In vitro evaluation of the hepatic disposition of colistin

Objective Colistin (polymyxin E) is a polycation antibiotic which is increasingly used (administered as Colistin Methane Sulfonate ‐ CMS) as a salvage therapy in (critically ill) patients with multidrug resistant gram‐negative infection[1]. Because colistin was introduced into clinical practice over 50 years ago, detailed trials on efficacy, safety and pharmacokinetics including underlying mechanisms are lacking. For instance, the metabolic fate of colistin is still largely unknown. Colistin is handled by glomerular filtration in the kidney, but almost completely reabsorbed in the renal tubuli. Less than 1% is eventually cleared by the kidney[2]. The purpose of the present study was to characterize the hepatic disposition of colistin and the role of transporters. Methods To characterize the metabolism of colistin, initial incubations were performed using colistin sulphate with rat liver microsomes (up to 1 mg protein/ml). Subsequently, rat and human hepatocytes in suspension (8 million/ml) were used as more complete in vitro drug metabolism tools. To measure hepatic uptake rates of colistin, the oil‐spin method was also applied to suspended hepatocytes (1 million cells/ml). Organic Anion Transporting Polypeptide (OATP) inhibitors (atazanavir and rifampicin) and Multidrug Resistance‐associated Protein (MRP) inhibitors (MK571 and benzbromarone) were used to evaluate the role of these uptake and efflux transporters in the disposition of colistin in rat and human hepatocytes. All samples were analysed by a validated LC‐MSMS method. Results Colistin, when incubated at 2 mg/L, which is the PKPD target correlating with a high probability of clinical efficacy, was stable up to 4 hr during the incubation with rat liver microsomes. However, a slow time‐dependent decrease in colistin concentration (C 0 = 2 μg/mL) was found in suspended rat and human hepatocytes, with half‐life values of 9.9 h and 18 h, respectively. Based on these in vitro metabolism data, the predicted human in vivo blood clearance, assuming the well‐stirred model (f u =0.5), was 0.10 ml/min.kg body weight, explaining only a minor part of the total blood clearance based on clinical data [0.67 ml/min.kg body weight]. Both a saturable (transporter‐mediated) and non‐saturable (passive diffusion) component were found to contribute to colistin uptake by hepatocytes. When incubated at the clinically relevant concentration, the intracellular accumulation of colistin in rat and human hepatocytes was 1.9 and 6.2‐fold higher, respectively, in the presence of the MRP inhibitor benzbromarone (30 μM). Conclusion Colistin was slowly metabolised in (rat > human) hepatocytes, apparently without involvement of CYP450 enzymes. MRP‐mediated efflux dominates the hepatic disposition of colistin, while one or more uptake transporters may also be involved. Transporter mediated uptake will be explored using Chinese hamster ovary (CHO) cell lines. These in vitro results will be used to simulate the in vivo disposition of colistin using Physiology‐based Pharmacokinetic modelling in Simcyp®. This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal .