis demonstrates for the first time, in both pre-clinical and clinical studies the targeting advantage that can be achieved by administering CMS directly into the lungs when compared to after IV administration. In rats and CF subjects, the therapeutic availability and drug targeting index of CMS and formed colistin were magnitudes higher than unity which indicates a greater exposure of CMS and formed colistin in the lungs (ELF, sputum) which represents an effective increase in targeting to the lungs while minimising systemic exposure following pulmonary administration when compared to IV dosing. Therefore the studies undertaken in this thesis have characterised the pharmacokinetics of CMS and formed colistin in the lungs and plasma following pulmonary and IV administration of CMS to Sprague-Dawley rats and CF subjects. The targeting benefit that can be achieved following administration via the pulmonary when compared to the IV route has been demonstrated both in the pre-clinical and clinical setting.